ANNUAL REPORT 2013 - NHKAccomplishments in 2013 2 ｜ NHK STRL ANNUAL REPORT 2013 Television...

20

13

Nippon Hoso Kyokai [Japan Broadcasting Corporation]

August 2014

2013

NHKScience & TechnologyResearch Laboratories

Annual Report

2013


Annual Report

Annual R

eport

Seijogakuen-mae Soshigaya-Okura Odakyu Line

To ShinjukuNHK STRL

Bus stopBus stop

Bus stopToho

Natl. Ctr. for Child Health

and Development

TomeiExpressway

Kinuta Koen（Park）

Ring Road No. 8/Kanpachi Dori

Setagaya Dori

Yoga

Tokyu

Den-en-toshi line

To Shibuya

Shuto

Expressway

Yoga I. C.

Access to NHK STRL

■Odakyu line, from Seijogakuen-mae station, south exit:［Odakyu Bus/Tokyu Bus］・Shibu 24（渋24） toward Shibuya Station［Tokyu Bus］・To 12（等12） toward Todoroki-soshajo・Yo 06（用06） toward Yoga Station（weekdays only）・Toritsu 01（都立01） toward Toritsu Daigaku Station, north exit■Tokyu Den-en-toshi line, fromYoga station:［Tokyu Bus］・To 12（等12） toward Seijo-gakuen-mae station・Yo 06（用06） toward Seijo-gakuen-mae station（weekdays only）

In all cases, get off the bus at the “NHK STRL”（NHK技術研究所）bus stop

Directions

Nippon Hoso Kyokai（NHK）Science & Technology Research Laboratories（STRL）1-10-11 Kinuta, Setagaya-ku, Tokyo

Tel: +81-3-5494-1125http://www.nhk.or.jp/strl/english

Edited and Published by:

Table of Contents

Greetings …………………………………………1

1 Television conveying a strong sensation of reality ……4

……14

…

5noisiV-iHrepuS1.1

5tamrofnoisiV-iHrepuS1.1.1

5saremaC2.1.1

6syalpsiD3.1.1

7Recording systems4.1.1

16Hybridcast1.1.3

18SNS and Program viewing analysis2.1.3

19Security technologies3.1.3

20

24

IP content delivery technology4.1.3

1.1.5 Video Coding 8

9Sound systems providing a strong sense of presence

6.1.1

10Satellite broadcasting technology7.1.1

11

13

Terrestrial transmission technology8.1.1

1.1.9 Wired transmission technology

3 Convergence of broadcasting and telecommunications

…………16

3.1 Technologies for the convergence of broadcasting and telecommunications

Accomplishments in 2013……………………2

4 User-friendly information presentation 22

4.1 User-friendly information presentation 22

4.2 Speech recognition for closed captioning 23

4.3 Speech and audio signal processing for the elderly

25Language processing for barrier-free services4.4

4.5 Content retrieval and recommendation technology 26.

4.6 Viewers’ mental state estimation technology 27

5 Advanced content production

technology …………………………………28

5.1 Contribution transmission technology 28

5.1.1 120-GHz-band FPU for uncompressed SHV 28

5.1.2 Bidirectional Field Pick-up Unit (FPU) transmission technology 29

5.1.3 700-MHz-band frequency migration 29

5.2 IP technologies for IP based contribution networks 31

5.1.4 Wireless contribution link technology 30

32Enhanced ISDB-T3.5

6 Devices for next-generation

broadcasting…………………………………33

6.1 Advanced image sensors 33

6.1.1 Super-high-sensitivity image sensors 33

6.1.3 Core technology for 3D-structured imaging devices 35

6.2 Advanced storage technology 35

6.2.1 Magnetic recording technology 35

6.2.2 Holographic memory 36

6.1.2 Organic image sensors 34

6.3 Next-generation display technologies 36

6.3.1 Flexible displays 36

6.3.2 Advanced display devices 38

7 Research-related work ……………………39

7.1 Joint activities with other organizations 39

7.2 Publication of research results 42

7.3 Applications of research results 45

NHK Science & Technology Research Laboratories Outline ……………………………48

：】Ｓｅｒｖｅｒ／ＮＨＫ技研／研究年報２０１3（英文版）／表紙（背幅３．５ｍｍ）／目次と表３

2 Three-dimensional images

2.1 Three-dimensional images

2.1.1 Integral 3D television 14

14

51Generating 3D content from multi-viewpoint images

2.1.2

16


To ShinjukuNHK STRL

Bus stopBus stop

Bus stopToho


and Development

TomeiExpressway



Setagaya Dori

Yoga

Tokyu

Den-en-toshi line

To Shibuya

Shuto

Expressway

Yoga I. C.

Access to NHK STRL



Directions




Table of Contents

Greetings …………………………………………1


……14

…

5noisiV-iHrepuS1.1


5saremaC2.1.1

6syalpsiD3.1.1


16Hybridcast1.1.3



20

24




6.1.1


11

13




…………16











technology …………………………………28





























14


2.1.2

16

Greetings

Shuichi FUJISAWAHead of NHK Science & Technology Research Laboratories

NHK STRL ANNUAL REPORT 2013　｜　1

NHK Science & Technology Research Laboratories (STRL), the sole research facility in Japan specializing in broadcasting technology, is contributing to building a rich broadcasting

culture in the country. Our aim is to fulfill the role of NHK as the public broadcaster through the development of broadcasting technology.

In 2013, Tokyo won the right to host the 2020 Summer Olympic and Paralympic Games. It may be no exaggeration to say that broadcasting technologies - such as the now popular satellite broadcasting, Hi-Vision TV, flat panel displays and digital broadcasting - have progressed hand in hand with the history of the Olympics. In anticipation of the upcoming games, we are reminded of our initial enthusiasm for broadcasting technology.

We are conducting research to resolve near-term operational issues and, at the same time, are undertaking long-term studies to create new services for the future. We are also building a technological infrastructure for the next generation of broadcasting, as we carry out research on a wide range of topics from cognitive science to devices and systems.

FY 2013 was the second year of NHK STRL’s three-year plan established in accordance with the 2012-2014 NHK corporate policy. Based on this plan, we pursued R&D on topics including Hybridcast, which is an infrastructure technology for services that take advantage of broadcasting and telecommunications. We also made progress on technologies for delivering highly realistic television broadcasts, such as Super Hi-Vision and 3D television.

In September 2013, we launched experimental services of NHK Hybridcast. We are also accelerating our R&D on technologies for Super Hi-Vision with the goal of beginning test broadcasts in 2016. Moreover, to strengthen ties with broadcasting and research facilities in other countries, we are collaborating with other organizations on many experiments and sharing our research results internationally.

It is my hope that this annual report will help you better understand NHK STRL’s research and development activities and allow both of us to build collaborative relationships in research and development. I also hope the document will aid you in making use of the results of our efforts.

Finally, I would like to express my sincere gratitude for your support and appreciate your continued cooperation in the future.

Accomplishments in 2013

2　｜　NHK STRL ANNUAL REPORT 2013　

Television conveying a strong sensation of realityNHK STRL is researching 8K Super Hi-Vision (SHV), a new form of television that delivers

an immersive sensation of reality. We aim to conduct test broadcasting of SHV in 2016 and begin full service broadcasting in 2020. We developed a wide-color-gamut SHV camera and a wide-color-gamut SHV projector with a laser light source in our research on video formats. We also developed a device interface to transmit various UHDTV signals including “full-spec” SHV signals over a 10-Gbps multilink, which was adopted as an ARIB Standard. In our work on SHV cameras, we developed a compact single-chip camera system weighing 5 kg and a cube-style compact camera head weighing 2 kg. We used the silent, high-sensitivity “theater camera” to record a performance of Teatro alla Scala, which met with a favorable reception at live viewings. With the goal of displaying full-specification SHV video with light-emitting displays, we studied on organic light-emitting diode displays that can operate at the 120-Hz frame frequency of SHV and demonstrated the feasibility of these displays by identifying the design values and circuit structure of the oxide TFT. In our work on audio, we developed a display-integrated 12-loudspeaker system capable of reproducing 22.2 multichannel sound.

In our work on large-capacity transmission technologies for next-generation terrestrial digital broadcasting, we began measurements of the channel characteristics at an experimental station for broadcasting dual-polarized MIMO transmissions in Hitoyoshi, Kumamoto Prefecture, and found that SHV signals compressed and encoded to 91 Mbps could be sent over a long distance (27 km) and correctly received. We also developed a method to transmit uncompressed SHV signals over optical fiber.

→See p. 4 for details.

Three-dimensional imagesWe are studying more natural and viewable three-dimensional television that does not

require special glasses. To improve the quality of 3D images, we developed an imaging equipment system with a large number of pixels by using multiple cameras and displays and integrating the images. We also began a study on coding technologies for the integral method with an eye to eventual standardization.


Convergence of broadcasting and telecommunicationsWe continued with our R&D on technologies for the convergence of broadcasting and

telecommunications and the realization of new broadcasting services that enhance the value of programs by taking advantage of broadcasting and telecommunications networks. We proceeded with our R&D on Hybridcast and worked on standardization of the next version of this system at IPTV Forum Japan. We studied the SHV data broadcasting format for 8K Hybridcast and contributed to the Moving Picture Experts Group’s MPEG Media Transport (MMT), a next-generation multiplexing technology that was released as an international standard.


User-friendly information presentationWe are researching a tactile presentation system with which people with visual

impairments can apprehend the shape of displayed objects by touch. We implemented a function to indicate the hardness of virtual objects with multi-point force presentation equipment that gives tactile information on multiple points of the fingers. In our work on automatic translation technology from Japanese text to sign-language CGs, we released a Japanese and sign-language dictionary system on NHK Online in order to gather opinions on our sign-language CGs for the purpose of improving services. In our studies on the closed-captioning using speech recognition technology, we developed an algorithm that automatically corrects recognition errors by using news scripts. This system will make it easier for local broadcast stations to offer closed caption services. Regarding the system to assist conversion of news programs scripts into easy Japanese for non-native Japanese speakers and children, we developed technologies to automatically show hiragana alongside kanji and highlight proper names on the “NEWS WEB EASY” website. In our research on high-quality speech synthesis, we began a study on speech processing technology that enables the expression of various emotions.


Shooting test of the theater camera (right)

FPD-integrated 12-loudspeaker system

Integral display equipment using four projectors

Sign-language CG website


Advanced content production technologyRegarding our work on wireless transmission of uncompressed SHV signals, we improved

the temperature characteristics of the 120-GHz-band field pick-up unit (FPU) that works with polarization multiplexing and demonstrated that it could transmit over a distance of 1 km in fine weather. In our research on a digital radio microphone capable of high-quality sound transmissions with low latency, we conducted experiments on its OFDM transmission system; this work led to the establishment of a domestic standard. In our work on enhanced ISDB-T, we studied a One-Seg system whose service area can be limited. We performed field experiments envisioning a case in which local governments and other organizations use service-area-limited broadcasting to provide urgent information in times of disaster and found that the system worked as expected. →See p. 28 for details.

Devices for next-generation broadcastingIn our work on high-sensitivity image sensors, we studied photoconductive films able to

multiply electric charges by applying a low voltage for increasing the sensitivity of solid-state image sensors.

In our work on organic image sensors, we developed “continuously stacked” sensor technologies in which layers of three different organic photoconductive films and transparent thin film transistor circuits are alternately stacked on a glass substrate. We continued with our development of 3D-structured imaging devices to enable pixel-parallel signal processing that simultaneously provides high resolution and high frame rates. We completed a verification of the devices’ operation principle.

In our work on recording devices, we doubled the multiplexing number of holographic recording and reproduction experimental equipment to 600 and accelerated the reproduction signal processing system. This work enabled the data production rate to be increased to 500 Mbps.

In our research on advanced storage devices, we developed a magnetic recording device utilizing magnetic nano-domains on magnetic nanowires, with the goal of realizing high-speed magnetic recording device with no moving parts. We achieved real-time detection of the motion of the nano-domains while using a pulse current to drive them.

In our work on next-generation display devices, in particular, bendable sheet-type displays, we prototyped a panel using red phosphorescent organic light emitting diodes (OLEDs) featuring low power consumption and a long service life. We also developed an “inverted” OLED that has an electrode structure opposite to that of ordinary OLEDs and confirmed that it had not deteriorated 250 days after being sealed in a plastic substrate. →See p. 33 for details.

International collaboration and standardization activitiesWe are working with international standardization organizations such as ITU-R and

SMPTE to standardize SHV and other technologies. Following its release of Recommendation BT.2020 for SHV video formats, ITU-R established Recommendation BS.2051 in February 2014 on hybrid sound systems combining channel-based and object-based methods, including the 22.2 multichannel sound loudspeaker arrangement.

At MPEG, the essential part of MMT multiplexing technology for implementing broadcast-broadband hybrid transmissions, including the packet structures and control signals for hybrid distribution, was approved as an international standard.

The Advanced Television Systems Committee (ATSC), an organization standardizing TV broadcasting systems in the U.S., began work on standards for next-generation digital terrestrial broadcasting in FY 2013. NHK presented our research results on transmission and multiplexing technologies for SHV terrestrial broadcasting at the ATSC 3.0 conference and participated in their meetings. →See p. 39 for details.

Public relations activitiesThe STRL Open House, held each year in May, featured 37 exhibits, three interactive

exhibits, and 13 poster displays of the laboratories’ latest research results. The theme for the FY 2013 exhibit was “Expect, Explore, Experience.” The event attracted a total of 20,607 visitors.

We exhibited our latest achievements such as SHV and Hybridcast at international events, including the NAB Show 2013 broadcast equipment exhibition held in the U.S., IBC 2013, the largest broadcast equipment exhibition in Europe, in the Netherlands, ABU 2013 in Vietnam, and the Sochi Winter Olympics in Russia.

STRL’s research results were also exhibited at events held by NHK broadcasting stations throughout the country. →See p. 42 for details.

Baseband signal processing unit

Compact 120-GHz-band RF unit set on a camera tripod

120-GHz-band FPU for uncompressed SHV

Digital radio microphone (Two-piece model)

Digital radio microphone (Hand-held model)

Digital radio microphone receiver

OFDM digital radio microphone prototype

Real-time reproduction of ultra-high resolution video using a holographic recording device

STRL Open House

NHK STRL is researching 8K Super Hi-Vision (SHV), a television system that delivers immer-sive video and sound. We aim to conduct test broadcasting of SHV in 2016 and begin full ser-vice broadcasting in 2020.

In our research on video formats, we developed a wide-color-gamut SHV camera and a wide-col-or-gamut SHV projector with a laser light source. The combination of these devices enables accu-rate reproduction of highly saturated object col-ors that could not be handled with Hi-Vision. We studied technologies to convert the SHV video signal format used in current program produc-tion to the full-resolution format in order to sup-port the broadcasting signal format to be used in 2016 and devised a method suitable for real-time processing to interpolate color signals precisely.

We developed a single-chip color camera sys-tem with a wavelength division multiplexing unit and a camera control unit for practical use. We also prototyped a cube-shaped compact camera head weighing only 2 kg. We improved the sensi-tivity of the image sensor to be used in full-speci-fication SHV cameras (33 megapixels, 120 Hz) by using a nanofabrication semiconductor process. We also had the opportunity to use the silent, high-sensitivity “theater camera” prototyped in FY 2012 to shoot Teatro alla Scala’s performance of “Rigoletto” for a live viewing event. The SHV video of this performance was favorably received by the audience.

We studied light-emitting displays for full-specification SHV video and demonstrated the feasibility of 120-Hz organic light-emitting diode (OLED) displays by identifying the design param-eters and circuit structure of the oxide TFT.

In our effort to give a high degree of mobil-ity to SHV program productions, we developed a compression recording technology for the cam-era and a high-speed recording technology using multi-parallel NAND flash memory. We also pro-totyped a compact video recorder.

Regarding video coding, we upgraded the SHV video encoder to embody the MPEG-H High Effi-ciency Video Coding (HEVC)/H.265 scheme that we developed in FY 2012. We updated the bit-stream generation function to be compliant with the international standard for HEVC and added a multiplexing function for the connection to the transmitter. We also compared its performance with that of a conventional MPEG-4 AVC /H.264 encoder and confirmed that its image quality is superior.

In our work on audio, we added functions to

save, read, and adjust frequently used production patterns to a 22.2 multichannel sound mixing system in order to shorten the production time. We also implemented 22-directional reverbera-tion waveforms measured in various places in a 3D reverberator and developed a display-inte-grated 12-loudspeaker system capable of repro-ducing 22.2 multichannel sound.

Regarding our standardization activities, we led the effort to establish an ITU-R Recommenda-tion on sound systems combining channel-based and object-based methods, including loudspeak-er layouts for 22.2-multichannel sound. The Rec-ommendation was published in February 2014. We also proposed to the Association of Radio In-dustries and Businesses (ARIB) an interface for connecting UHDTV video devices in studio. The interface transmits various UHDTV signals, in-cluding full-specification SHV signals, over mul-tiple 10-Gbps links. ARIB established a standard for it in March 2014.

In our work on satellite broadcasting in the 12-GHz band, we studied the 16 Amplitude and Phase Shift Keying (APSK) modulation scheme with high frequency usage efficiency and pro-posed to the Information and Communications Council a transmission system for SHV that is ca-pable of transmitting 100-Mbps signals through a single channel of the satellite transponder. We also worked on international standardization of multiplexing technologies; our proposal was adopted by MPEG as an international standard for its new media transport scheme called MPEG Media Transport (MMT).

We are also developing large-capacity trans-mission technologies for the next generation of digital terrestrial broadcasting. This year, we in-stalled an experimental station for testing dual-polarized multiple-input multiple-output (MIMO) transmissions in Hitoyoshi, Kumamoto Prefec-ture, and began measurements of the channel characteristics. The experiments showed that SHV signals compressed to 91 Mbps could be transmitted over a long distance (27 km).

In our work on optical transmission of uncom-pressed SHV signals, we developed a method to transmit 72-Gbps full-resolution SHV signals on eight wavelengths (10 Gbps per wavelength). We also proposed to the Japan Cable Television En-gineering Association (JCTEA) a method to parti-tion compressed SHV signals and transmit them through multiple channels of cable television networks.


1 Television conveying a strong sensation of reality

1.1 Super Hi-Vision

1.1.1 Super Hi-Vision format

We have made progress with our R&D and standardization activities related to the 8K Super Hi-Vision (SHV) video system.

■ Wide color gamutThe SHV video format employs wide-gamut system colo-

rimetry that covers a wider range of colors than for Hi-Vision and can reproduce most real object colors. We designed the spectral sensitivity characteristics that would be needed for a camera supporting the wide-color-gamut and used the results to develop prisms for full-resolution SHV cameras with three 33-megapixel image sensors (1). We built a wide-color-gamut SHV system consisting of this wide-color-gamut camera and a wide-color-gamut SHV projector (see 1.1.3). At the NHK STRL Open House, we used the system to capture and display im-ages of objects with highly saturated colors and demonstrated the improvement in accuracy of the reproduced colors includ-ing highly saturated ones.

In order to convert the colors of the SHV wide-gamut sys-tem colorimetry into those of the Hi-Vision color space, we are studying a method that decreases only chroma without altering lightness and hue to maintain image quality. Since sufficient data had not been reported on the characteristics of perceptual hues in the color space outside the Hi-Vision color-gamut, we performed subjective evaluation tests using the wide-color-gamut SHV projector. The results showed that some hues have different characteristics from the ones pre-dicted by the conventional uniform color space.

■ High resolutionWe are studying technologies to convert SHV video captured

by 8-megapixel four-sensor SHV cameras or 33-megapixel single-sensor SHV cameras, which has a Bayer pattern pixel structure, into full-resolution SHV video with high quality. In the Bayer pattern, two green, one red and one blue pixels are arrayed in a square grid. We devised a method suitable for re-al-time processing to interpolate color signals precisely (2) and confirmed that it can generate color images that look natural and have better resolution and minimal artificial colors.

To better understand the resolution characteristics of SHV cameras, we developed a new method to measure not only horizontal or vertical but also multi-dimensional resolution characteristics precisely by analyzing an image of a slightly-slanted knife-edge target captured by the cameras (3).

To clarify the relation between spatial video resolution and preferred viewing distance during TV viewing, we measured the preferred viewing distance when displaying a program consisting of still images at different levels of spatial resolu-tion. The results showed that a shorter viewing distance tends to be preferred for higher spatial resolution (4).

■ InterfacesWe are standardizing an interface for connecting video de-

vices in the studio to transmit various UHDTV signals over 10-Gbps multilinks at the Association of Radio Industries and Businesses (ARIB). The UHDTV signals include full-specifica-tion SHV signals (7680 × 4320 pixels, 120-Hz frame frequency, 12 bit), 4K signals (3840×2160 pixels), 60-Hz frame frequency, and the 4:4:4, 4:2:2 and 4:2:0 formats. The ARIB Standard STD-B58 (5) was established in March 2014, and it includes specifica-tions of the data mapping and serialization, transmitter and receiver characteristics, and optical connector characteristics.

■ Subjective evaluation of image qualityWe studied methods for subjectively evaluating the image

quality that would be suitable for SHV video, which has high resolution and a wide field of view. Through subjective evalu-ation tests of coded video, we found that subjective evaluation scores depend on the viewing distance and horizontal view-ing position. Based on this result, we proposed that for a sub-jective evaluation of SHV video using a display of about 80 inches, three evaluators should observe horizontally trisect-ed parts of an image at a viewing distance of 0.75 times the screen height. This research was conducted as part of a project commissioned by the Ministry of Internal Affairs and Commu-nications, titled “The research and development project for the expansion of radio spectrum resources”.

■ Broadcasting format standardizationWe contributed to the standardization of ultra-high defini-

tion television (UHDTV) broadcasting systems at the Informa-tion and Communications Council and ARIB. In March 2014, the Council issued specifications for video formats including SHV with a 120-Hz frame frequency and High Efficiency Video Coding (HEVC).

[References](1) T. Soeno, K. Masaoka, T. Yamashita, T. Nakamura, R. Funatsu, Y.

Nishida, M. Sugawara, A. Saita: “Wide-color-gamut Super Hi-Vision Camera,” ITE Annual Convention 2013, 10-5 (2013) (in Japanese)

(2) S. Tajima, R. Funatsu, Y. Nishida: “Chromatic interpolation based on anisotropy-scale-mixture statistics,” Signal Processing, Vol. 97, pp. 262–268 (2014)

(3) K. Masaoka, T. Yamashita, Y. Nishida, M. Sugawara: “Modified slant-ed-edge method and multidirectional modulation transfer function estimation,” Optics Express, Vol. 22, No. 5, pp. 6040–6046 (2014)

(4) M. Emoto, M. Sugawara: “Still Images of High Spatial Resolution Enable Short TV Viewing Distances,” ITE Transactions on Media Technology and Applications, Vol. 2, No. 2, pp. 185-191 (2014)

(5) ARIB Standard STD-B58 (ver. 1.0), “Interface for UHDTV Production Systems” (2014) (in Japanese)

1.1.2 Cameras

We are researching cameras, including single-chip color compact cameras, video production technology, and wide col-or gamut devices for 8K Super Hi-Vision (SHV). We are also proceeding with R&D on image sensors for full-specification SHV.

■ SHV imaging systemToward the development of a compact SHV camera for prac-

tical use, we fabricated a camera system with more operability by adding a wavelength division multiplexing unit and a cam-era control unit to the compact single-chip color camera head prototyped in FY 2012. We exhibited this system at the NHK STRL Open House, NAB Show 2013, and IBC 2013.

We studied the characteristics of the on-chip color filter ar-


1　Television conveying a strong sensation of reality　｜　1.1 Super Hi-Vision

rays for single-chip color imaging compliant with the color gamut specified in Rec. ITU-R BT.2020. In FY 2013, we devised a method to reproduce the wide color gamut with higher preci-sion by combining signal processing and optical filter charac-teristics (1).

We built an experimental device that can process signals from the 133-megapixel single-chip color image sensor devel-oped in FY 2012. We optimized the operating conditions of this image sensor and confirmed that its signals are output cor-rectly.

We shot Teatro alla Scala’s performance of the opera “Rigo-letto” at NHK Hall and showed the video at a live viewing in September. The videos were made with a silent-operating high-sensitivity SHV camera, named the “theater camera” (2), prototyped in FY 2012. This camera captured the performance without generating any noise that would disturb the audience. We also used the full-resolution camera with the wide color gamut (also prototyped in FY 2012) to shoot SHV evaluation footage.

■ Increasing the sensitivity of full-specification SHV imaging devicesWe have developed image sensors for full-specification SHV

in accordance with Rec. ITU-R BT.2020. Since FY 2012, we have been studying ways to increase the sensitivity of SHV image sensors to the level of current Hi-Vision cameras. We evaluat-ed the sensitivity of the 33-megapixel, 120-Hz frame frequency SHV image sensor we fabricated in FY 2012 with a semicon-ductor process having a 0.11μm design rule (3). We improved the sensitivity by 1.4 times over earlier devices by nanofabri-cating the charge-voltage converter in the pixels and reduced noise from the circuitry. To gain even higher sensitivity, we designed and built a full-specification SHV image sensor by making more extensive use of the nanofabrication semicon-ductor process.

We worked to reduce the power consumption of the analog-to-digital (A/D) converter embedded in the image sensor. We developed a new structure for the converter and proved that it could lower power consumption through simulations and measurements of a prototype.

We also tried to improve the A/D converter characteristics by incorporating a digital error correction function and we ex-perimentally verified the effectiveness of a prototype circuit.

One of our papers from FY 2012 on full-specification SHV image sensors won the internationally prestigious Walter Ko-sonocky Award.

■ Full-specification single-chip color SHV imaging devices We are researching technologies for a single-chip image

sensor that will be part of a full-specification SHV camera (133 megapixels, 120 Hz).

To minimize the effect of the lower sensitivity of a single-chip image sensor with many pixels, we conducted simula-tions on a backside-illuminated structure enabling an optical aperture ratio of 100% and found that it would have 1.65 times the sensitivity of a frontside-illuminated structure.

To evaluate the characteristics of single-chip color imaging devices and enrich the variety of SHV content, we prototyped a single-chip color image sensor with 33 megapixels and mount-ed it on a cube-shaped compact camera head weighing only 2 kg. We used the camera to shoot SHV video footage of an opera for live viewings and a New Year’s Eve music show shot in front of a live audience.

Research on the full-specification SHV imaging devices and the full-specification single-chip color SHV imaging device was conducted in cooperation with Shizuoka University.

[References](1) T. Hayashida, T. Soeno, T. Nakamura, R. Funatsu, T. Yamashita, T.

Yasue, H. Shimamoto, K. Masaoka: “A Study of Spectral Sensitivity Correction of Single-chip Color Camera for Wide Gamut Imaging,” ITE Winter Conference 2013, 11-10 (2013)

(2) T. Nakamura, T. Hayashida, R. Funatsu, T. Soeno, T. Yamashita, T. Yanagi, T. Yoshida: “Development of a Super Hi-Vision camera to-ward shooting theater contents,” ITE Winter Conference 2013, 11-12 (2013)

(3) T. Yasue, T. Hayashida, J. Yonai, K. Kitamura, T. Watabe, T. Kosugi, T. Watanabe, H. Ootake, H. Shimamoto, S. Kawahito: “Sensitivity of 120-Hz Super Hi-Vision Image Sensor using 0.11μm CIS Process,” ITE Annual Conference 2013, 10-4 (2013)

1.1.3 Displays

We are continuing our research on 8K Super Hi-Vision (SHV) direct-view displays and projectors.

■ High frame rate technologiesWith the goal of displaying full-specification SHV video with

light-emitting displays, we continued with our research and development on liquid crystal displays (LCDs) and organic light-emitting diode displays (OLEDs) that can support a 120-Hz frame frequency. In FY 2013, we prototyped an SHV LCD that can reproduce SHV video at 120 Hz. We also estimated the display performance of moving pictures at 120 Hz, on the basis



Figure 1. Shooting test of the theater camera (right)

Figure 2. Cube-shaped compact SHV camera head

of an evaluation of the moving picture resolution of LCDs and plasma display panels (PDPs) (1). In particular, we found that a light emitting aperture time of 25 – 50% per frame is required for the hold-type SHV displays. We began designing an active matrix pixel circuit for a 100-inch diagonal OLED. We conduct-ed circuit simulations to analyze the drive performance of a panel equipped with oxide thin-film transistors (TFTs), which are advantageous to large panels. We determined the design parameters (size of electrodes and mobility of TFT) and circuit structure of the oxide TFT required for the 120-Hz simultane-ous dual-line driving method and confirmed the feasibility of 120-Hz OLEDs (2).

■ Wide-color-gamut SHV projectorWe are developing a projector capable of displaying full-

specification SHV video.To build displays supporting the wide-gamut system colo-

rimetry, we developed a projector with a laser light source, in-stead of an ultra high performance (UHP) lamp, that was based on the projector for displaying 120-Hz frame-frequency SHV video that we developed in FY 2012 (3). We used laser diodes with central wavelengths of 639 nm for red, 532 nm for green, and 462 nm for blue. While leaving the optical system of the projector head as it was, we installed an external light source unit, from which laser light is injected through optical fiber to the projector head. We exhibited a wide-color-gamut SHV system consisting of the wide-color-gamut SHV camera (see 1.1.1) and this projector at the NHK STRL Open House (Figure 1) and demonstrated that it can accurately reproduce highly saturated object colors that Hi-Vision fails to handle.

[References](1) T. Usui, H. Satoh, K. Ishii, Y. Takano, T. Yamamoto: “Evaluation of

Moving Picture Resolution for a Super Hi-Vision Plasma Display Panel,” ITE Annual Conv., No. 5, 16-7, (2013) (in Japanese)

(2) H. Satoh, K. Ishii, T. Usui, Y. Takano, H. Tsuji, T. Yamamoto: “Simu-lation of 120Hz driving performance for SHV OLED Display,” IEICE General Conf., (2014) (in Japanese)

(3) Y. Kusakabe, Y. Iwasaki, Y. Nishida: “Wide-color-gamut Super Hi-Vision Projector,” ITE Annual Convention 2013, 16-1 (2013) (in Japa-nese)

1.1.4 Recording systems　

■ Super Hi-Vision recorderToward our goal of developing video recorders for 8K Super

Hi-Vision (SHV) single-chip cameras, we created compression recording technology using solid state memory and prototyped a compact video recorder (1).

Regarding the compression recording technology, we em-ployed a method in which two green (G) signals of the dual-green format from a single-chip SHV camera are upconverted to full resolution images by using data interpolation and then, these full-resolution images are compressed. This method re-duced the degradation of image quality by about 1 to 2 dB in terms of the peak signal-to-noise ratio (PSNR), as compared with the conventional method which compresses 4K image units for each color. We also developed a high-speed record-ing interface to connect a memory controller and solid state memory, which are separated in our design. The solid state memory is detachable at the interface. By recording of up to 16 channels into the solid state memory in parallel and using large-sized block recording (suitable for recording video), we achieved a recording speed in excess of 12 Gbps per memory package in the solid state memory. With a 25% compression rate, we were able to record a 45-minute SHV video in the re-cording capacity of 2 TB.

Using this technology, we prototyped an SHV studio re-corder (Figure 1) and a camera-integrated recorder. The studio recorder has separate signal processings for encoding and de-coding; thus, it can perform recording and reproduction simul-taneously. This is convenient in that a user can view and check the video while recording. The studio recorder also supports

back-up on a PC and connection to an editing machine, as it employs a format in which compressed SHV video is recorded in frame units. We also equipped the camera-integrated re-corder with a simpler reproduction function than the studio recorder to make it more compact and power-saving. Using the prototype recorders, we demonstrated real-time encoding and decoding of SHV video as well as recording and reproduction of compressed video.

[References](1) E. Miyashita, T. Kajiyama: “Compact Camera Recorder for Super

Hi-Vision,” IEEE International Conference on Consumer Electronics (ICCE), 2014 Digest, pp. 165-166 (2014)



Figure 1. Wide-color-gamut SHV system (left: objects, middle: camera, right: screen)

Figure 1. SHV studio recorder prototype and its memory package

1.1.5 Video Coding

We are researching video compression coding methods for 8K Super Hi-Vision (SHV) broadcasting.

■ HEVC high efficiency video encoder We improved the SHV video encoder using the MPEG-H High

Efficiency Video Coding (HEVC)/H.265 scheme developed in FY2012 and conducted performance evaluations. We made the encoder compliant with the HEVC international standard and added a bit-stream generation function using MPEG-2 Trans-port Stream (TS) multiplexing. We also compared the perfor-mance of the HEVC encoder with that of a conventional MPEG-4 Advanced Video Coding (AVC)/H.264 encoder in objective and subjective evaluations. The objective evaluations showed the advantage of HEVC over AVC on the basis of the peak sig-nal to noise ratio (PSNR) criterion. In contrast, the results of subjective evaluations demonstrated that HEVC scored lower than AVC under some conditions, though it scored higher than AVC in most cases. While the HEVC scheme improves coding efficiency by using large blocks, coarse quantization in these blocks degrades the texture of the original image, causing block noise, at low bit rates. This would be the reason HEVC could not improve the image quality in certain cases. We plan to improve the image quality of HEVC by taking account of the analysis. This research was conducted under the government-commissioned project titled “The research and development project for the expansion of radio spectrum resources” of the Ministry of Internal Affairs and Communications, Japan.

We have begun development of an SHV video decoder using the HEVC scheme and completed the basic design in FY2013. We studied the logic design and circuit size for a function to analyze bit streams and extract image data as well as the in-verse transform and motion compensation functions to gener-ate a decoded image from the bit stream. This research was conducted in cooperation with Mitsubishi Electric Co. Ltd.

We also studied ways to reduce the huge amount of compu-tations entailed by HEVC coding. We found that the choice of prediction modes shows a bias that depends on the reference frame structure of encoded frames. Considering this, we have developed a software encoder that skips the prediction mode selection to achieve approximately a 10% reduction in compu-tations while keeping image degradation at a minimum.

In parallel with these R&D activities, we studied domestic television systems for SHV broadcasting. We devised a cod-ing method to enable 120-Hz frame frequency broadcasting in the future while maintaining backward compatibility with 60-Hz receivers to be used at the start of SHV broadcasting. We also conducted subjective evaluations of multi-format coding (8K, 4K, and 2K) to clarify the bit rates needed for broadcasts (Table 1). We submitted our results to the working group on video coding systems of the Association of Radio Industries and Businesses (ARIB).

■ IP transmission for SHVIn preparation for SHV live viewings of 2014 FIFA World

CupTM in Brazil, we conducted IP transmission experiments in which AVC bit streams were sent between Brazil and Japan in August 2013 and February 2014, in cooperation with NTT and Rede Nacional de Ensino e Pesquisa (RNP) — a Brazilian re-search institution on academic networks. We also performed three live viewings in Japan by IP transmission: “Rigoletto” performed by Teatro alla Scala on September 15 at NHK Sendai station; the ISU Grand Prix of Figure Skating 2013, NHK Trophy on November 8 – 10 at NHK Nagoya station; and the 64th New Year’s Eve Music Show (Kouhaku) on December 31, at Grand Front Osaka.

■ Reconstructive video coding systemWe are developing a video coding system using super-reso-

lution and image reconstruction technologies.To support 120-Hz video, we developed a spatio-temporal

video format converter capable of reducing and reconstructing the spatial resolution and frame frequency simultaneously in real-time. This device, which can handle up to 4K 120-Hz vid-eo, is implemented with a spatio-temporal hybrid super-reso-lution technique that recovers the spatial resolution and frame



7320×4320/60Hz 80〜 100 Mbps 3840×2160/60Hz 30〜 40 Mbps 1920×1080/60Hz 10〜 15 Mbps

Table 1. Estimated bit rate needed for each video format

補助情報符号化

Transmitted video Transmission Spatio-temporal video format converter Received video

4K120Hz

4K60Hz

2K120Hz

4K120Hz

4K60Hz

2K120Hz

2K60Hz

Spat

io-te

mpo

ral r

esol

utio

n re

duct

ion

and

optim

izat

ion

Spat

io-te

mpo

ral h

ybrid

imag

e re

cons

truct

ionLow-resolution video Video stream

Side stream

Local decoded video

Side data

Video encoding

Side data encoding

Side data decoding

Video decoding

Approx. 1 to 10% of video stream

Optimal selection dataSynchronization data

Synchronization of received video and side dataOptimal reconstruction based on selection data

Figure 1. Reconstructive video coding by spatio-temporal data hybrid super-resolution

rate (Figure 1). We also studied a coding method for 120-Hz video combining a transmission-side frame sub-sampling to 60 Hz and a receiver-side frame interpolation. Since the frame interpolation from the interleaved frames may cause image degradation due to mismatches in the motion vector detection, we devised a method for finding the optimal motion vector that gives the best result from among multiple motion vector candi-dates. We confirmed that this method can attain a high-quality frame interpolation.

We also developed a new video transmitting algorithm combining the spatial resolution conversion process with the bit-depth reduction and reconstruction method we had previ-ously developed. The algorithm is optimized for prospective hardware implementations by minimizing the complexity of the constituent processes. With the help of side information, which is optimized by testing the reconstruction process on the transmission side, the algorithm is capable of robust and

precise super-resolution reconstruction and bit-depth restora-tion and can be used to make a high-quality video coding sys-tem that has an extremely high compression rate.

[References](1) Y. Sugito, K. Iguchi, A. Ichigaya, K. Chida, S. Sakaida, Y.Shishikui,

H. Sakate, T. Itsui, N. Motoyama, S. Sekiguchi: “Development of the Super Hi-Vision HEVC/H.265 Real-time Encoder,” SMPTE 2013 An-nual Technical Conference & Exhibition (2013)

(2) K. Iguchi, A. Ichigaya, Y. Sugito, S. Sakaida, Y. Shishikui, N. Hiwa-sa, H. Sakate, N. Motoyama: “HEVC Encoder for Super Hi-Vision,” IEEE International Conference on Consumer Electronics 2014 (ICCE 2014), p.61-62 (2014)

(3) Y. Matsuo, T. Misu, S. Iwamura, S. Sakaida: “Ultra High-definition Video Coding using Bit-depth Reduction with Image Noise Reduc-tion and Pseudo-contour Prevention,” IEEE Visual Communications and Image Processing (VCIP) 2013, FP 489 (2013)

1.1.6 Sound systems providing a strong sense of presence

We are researching three-dimensional 22.2 multichannel sound system for 8K Super Hi-Vision (SHV).

■ SHV sound production systemWe are progressing with our R&D on sound production

systems that can efficiently produce SHV sound. In FY 2013, we added a function to save, read, and adjust frequently used production patterns (templates) to a 22.2 multichannel sound mixing system in order to shorten the sound production time. The new function improved the operational efficiency of mov-ing multiple sound sources from side to side or up and down simultaneously. We enhanced the 3D reverberator function-ality by incorporating 22-directional impulse responses (IRs) measured in concert halls, studios, field and various other places. We also explored ways to obtain all reverberation IRs of 22 directions without actually measuring them and devel-oped a method to generate multiple reverberation IRs by shift-ing the reflected sounds in a single reverberation IR randomly on the time axis (1).

We studied the relation between the distribution of spa-tial and temporal reverberation and the spatial impression of sound in order to make it easy to control the spatial impression SHV sound during production.

We continued with our study on the shape and signal pro-cessing of a compact spherical microphone that can be in-stalled in a camera. In FY 2013, we examined a signal process-ing method to compensate for the deterioration in directivity of a smaller microphone (2) and designed and prototyped a device using this method.

■ FPD-integrated reproduction system for home useWe are researching the reproduction of 22.2 multichannel

sound so that its immersive sensation can be experienced at home. We developed a loudspeaker system in which 12 loud-speakers capable of reproducing 22.2 multichannel sound are mounted the front of the frame surrounding the flat panel dis-play (FPD), not on the side or back of the display. By applying the technology of binaural reproduction, we hope to complete this FPD-integrated home reproduction system by 2016.

We conducted subjective evaluations of the sound images reproduced by the FPD-integrated home reproduction system and found that loudspeakers arranged above and below the display are the most appropriate.

For sound image reproduction at the side and back of the display, we developed a method to expand the listening area by improving the binaural reproduction algorithm used by the system. We also developed a technology to help in front-back discrimination of the sound image by controlling the frequen-

cy amplitude response of sound transmitted to both ears. We analyzed the stability of binaural sound reproduction

over the loudspeakers. The results demonstrated the stability increases as the number of loudspeakers increases and that the 12-loudspeaker frame system is more effective than a two-loudspeaker system.

■ Loudspeaker unit for binaural reproductionWe developed a new loudspeaker unit to improve the perfor-

mance of the FPD-integrated home reproduction system. We changed the structure of full-band unit and enhanced the re-production quality. To make up for the lack of bass sound from the small-bore unit, we developed a compact subwoofer with a wider range of lower frequency sound by increasing the weight of the diaphragm. This research was conducted in cooperation with Foster Electric Company, Ltd.

■ Loudness meter for 22.2 multichannel soundWe prototyped a 22.2-multichannel sound loudness meter to

measure the loudness levels that listeners perceive. We tested the device in an SHV public viewing of the popular New Year’s Eve music show, Kouhaku Uttagassen, and confirmed that it could suppress loudness differences between programs.

■ StandardizationAt the April meeting of ITU-R, we proposed common loud-

speaker layouts for channel-based and object-based sound systems and a framework of Recommendations on sound sys-tems, in cooperation with foreign broadcasters and research institutions. At the November meeting, we proposed a draft new Recommendation on an advanced hybrid sound system combining channel-based and object-based methods includ-ing a specification on the 22.2-multichannel sound loudspeak-er arrangement. The draft was approved as an ITU-R Recom-mendation in February 2014 (3).

At the Association of Radio Industries and Businesses (ARIB), we led the effort toward a studio standard for 22.2 mul-tichannel sound. The standard was established in March 2014.

At SMPTE, our efforts led to re-approval of the ST2036-2 standard for 22.2 multichannel sound.

At MPEG, we proposed adding a 22.2 multichannel configu-ration to the MPEG-4 Advanced Audio Coding (AAC) standard to enable 22.2 multichannel sound broadcasting using the MPEG-4 AAC LC profile. Our proposal was approved as a re-vised International Standard (4).

In accordance with the standardization of SHV broadcast-ing standards, we began development of a 22.2 multichannel



sound encoder using MPEG-4 AAC.

■ “Ultra-reality” meterWe continued with our study on objectively evaluating oth-

erwise subjective factors such as the sense of presence and emotional effect. In FY 2013, we enhanced the model for pre-dicting acoustic impressions from acoustic feature values into a model using time-series data from the start of listening. We found that the prediction accuracy for each instant of time im-proved when we used the top 5% values of time-series data arranged in descending order of acoustic feature values, rather than using their instantaneous values or average values. This research was conducted under contract by the National Insti-tute of Information and Communications Technology (NICT) for its project, “Research and development of ultra-realistic communication technology through innovative three-dimen-sional image technology.”

[References](1) C. Mori, T. Nishiguchi, K. Ono, K. Hamasaki: “Study on the directiv-

ity of microphone for measuring impulse responses used in 3D re-verberator,” Preprint of the Acoustical Society of Japan, 1-P-22, pp. 867-868 (2013 Autumn) (in Japanese)

(2) K. Ono, T. Nishiguchi, K. Matsui, K. Hamasaki: “Spherical micro-phone for Super Hi-Vision 22.2 multichannel sound,” AES 135th Convention, Convention paper 8922 (2013)

(3) ITU-R Rec. BS.2051 “Advanced sound system for programme pro-duction” (2014)

(4) ISO/IEC 14496-3:2009/Amd.4:2013(E) New levels for AAC profiles (2013)

1.1.7 Satellite broadcasting technology

We are researching satellite transmission technologies and multiplexing technologies for 8K Super Hi-Vision (SHV) satel-lite broadcasting.

■ Transmission systems for satellite broadcastingFor ultra-high-definition television broadcasting in the 12-

GHz band, we increased the symbol rate of the satellite trans-mission scheme by reducing the roll-off factor and using 16 Amplitude and Phase Shift Keying (APSK). These improve-ments are part of the ARIB standard, “Transmission system for advanced wide band digital satellite broadcasting” that is under examination by the Information and Communications Council.

We prototyped a transmitter and receiver with a roll-off fac-tor as small as 0.01 and helped to perform the verification tests of their transmission scheme at ARIB. The tests led to ARIB adopting a roll-off factor of 0.03 and symbol rate of 33.7561 Mbaud. We also submitted to ARIB a scheme for 100-Mbps transmissions over a single channel of the satellite transpon-der that uses 16 APSK and a coding rate of 7/9 (an error-cor-recting code parameter).

We conducted transmission tests using the N-SAT-110 and BSAT-3a broadcasting satellites and verified the feasibility of this scheme (Figure 1)(1). The test results were incorporated in a report issued by the broadcasting system committee of the Information and Communications Council. As part of our ARIB related activities, we also studied on-premises cable distribu-tion of ultra-high resolution television satellite broadcasting.

Part of this research was conducted as part of a govern-ment-commissioned project of the Ministry of Internal Affairs and Communications titled, “The research and development project for the expansion of radio spectrum resources”.

We also worked on improving the multi-level coded modu-lation for 8 Phase Shift Keying (PSK). We devised a hybrid set partitioning method, which adjusts the error-correcting capa-bility of each bit by interchanging bits allotted to 8PSK symbols and showed that it could outperform the normal 8PSK system in advanced wide-band digital satellite broadcasting (with a coding rate of 3/4).

Figure 2. FPD-integrated 12-loudspeaker system

Figure 1. Functional verification of production system



Markers indicate code rates of: 9/10, 7/8, 5/6, 4/5, 7/9, 3/4, 2/3, 3/5,

1/2, 2/5, 1/3 (from the top)

Required C/N (dB)

QPSK

π/2 shift BPSK

8PSK

16APSK

32APSK

Info

rmat

ion

bit r

ate

(Mbp

s)

0-5 0 5 10 15 20 25

20

40

60

80

100

120

140

160CS satellite return

Figure 1. Required C/N vs. information bit rate in the ARIB satellite trans-mission test

■ Media transport scheme for multiple delivery channelsWe are researching a media transport scheme using mul-

tiple delivery channels without needing to pay attention to the sort of network (broadcasting and broadband) the channels belong to. This scheme is also intended to accommodate the diversifying terminal environment and channels of viewers.

Synchronized presentation of program components trans-mitted through multiple channels has been difficult for con-ventional media transport schemes such as MPEG-2 Transport Stream (TS), which assumes a single channel, and Real-time Transport Protocol (RTP), which does not support multiplexed transmission of program components and signaling informa-tion. In contrast, a multiplexing technology called Moving Pic-ture Experts Group Media Transport (MMT) can easily com-bine high-resolution video transmitted by broadcasting with other program components transmitted over broadband net-works.

We presented the results of our study on a next-generation broadcasting system using MMT at international conferences(2) and demonstrated the effectiveness of MMT as a multiplexing technology for SHV satellite broadcasting.

We conducted a study on migrating the signaling informa-tion used for current broadcasting systems to MMT. The re-sults were incorporated in a report issued by the broadcasting system committee of the Information and Communications Council.

We also worked on international standardization of our re-search results. Part of our proposal was incorporated in the ISO/IEC 23008-1 international standard on MMT.

■ Advanced satellite broadcasting systemWe studied a micro-strip antenna for feeding power to right-

and left-hand circularly polarized receiving antennas for 12-GHz satellite broadcasting. Computer simulations showed that a cross polarization discrimination (XPD) level of over 25 dB could be obtained by using sequentially rotated four-element

arrays. We are developing engineering models of the transponder

and antenna for a 21-GHz-band broadcasting satellite. We prototyped a wideband band pass filter (BPF) and high-power traveling wave tube (TWT) for the 21-GHz band and conducted thermal tests on them(3). We also fabricated a partial prototype of the beam forming network consisting of a 21-GHz array-fed imaging reflector antenna and a prototype of a 32-element horn array and evaluated its electrical characteristics. We also analyzed the surface distortion caused by temperature fluc-tuations of the reflecting mirror. This research was conducted as part of a government-sponsored project of the Ministry of Internal Affairs and Communications, titled “Research & de-velopment of efficient use of frequency resource for next-gen-eration satellite broadcasting system”.

To reduce waveguide loss in the array feeding unit, we stud-ied a thermal transportation method that could deal with the case in which the TWT and output filter are close to each other. We conducted a thermal vacuum experiment on the TWT and filter using a thermal dummy and found that heat transfer to the satellite bus could be improved by using solid-state materi-als for the thermal filter inserted between the heat-generating part and the onboard equipment panels. This research was conducted in cooperation with the Japan Aerospace Explora-tion Agency.

[References](1) Y. Suzuki, K. Tsuchida, Y. Matsusaki, A. Hashimoto, S. Tanaka, T.

Ikeda, N. Okumura: “ARIB Evaluation Tests of Transmission Sys-tem for Ultra High Definition Television Satellite Broadcasting,” ITE Technical Report, Vol. 38, No. 14, pp.33-38, (2014) (in Japanese)

(2) S. Aoki, K. Otsuki, H. Hamada: “Effective Usage of MMT in Broad-casting Systems,” IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, mm13-11(2013)

(3) M. Kamei, Y. Matsusaki, M. Nagasaka, S. Nakazawa, S. Tanaka, T. Ikeda: “Developments of Wideband BPF for Output Filter of 21GHz-band Broadcasting Satellite,” IEICE General Conference, B-3-15 (2014)

1.1.8 Terrestrial transmission technology

We are researching transmission technologies for the next generation of terrestrial broadcasting that will broadcast 8K Super Hi-Vision (SHV) for fixed reception and Hi-Vision for mo-bile reception over a single channel.

■ Transmission technology for fixed receptionWe are researching large-capacity transmission technolo-

gies for terrestrial broadcasting of SHV. We had previously de-veloped a transmission technology combining ultra-multilevel orthogonal frequency division multiplexing (OFDM) technolo-gy, which has a carrier symbol modulation order of up to 4096, with dual-polarized multiple-input multiple-output (MIMO) technology, which simultaneously uses horizontal and vertical polarizations to transmit twice as much information as sin-gle-input single-output (SISO) and measured its transmission characteristics in field experiments.

In FY 2013, we applied transmit diversity technology using space time coding (STC) to single frequency network (SFN) in which two transmitting stations cover a certain area with sig-nals of the same frequency. This method reduced the degrada-tion caused by overlapping reception of waves from two trans-mitting stations. We demonstrated over-the-air transmission of SHV signals using the STC-SFN method at the NHK STRL Open House. To improve the bit error ratio (BER) performance, we designed a circuit that performs a log likelihood ratio (LLR) iterative computation at every sum-product decoding. The cir-cuit performed better than conventional LDPC decoding meth-ods. We studied a spatially coupled low density parity check

(LDPC) code featuring a parity-check matrix with diagonally arranged nonzero elements, devised codes with various code rates, and verified their performance through computer simu-lations. Regarding the carrier modulation scheme, we studied a non-uniform constellation (NUC), in which signal points are distributed non-uniformly and confirmed through computer simulations that the NUC is superior to a conventional uniform constellation. To increase the transmission capacity, we de-signed a circuit to extend the OFDM bandwidth by about 214 kHz (1/28×6 MHz).

■ Long-distance SHV transmission test at Hitoyoshi experimental station

We installed an experimental transmission station for du-al-polarized MIMO transmission (Table 1) at NHK’s TV relay station in Hitoyoshi City, Kumamoto Prefecture and began measurements of the MIMO channel characteristics for long-distance transmissions.

We measured the received electric field strength, delay pro-file, condition number, etc., at 52 sites in Hitoyoshi with the reception antenna set at different heights. At 30 of these sites, we also measured the error rates and the required carrier-to-noise ratio (C/N). We also performed the world’s first long-distance SHV terrestrial transmission(1). The SHV signal was encoded at 91 Mbps by using the MPEG-4 AVC/H.264 scheme and broadcast from the experimental transmission station to a receiver 27 km away in Yunomae Town, Kumamoto Prefec-ture. The video decoded and displayed at the reception point



showed that SHV content could be transmitted stably over the long distances covered by current digital broadcasting (Figure 1). This research and development is being performed under the auspices of “Research and Development of Basic Technol-ogy Encouraging Effective Utilization of Frequency for the Next Generation Broadcasting System” program and it funded by the Ministry of Internal Affairs and Communications, Japan.

■ Transmission technology for mobile receptionWe are studying 2×2 MIMO-OFDM, which uses two anten-

nas each for transmission and reception, for the purposes of mobile reception.

In FY 2013, we conducted computer simulations to compare the mobile reception characteristics of STC-MIMO, which en-codes and transmits same data from the transmission anten-nas, and Space Division Multiplexing (SDM)-MIMO, which transmits different data from the transmission antennas, at the same transmitting efficiency (number of bits per carrier × error correction coding rate) (2). The results showed that superiority reverses between STC-MIMO and SDM-MIMO at a transmit-ting efficiency of 3 bit/Hz in a mobile reception environment with a 20-Hz maximum Doppler frequency. We also devised a new decoding method that shuffles the decoding order of the maximum-likelihood decoding with QR decomposition. The method improved the characteristics of SDM-MIMO by 0.9 dB, compared with conventional methods (Figure 2).

The transmission characteristics of MIMO transmission

method deteriorate if the two channels from each transmis-sion antenna to each reception antenna are similar. In FY 2013, we measured the channel characteristics for a case in which two pairs of transmission antennas or reception antennas are set at some distance from each other and for a case in which they are installed at the same location and the vehicle drives on a road under non-line-of-sight conditions and analyzed the

Experimental station

Current terrestrial digital broadcasting・Modulation scheme: 64QAM・Use single polarization

Experimental station (UHF46ch)・Ultra-multilevel modulation (4096QAM)・Use dual polarization (Dual-polarized MIMO)

Dual-polarized receiving antenna

4096QAM

64QAMLong distance (27km)

Twice of terrestrial digital broadcastingTwice of terrestrial digital broadcasting

⇒

Dual-polarized transmitting antenna

Large-capacity transmission (approx. 4 times of current digital broadcasting)

Figure 1. Comparison of the experimental station and current terrestrial digital broadcasting

Item

Modulation schemeOccupied bandwidthTransmission frequencyTransmission powerCarrier modulationFFT sizeGuard interval ratio

Error correction code

Transmission station

OFDM5.57 MHz671.142857 MHz (UHF 46ch)Horizontal polarization: 10W Vertical polarization: 10W BPSK, QPSK, 16QAM, 64QAM, 256QAM,1024QAM, 4096QAM8k, 16k, 32k, 64k 1/8, 1/16, 1/32 Inner code: LDPCCoding rate r=2/3, 3/4, 5/6Outer code: BCHEstablished at NHK Hitoyoshi TV relay station

Specifications

Table 1. Experimental station specifications

Transmitting efficiency (bits/Hz)

Req

uire

d C

/N (d

B)

STC−MIMOSDM−MIMO (Conventional method)SDM−MIMO (Proposed method)

30

25

20

15

10

5

00 1 2 3 4 5 6 7 8 9 10

Figure 2. Required C/N of STC-MIMO and SDM-MIMO at the same trans-mitting efficiency



space correlation indicating the similarity of channels. The re-sults showed the median value of the spatial correlation is 0.2 or less regardless of the distance between the antennas, which means the channel characteristics are different, and this dem-onstrated the effectiveness of MIMO transmission.

We are studying a method to multiplex transmission signals for fixed reception and mobile reception using frequency divi-sion multiplexing. With this method, however, it is necessary to compensate for the degradation caused by inter-carrier-inter-ference (ICI) from Doppler effects during mobile reception. In FY 2013, we used a basis expansion model for the channel es-

timation and conducted computer simulations to evaluate the mobile reception characteristics when belief propagation is used for ICI compensation. The results showed that is scheme improved the required C/N by 3 dB at 100 km/h compared with the conventional channel estimation without ICI compensa-tion and that it is an effective compensation method (Figure 3).

■ Advanced Television Systems Committee (ATSC) 3.0 standards

We submitted a proposal for ATSC 3.0, the U.S. standards for next-generation terrestrial digital broadcasting (3), and con-tributed to its standardization activities. Our proposal includes an OFDM scheme with a segmented structure capable of hier-archical transmission, which is the feature of Integrated Ser-vices Digital Broadcasting-Terrestrial (ISDB-T), dual-polarized MIMO and MISO, ultra-multilevel modulation up to 4096QAM, spatially coupled low-density parity check (LDPC) code, NUC mapping, inter-polarization interleaving and dispersion, and the Space Time Coding (STC) - Single Frequency Network (SFN) method.

[References](1) NHK INFORMATION: “8K (Super Hi-Vision) Long-distance transmis-

sion test is successfully achieved,” http://www.nhk.or.jp/strl/eng-lish/data/20140203.pdf

(2) Y. Narikiyo, H. Miyasaka, M. Nakamura, H. Sanei, M. Takada: “Com-parison of SFBC and SDM MIMO reception performance under mo-bile reception environment by computer simulations” ITE Technical Report, Vol. 38, No. 5, pp. 113-116 (2014) (in Japanese)

(3) ATSC: “Summaries of Responses to ATSC 3.0 Physical Layer Call for Proposals,” http://www.atsc.org/cms/index.php/the-news/327-sum-maries-of-responses-to-atsc-30-physical-layer-call-for-proposals

1.1.9 Wired transmission technology■ Optical transmission of uncompressed SHV signals

We are researching a system to transmit uncompressed Super Hi-Vision (SHV) signals over optical fibers to avoid de-lays or image degradation resulting from image compression. The system is for transmitting program contributions within and between broadcast stations or from relay destinations to broadcast stations.

In FY 2013, we developed a method to transmit a 72-Gbps full-resolution SHV signal over eight wavelengths (10 Gbps per wavelength). Long-distance transmissions over optical fiber are prone to errors that degrade signal quality. To enable lon-ger distance transmissions, we developed a technology to add error-correcting code to the signals. With this technology, the maximum transmission distance can be extended from 80 km to approximately 100 km.

For our study on SHV transmission over Ethernet, we de-veloped a control method to reduce the jitter of video clock

reproduced by the receiver and demonstrated its effectiveness through simulations and experiments using 10-gigabit Ether-net (10 GE).

We also contributed to a study on interfaces for UHDTV production systems that is being conducted by a Ultra-high-definition TV studio facilities working group at the Association of Radio Industries and Businesses (ARIB).

■ Cable TV transmission of SHV signalsWe are researching a method to transmit partitioned SHV

signals over multiple channels (1) so that SHV programs can be distributed through existing coaxial cable television networks. In FY 2013, we designed the service information structure of transmission and programs to receive the signals correctly and gave the results of our study to the Japan Cable Television Engineering Association (JCTEA).

We also developed a method to time-multiplex multiple digi-tal broadcasts in a baseband signal and transmit it over an op-tical fiber, as a way to distribute digital broadcasting to homes over FTTH (Fiber To The Home). In FY 2013, we prototyped test equipment to transmit multiple programs including SHV sig-nals over an optical network using 10 GE technology.

[References](1) Y. Hakamada, N. Nakamura, T. Kurakake, T. Kusakabe, K. Oyamada:

“UHDTV (8K) Distribution Technology and Field Trial on Cable Tele-vision Networks,” ITE Trans. on MTA, Vol. 2, No. 1, pp. 2-7 (2014)

Moving speed (km/h)

Req

uire

d C

/N fo

r BER

=10-7

af

ter L

DPC

dec

odin

g (d

B)

FFT size: 32kCarrier modulation method: QPSKError correction: LDPC codeCoding rate: 1/2Transmission frequency: 600MHz

with plain channel estimationwith ICI compensation

15

10

5

00 50 100 150

Figure 3. Characteristic improvement due to ICI compensation

Receiver

Transmitter

Reproduced video signals

Regenerated clock frequency for video signals

Video signal transmission system using Ethernet

Figure 1. Clock regeneration test in Ethernet transmission



We are continuing our research and develop-ment of more natural and viewable three-dimen-sional television that does not require special glasses, with the goal of developing a new form of broadcasting delivering a strong sense of pres-ence.

The integral method, which uses an array of many tiny lenses for capture and display, requires imaging equipment with a large number of pixels to generate high-quality 3D images. We prototyped image capture and display devices in FY 2011 that were based on a video system with an equivalent of 8,000 scan lines. We have been working on the elemental technologies to achieve higher resolu-tion and a wider viewing zone since FY 2012. In FY 2013, we built equipment to capture partitioned viewing zones with multiple cameras as well as compact equipment with the lens array adhering to the image sensor. We conducted a study on ex-panding the viewing zone by combining a num-ber of small projectors and making a large screen from multiple flat displays.

Our study on the reproduction quality of inte-gral 3D images started in FY 2012. In FY2013, we measured the ocular accommodation response to 3D images and confirmed that the accommoda-tion conforms to the depth of the 3D images. We

also conducted subjective evaluations using com-puter simulations on the relation between display parameters such as the pitch and focal length of lenses in the array and the resolution, and we ob-tained basic data relating the depth and resolution of 3D images.

Coding of the integral method has been included in a subject of study by the MPEG-Free-viewpoint Television (FTV) ad hoc group, which started its activities in July 2013. We subsequently began to study coding technologies for the integral method and decided to participate in activities of the ad hoc group.

We have been studying multi-viewpoint video technology since FY 2012 as a way of generating integral 3D images from 3D CG models of objects built from images taken by multiple cameras and researching its application to video production. To improve the accuracy of the 3D models, we built a camera array consisting of multiple cameras and infrared cameras in FY 2013. We also conducted experiments on generating integral 3D images from images captured by multi-viewpoint robotic cameras and improved the stability and speed of processing. This technology was successfully used in live sports broadcasting.


2.1.1 Integral 3D television

We are researching a form of spatial imaging technology, called the integral method, as a way of creating a more natu-ral and viewable 3D television that does not require special glasses. In FY 2013, we studied ways to improve the quality of 3D images and their reproduction.

■ Increasing 3D image qualityGenerating high-quality 3D images using the integral meth-

od requires many more pixels than for ordinary 2D imaging, both for capture and display. We studied ways to make 3D im-ages with more pixels by using multiple imaging devices. For capture, we used two sets of an image sensor and a lens ar-ray. This structure, however, suffers from a loss of information and a difference in luminance level between image sensors. To solve this problem, we interpolated the missing information and reduced the luminance level difference. Thus, we were able to use a large number of pixels, two image sensors worth, to make 3D images (1). For the display, we prototyped equipment consisting of four projectors (Figure 1). We found that we could expand the viewing zone (the areas in which the 3D image can be viewed) by a factor of 2.5 in the horizontal direction and 2.0 in the vertical direction, by using projective and affine trans-formations to correct the distortion in the images from each projector (2).

Part of this research was conducted under contract with the Ministry of Internal Affairs and Communications for its project titled, “R&D on spatial information acquisition system using



Lens array

4 projectors

Figure 1. Integral display equipment using four projectors

multiple image sensors.”

■ Reproduction quality of 3D imagesWe have been studying the quality of 3D images reproduced

using the integral method since FY 2012. The integral method reproduces the light rays from an object and forms the equiva-lent 3D image in the air near the viewer. It is expected to have the advantage of not straining the viewers’ eyes, because the ocular accommodation of viewing such a 3D image is said to be at the same depth as when one observes the actual object.

In FY 2012, we measured the ocular accommodation re-sponse when viewing integral 3D images with both eyes in or-der to verify the relation between the accommodation position and the image’s depth. The results showed that the accommo-dation conforms to the depth of the 3D image. Meanwhile, the accommodation is directed to the point of regard by moving the right and left eyeballs. To eliminate this factor, we mea-sured the accommodation response in viewing integral 3D im-ages with a single eye in FY 2013. The results confirmed that the accommodation has a tendency to conform to the image’s depth position even with monocular vision (3).

To evaluate the reproduction quality of 3D images against the display parameters of the integral method, we conducted

subjective evaluations on image quality versus the depth of the image by using the pixel pitch of an image displayed by a lens array as a parameter. The experiment provided us with enough data to derive the relation between the pixel pitch and the reproduction range of the depth. We plan to conduct evalu-ations with a wider range of parameters to further study the relationship between pixel pitch and the reproduction range of the depth.

We began a study on coding technologies for the integral method in FY 2013 and decided to participate in the standard-ization activities of the MPEG-FTV ad hoc group, which studies the integral method’s coding and other subjects.

[References](1) H. Hiura, T. Yamashita, J. Arai: “The Expansion of the Capturing

Area for Integral Imaging using Two Imaging Units,” ITE Technical Report, Vol. 38, No. 11, pp. 27-30 (2014) (in Japanese)

(2) N. Okaichi, H. Hiura, M. Miura, J. Arai: “Integral 3D display with multi-projection system using distortion compensation,” ITE An-nual Convention 2013, p.11-5 (2013) (in Japanese)

(3) H. Hiura，T. Mishina，J. Arai, Y. Iwadate: “Accommodation response measurements for integral 3D image,” Proc. SPIE, Vol. 9011, 9011-48 (2014)

2.1.2 Generating 3D content from multi-viewpoint images

We are studying ways to acquire integral 3D images of ob-jects that are difficult to capture with optical equipment, such as very distant or very large objects.

In FY 2012, we studied how to capture objects with infrared patterns projected on them using a camera array consisting of two infrared cameras and two color cameras (Figure 1) and estimate the depth information. In FY 2013, we improved the stability and accuracy of the depth estimation by extending the two-dimensional smoothing filter it uses to the temporal dimension to make the three-dimensional spatial-temporal smoothing filter (1). We also studied how to generate 3D models by integrating depth information gotten from multiple camera arrays.

We continued our work from FY 2011 on generating 3D models of objects from multi-viewpoint images. In FY 2013, we developed a new method to improve the accuracy of the 3D models by evaluating the reliability of the depth estimations obtained from the multi-viewpoint images and combining only the areas of the images that are free of estimation errors (2). This method reduced the depth estimation errors and enabled the generated integral 3D images to appear to show camera work such as zooming and panning. The multi-viewpoint im-ages in this study were captured by a multi-viewpoint robotic camera (3).

Regarding this robotic camera, we implemented a “weak” camera calibration method that does not require special pat-terns in FY 2012. In FY 2013, we improved the accuracy of this camera calibration. Time slice video (appearing as if one cam-era had captured footage of the object while rotating around it) captured by the multi-viewpoint robotic cameras was used during the production of the “Figure Skating 2013, NHK Tro-phy” program.

[References](1) K. Hisatomi, K. Ikeya, M. Katayama, Y. Iwadate, K. Aizawa: “Depth

Estimation by Cost Volume with Spatial- temporal Cross-based Lo-cal Multipoint Filter using Projecting Infrared Patterns,” European Conference on Visual Media Production (CVMP), IO (2013)

(2) K. Ikeya, K. Hisatomi, M. Katayama, Y. Iwadate: “Integral 3D Con-tents Production from Multi-View Images: 3D Modeling and 3D Im-age Conversion of Sports Scene,” ITE Journal, Vol. 67, No. 7, pp. 229-240, (2013) (in Japanese)

(3) K. Ikeya, K. Hisatomi, M. Katayama, Y. Iwadate: “Multi-Viewpoint Robotic Camera System,” MIRU2013, OS4-5 (2013)


2 Three-dimensional images　｜　2.1　Three-dimensional images

　Figure 1. Camera array

Infrared projector

Color camera

Infrared camera


3.1.1 Hybridcast

We are continuing our research on Hybridcast, which is a service platform to provide flexible and extensible new servic-es that take advantage of the characteristics of broadcasting and telecommunications networks. In FY 2013, we released the first version of this service to the public and continued with R&D to make the service more efficient and standardization. We showed the appeal of Hybridcast by demonstrating it to visitors at the NHK STRL Open House and other exhibitions at home and abroad.

■ Practical use of HybridcastThe IPTV Forum published the first edition of Hybridcast

technical specifications in March 2013. (1)(2)(3) Operational speci-fications for receivers were also created and standardized, and manufacturers started selling TV sets supporting Hybridcast in June. NHK launched Hybridcast services (Figure 1) in Sep-tember 2013 and followed up by launching more in December.

Prior to the launch, NHK STRL had participated in Hybridcast promotion projects and contributed to the studies on services to be offered by NHK, facilities and equipment, and operational

NHK STRL is studying technologies that take advantage of broadcasting and telecommuni-cations networks to offer new services and en-hance programming.

We participated in at IPTV Forum’s standard-ization activities that led to Hybridcast techni-cal specifications being issued in March 2013. TV sets based on these specifications came to mar-ket not long after, and NHK launched Hybridcast services in September. We also worked on ad-vanced Hybridcast and Hybridcast for 8K Super Hi-Vision broadcasting. In addition, we contrib-uted to the standardization of advanced Hybrid-cast at IPTV Forum Japan.

To provide more attractive hybrid services, we are conducting R&D on technologies that use program related information and social network-ing services (SNS). We studied methods to direct Internet users to various programs and related services by linking the broadcasters’ program-related information with open data. We also stud-ied ways to promote TV viewing by analyzing the viewing behaviors of SNS users. We developed a system to allow users to share CG content while communicating with each other through an SNS. We studied a technology to provide program-re-lated information matching viewers’ preferences by monitoring their viewing behavior and infer-ring their interests. We developed a method to analyze the viewers’ behavior based on gazes, prototyped a viewing behavior monitoring sys-

tem, and experimentally studied the relation be-tween gazes and viewers’ interests.

Along with our efforts to improve the conve-nience of hybrid services, we are developing technologies to provide secure and reliable ser-vices. We developed a new authorization scheme to control different license management systems for broadcast and telecommunications servic-es in an integrated framework. We developed a verification system for the next-generation Con-ditional Access System (CAS) and contributed to standardization activities at the Information and Communications Council. As a cryptogra-phy technology to provide more secure and reli-able services, we developed an attribute-based encryption scheme with which personal infor-mation is stored securely in the cloud and only service providers who meet certain conditions can decrypt the information. To protect content, we developed a cryptography method capable of tracing the key owner.

For program distribution over a communi-cations network, we developed a distributed broadcasting system that enables easy time-shift viewing of past programs over a network and a push content delivery method to distrib-ute information via communication channels in synchronization with broadcast content. We also developed a hybrid live streaming system linking peer-to-peer (P2P) and content delivery network (CDN) distribution.



Figure 1. Example of Hybridcast services

aspects. We demonstrated a wide variety of Hybridcast servic-es together with commercial broadcasters and pay TV service providers at the NHK STRL Open House 2013, using Hybrid-cast compliant receivers developed in cooperation with man-ufacturers and telecommunications carriers. We exhibited a “home screen” service, active program guide, scroll news, and “TV60 Minogashi-Natsukashi,” a service to provide past pro-grams on demand. . We also exhibited an interactive picture drawing application (Figure 2), musical instrument sessions application (Figure 3), multilingual closed-captions and the TV note service, which TV links tablet devices and synchronizes service provision with the progress of programs. We improved the public’s understanding of Hybridcast and showed the fea-sibility of various services. Examples of Hybridcast services to be offered by commercial broadcasters were also exhibited (4).

Dissemination of Hybridcast and the growth of its services will require an environment to enable participation by a wide range of application developers. Whereas a versatile HTML authoring tool can be used to develop the Hybridcast applica-tions themselves, which are written in HTML 5, an operational verification is also needed to preview how the developed ser-

vice functions on various devices. Thus, we developed an ap-plication development support tool capable of previewing the linkage of programs and portable devices on a PC and exhib-ited it at the NHK STRL Open House (Figure 4) (5).

■Technologies for advanced HybridcastWe are studying the technical specifications for the next ver-

sion of Hybridcast (called advanced Hybridcast) and Hybrid-cast for the 8K Super Hi-Vision era.

The new functions include: a system model of non-broad-cast managed applications allowing various service providers other than broadcasters to participate; a recording and re-production functionality; precise synchronization using time stamps; and a sophisticated device linkage interface.

We developed an application distribution management sys-tem (Figure 5) to study non-broadcast managed applications. It can register and distribute applications and issue signature keys to authenticate them. We exhibited a system model for other service providers than broadcasters to provide applica-tions at the NHK STRL Open House.

With the launch of 8K Super Hi-Vision (SHV) broadcasting in 2020 in mind, we studied possible 8K Hybridcast services that could take advantage of the large, extremely high-definition screen of this new form of broadcasting (Figure 6). We exhib-ited a live broadcasting service for covering marathons at the NHK STRL Open House. A high-definition map to show the current location is displayed together with multiple windows showing video coverage of the race and “tweets” from social networking services (SNS) provided by broadcasting and tele-


3　Convergence of broadcasting and telecommunications3.1 Technologies for the convergence of broadcasting and telecommunications

TV screen

Tablet display

Figure 2. Example of Hybridcast services at the NHK STRL Open House (Interactive picture drawing application)

Tablet devices

TV screen

Figure 3. Example of Hybridcast services at the NHK STRL Open House (Musical instrument sessions application)

Application preview window

Preview management window

Figure 4. Application development support tool

Figure 5. Application distribution management system

communications. Together with other divisions involved in the development

of SHV, we studied the SHV data broadcasting and designed a reference model of the protocol stack based on MPEG Media Transport (MMT) we also investigated data coding, transmis-sion methods and closed captions.

■ Hybridcast standardizationAt ITU-R, we contributed to a draft new Recommendation

for the general and technical requirements of hybrid broad-

casting and drafted a report on the different forms of hybrid broadcasting in the world, including Hybridcast and Europe’s HbbTV. In the Future of Broadcasting Television (FOBTV) group, we participated in a study on a system model based on the use cases submitted to date. At the Technical Plenary/Ad-visory Committee Meeting (TPAC) of the World Wide Web Con-sortium (W3C) working on HTML5 standardization, NHK re-ported on the establishment of Hybridcast specifications at the domestic IPTV Forum Japan in March 2013 and on the launch of Hybridcast services in September. These reports conveyed the advantages of Hybridcast to the world. On the domestic front, we worked on standardization of advanced Hybridcast and HTML5 browser specifications at the IPTV Forum Japan.

We will continue with our research on more sophisticated services and standardization for Hybridcast. We will also in-vestigate hybrid services for the SHV age and standardize SHV data broadcasting.

[References](1) IPTVFJ STD-0010, “Integrated Broadcast-Broadband System Specifi-

cation Version 1.0”, IPTV Forum Japan(2) IPTVFJ STD-0011, “HTML5 Browser Specification Version 1.0”, IPTV

Forum Japan (3) M. Takechi: “Technical Specifications and Trend of Hybridcast,”

NHK STRL R&D, No. 142, pp. 12-19 (2013) (in Japanese)(4) A. Baba, H. Oomata, K. Matsumura, M. Takechi, S. Sunasaki: “Pro-

totype Hybridcast Services at NHK STRL, Open House 2013,” ITE Technical Report, Vol. 37, No. 41, pp. 17-20 (2013) (in Japanese)

(5) K. Majima, K. Matsumura, Y. Hironaka: “Development of Hybrid-cast Application Development and Distribution System,” NHK STRL R&D, No. 142,pp. 20-27 (2013) (in Japanese)

3.1.2 SNS and Program viewing analysis

We did R&D on technologies to make use of program related information and social networking services (SNS) to provide more attractive services that exploit the advantages of broad-casting and telecommunications.

■ Information network for broadcasting programsWe studied ways to direct Internet users from one program

to various services or other programs by utilizing broadcast-ers’ program guide information and related information and to form a community using SNSs to bring excitement to the viewing experience.

We devised a method to convert the program guide informa-tion in the Electronic Program Guide (EPG) to the Linked Open Data (LOD) format, link it with open data such as National Diet Library Authority Data and Twitter, and associate programs with program related information and social information (1). We prototyped an application to guide users from one program to related programs and information by web-based visualiza-tion (Figure 1).

For behavior analysis of users, we analyzed the results of experiments on the “teleda” social TV platform developed in FY 2011 and found that users fall into different types, such as those who prioritize their own preferences and those who are influenced by others’ behavior. Based on the analysis, we com-piled service design guidelines.

We also devised a method to estimate users’ viewing trends such as viewing times and programs viewed from their tweets on Twitter and use those estimates to categorize users. Iden-tifying influential users on SNSs based on the categorization and providing them with effective program recommendation services should promote TV viewing among SNS users.

■ Virtualization of video content productionWe studied virtualization of video content production based

on the server side video creation so that Internet users can easily produce their own video content.

We built an SNS-style CG content creation system (Figure 2) (2), in which SNS users can communicate through videos produc-tion using the servers on the network and exhibited it at the NHK STRL Open House. This system allows users to re-create and share content derived from already posted content. Users



Broadcast videoMap

SNS

Ranking data displayed

Live distribution

Figure 6. Example of 8K SHV Hybridcast services (Marathon live broad-casting)

Figure 1. Program-related information navigation application with open data

can easily produce derived content by using a graphical user interface (GUI). We conducted experiments with about 100 us-ers after we had posted sample content and demonstrated the system and method. Regarding linkage with Hybridcast, we developed a system to display sign-language CG animations on tablet devices concurrently with a broadcast and exhibited the system at the NHK STRL Open House. We also developed a simple TVML (TV program Making Language) player by using Web GL. It enables real-time runnning of TVML scripts sent with broadcasting without installing special software on the terminal.

We also worked on enhancing the functionality of the CG creation engine. We developed a new TVML player using the latest graphics library. The player can display more characters through the use of high-speed processing and improve high-quality translucent display.

■ Acquisition and analysis of viewing behaviorsWe are researching technologies for the automatic acquisi-

tion and analysis of viewing status and the estimation of view-ers’ interests, in a domestic TV viewing environment with a view to developing new services that provide program-related information matching viewers’ preferences.

We developed a method for measuring a viewer’s body movements and head pose and a machine learning system that estimates whether the viewer is gazing at the TV screen

or not. Experiments showed that gaze behavior can be judged accurately (3). We also developed a method that combines Bayesian inference with an estimation based on the intensity of facial expressions, which has improved the accuracy of fa-cial expression recognition. The research on gaze behavior es-timation and on facial expression recognition were conducted in cooperation with Tokyo City University and the University of Tokyo, respectively. Part of the research on gaze behavior estimation was conducted under a government-commissioned project from the Ministry of Internal Affairs and Communica-tions, titled “Strategic Information and Communications R&D Promotion Programme (SCOPE).”

We prototyped a viewing behavior monitoring system incor-porating the method of estimating gaze behavior (Figure 3). We used the system to study the relation between gaze behav-ior and inferred viewer interests through TV viewing experi-ments. The results demonstrated the possibility of inferring viewers’ interests by correlating gaze behavior with closed caption data that appears on screen during gaze periods. We also developed a method to selectively extract keywords from closed caption data that belong to particular classes derived from Wikipedia category information. This method enabled selective extraction of keywords identifying the names of peo-ple and places, which are often for the focus of viewer interest.

[References](1) K. Ariyasu, T. Nakagawa: “A Proposal of TV Program Information

Network Formation Method by Linked Open Data,” Proc. of FIT2013, no.2, F-035, pp. 363-364，(2013) (in Japanese)

(2) M. Douke, H. Kaneko, S. Inoue: “SNS Style CG Content Creation Sys-tem Enables Users to the Easy Chain of Re-areation”, Proc. of ITE Annual Conference, 2-3, 2013 (in Japanese)

(3) Takahashi, Clippingdale, Okuda, Yamanouchi, Naemura, Shibata: “Automatic rating of video contents based on viewer’s behaviors in living space,” SITIS2013, pp. 6-13 (2013)

3.1.3 Security technologies

To support the convergence of broadcasting and telecom-munications, we are researching integrated authentication federation technology for broadcast and telecommunications, a next-generation conditional access system (CAS), cryptogra-phy, and device-linked technology.

■ Integrated authentication federation technology for broadcast and telecommunicationsWe made progress with our R&D on an integrated authenti-

cation technology to enable secure and easy use of personal-ized services by linking license management schemes used in

broadcast and telecommunications services. Depending on the types of channels and terminals used, dif-

ferent methods are used by broadcasters and telecommunica-tions providers for access control and content protection. To view the same content in different environments, individual contract management is required for each service. To simplify this situation for subscribers to these services, we developed a license distribution management system embodying a new authorization scheme that merges different license manage-ment methods - a conditional access system for broadcast ser-vices and a digital rights management system for telecommu-nications services - in an integrated framework (1). A prototype



Figure 2. SNS-style CG content creation system

Figure 3. Example of gaze behavior estimation system in operation

system based on this method was exhibited at the NHK STRL Open House (Figure 1). This system enables content viewing rights obtained from a single user authentication to be carried over to different services of broadcasting and telecommunica-tions. To put it to practical use, we also made proposals to incorporate the next-generation conditional access system in commercial standards.

We developed the “TV note” system to provide personal-ized services securely while effi ciently collecting information on viewers’ interests through a linkage with authentication services. The system was exhibited as a service model of Hy-bridcast at the NHK STRL Open House. We also studied the specifi cations of the authorization scheme and a device link-age method in order to apply the authentication framework to actual services.

■ Next-generation CAS technologyWe are researching next-generation CAS technology for the

SHV test broadcasting in 2016. In FY 2013, we developed a veri-fi cation system and contributed to the standardization of the scrambling subsystem and associated information subsystem for ultra-high defi nition television at the Information and Com-munications Council. The scrambling subsystem uses a cryp-tography algorithm on the transmission side, and the associ-ated information subsystem uses a conventional conditional access system to manage the contract information in a way that is consistent with the access control functions of current services. For ease of maintenance and improvement of securi-ty, we also devised a downloadable conditional access system that sends the software updates to the receivers involved in processing associated information through broadcasting and telecommunications networks. These methods were refl ected in a report issued by the Broadcasting System Committee of the Information and Communications Council.

■ Cryptography algorithm for hybrid services

We researched cryptography technologies to provide secure and reliable hybrid services taking advantage of broadcasting and telecommunications. Service providers need information about viewers in order to personalize services. Meanwhile, viewers should not receive unexpected services from un-known providers. An effi cient way to meet these requirements is to store viewers’ information on a cloud. We studied an at-tribute-based encryption scheme (2) that encrypts the viewer’s information to protect his/her privacy and allows only the pro-viders meeting a policy set by the viewer based on attributes such as reliability of service provider to decrypt the informa-tion. By modifying the way of describing attributes, we found a way to use a smaller ciphertext and secret key. We evaluated the encryption on a computer. This research was conducted in cooperation with the Advanced Industrial Science and Tech-nology.

In our research on cryptography to protect content, we de-veloped a method that can trace the owner of a decryption key that has been illegally copied. To deal with possible cryptanal-ysis, we developed a way to update the scrambling method that that will work on existing as well as new receivers without increasing the transmission volume.

■ Interactive content technology for device linkage servicesWe studied technologies to enable broadcast content to be

shown on devices other than receiver equipped TVs Called “Augmented TV”, this technology captures images of the TV screen with a camera of a mobile terminal and processes the captured image on the mobile terminal. It provides a new viewing experience by displaying computer graphics (CG) of program-related characters on the mobile terminal in accurate synchronization with the TV screen.

In FY 2013, we built a system to produce and present pro-gram and associated CGs in a virtual space displayed on a mo-bile terminal. The system works in an easy and integrated way by using TV program Making Language (TVML) (3). We also achieved frame-level precision synchronization (under 0.03 s) between the TV image and the mobile terminal by using a mo-bile terminal’s camera to capture “synchronization markers,” graphic time data, on the broadcast content shown on the TV display.

[References](1) C. Yamamura, Y. Nishimoto, A. Fujii: “Development of an authoriza-

tion system for license delivery control - application of its system to CAS/DRM -”, ITE Annual Conference, 18-11 (2013) (in Japanese)

(2) N. Attrapadung, G. Hanaoka, K. Ogawa, G. Ohtake, H. Watanabe, S. Yamada: “Ciphertext-Policy Attribute-Based Encryption over Range Attributes,” SCIS, 2E4- 1 (2014)

(3) H. Kawakita, H. Kaneko, S. Inoue, A. Fujii, T. Nakagawa: “Production Platform of 3DCG Content Connected with TV Program - System Ar-chitecture for Augmented TV -”, ITE Annual Conference, 2-5 (2013) (in Japanese)

3.1.4 IP content delivery technology

With the aim of distributing programs over a network, we are researching a distributed server-based system for provid-ing broadcast programs over a network and IP content delivery technologies for future hybrid services.

■ Distributed server-based broadcasting systemWe are researching a distributed server-based broadcasting

system for time-shift viewing of past programs over IP net-works. The accumulation, reprocessing and distribution of broadcast content is handled by servers so that viewers can

view any of the previously programs as they like. In FY 2013, we upgraded the time-associated zapping sys-

tem prototyped in FY 2012 to enable program zapping by tags such as cast name, in addition to zapping by on-air time. We developed a database to manage tags associated with broad-cast on-air time and incorporated it in a distributed fi le sys-tem capable of storing and retrieving video, sound and related information of programs using broadcast time as ID. The re-sulting acquisition of associated tags in synchronization with video and audio lets viewers zap quickly without stress. We also developed a program viewing player that shows the tags

Broadcast content viewing

Telecommunications content viewing

EMM server

Tablet terminal

Receiver with built-in downloadable CAS content viewingEMM: Entitlement Management Message (Broadcast license)DRM: Digital Rights Management (Telecommunications license)

Authentication / authorization server

VOD server / DRM server

Figure 1. Integrated access control system for broadcast and telecom-munications



associated with a program in addition to information for time zapping (Figure 1) and a system for collecting viewing logs and visualizing them by video frame. We will use this system to acquire viewing data categorized by broadcast on-air time and tag.

In addition, we advanced with our development of a pro-gram viewing system that converts the resolution of programs stored in the distributed file system on the server side accord-ing to the display capabilities of the viewing terminal.

■ Hybrid content distribution technologiesWe are researching a push content delivery method to en-

able synchronized distribution over communication channels of the information to be incorporated and displayed in each video frame of broadcast content (Figure 2).

We began study in FY 2012 on a distribution method to con-trol the timing of sending information on the distribution serv-er so that the information arrives to many viewers at the same time. In FY 2013, we devised a distribution control method to minimize the difference in arrival times (jitter) of periodically distributed information in a quasi-network environment. The results of simulations showed that this method suppresses jit-ter in a large-scale distribution service (1).

■ Hybrid P2P-CDN live streaming technology

In light of the results of live streaming experiments at the London Olympics in 2012, we developed a hybrid live stream-ing system linking peer-to-peer (P2P) distribution and content delivery network (CDN) distribution (Figure 3). A distribution support server enables terminals not supporting P2P to receive live streaming video from a terminal supporting P2P. We also developed a method that uses this server to suppress differ-ences in the video reproduction timing between terminals receiving P2P distributed content and the terminals receiving CDN distributed content. We evaluated the system implement-ed on a commercial cloud and confirmed that it worked (2).

[References](1) S. Tanaka, S. Nishimura, Y. Endo: “A Study of Determining Standard

RTT in Push Content Delivery Method for Synchronization between Broadcast Programs and Push Contents,” IEICE General Confer-ence, B-6-38 (2014) (in Japanese)

(2) S. Nishimura, S. Tanaka, Y. Endo: “Development of a Hybrid P2P-CDN Live Streaming System,” ITE Winter Annual Convention, 12-9 (2013) (in Japanese)



Broadcast

Telecommunications

Player name

Hybridcast receiverPush

Push delivery by high-precision synchronization

Synthesized display on broadcast content

Program related information (Player name, etc.)

Figure 2. Example of Push content delivery method

For zapping by time

For zapping by tag

Figure 1. Program viewing player for time-shift zapping system

Live encoder Cache server Distribution support server P2P distribution server

P2P stream

Direct receptionP2P network

P2P-compliant terminalsNon-P2P-compliant terminals

CDN Distribution edge

Request transfer

Segment reception

Reception request

Distribution status monitor

Direct reception

Distribution status monitor

Figure 3. Hybrid P2P-CDN live streaming system

4.1 User-friendly information presentation

NHK STRL is researching user-friendly information media that will allow people with vision or hearing impairments to get more information from broadcasts.

■ Haptic display technology of nonverbal informationWe are researching technologies to convey information

that is difficult to convey in words, such as diagrams and the shapes of artworks, to people with visual impairments by us-ing 2D and 3D information that can be apprehended by the tactile and kinesthetic senses.

Our work on tactile presentation of 2D information includes a method to convey an understanding of diagrams and graphs to people with visual impairments by vibrating a tactile display. With this method, we found that vibration per line segment in addition to vibration per facet can reduce the time required for people to grasp spatial positions in diagrams and line charts. For the powered mechanical leading method that conveys the overall composition of a diagram by guiding the fingers, we derived conditions that can be placed on the leading speed to reduce errors in spatial position recognition.

We developed a method to present 3D information using a multi-point force presentation method that gives stimuli to the finger. We conducted subjective evaluations with visually impaired people and objective evaluations using the length

The start of NHK’s Hybridcast is one step in our work to enhance broadcasting services. At the same time, it is important to develop the means for all viewers, including the disabled, the elder-ly, and non-native Japanese speakers, to be able to receive and enjoy its content. We are conduct-ing research on using information technology to deliver the benefits of broadcasting to all people in forms that are appropriate for them.

In our research on user-friendly presentation of information, we have studied tactile and hap-tic presentation methods for visually impaired people. We evaluated a method to facilitate un-derstanding of 2D information and a method to present 3D shape information by giving multi-point stimuli to fingers. For hearing-impaired people, we upgraded our Japanese sign-language CG translation system for weather reports.

We refined the speech recognition technology used for closed-captioning of information pro-grams and succeeded in improving the recogni-tion rate to 95.9%. We also developed an algo-rithm that uses news scripts for auto correction of recognition errors so that closed caption ser-vices can be economically offered by local broad-casting stations.

In our research on speech and audio signal pro-cessing for the elderly, we upgraded the equip-ment to adjust program sound at the receiver. We refined the speech synthesis technology toward the realization of automatic reading services of broadcast scripts and data broadcasting texts and started research on the speech processing technology to enable various emotional expres-sions.

In our research on natural language process-ing for barrier-free services, we developed tech-nologies to improve the efficiency of the rewrit-ing process of the news conversion assistance system into easy Japanese for non-native Japa-nese speakers in Japan and continued with our research on technology to automatically convert programming into easy Japanese. We also devel-oped a natural language analysis system to cate-gorize viewers’ opinions on broadcast programs.

In our research on content retrieval and rec-ommendation, we built a content map connect-ing programs with various relations by analyzing program description texts and studied a method to give reasons for retrieving and recommending programs. The program retrieval technology has been incorporated into a Hybridcast application. We also researched “Video bank” to assist in the retrieval and processing for raw video footage by adding metadata useful for video production to the raw video footage. The analysis technology we developed was used for managing raw video footage at a local station.

In our research on estimating the psychologi-cal state of viewers, we conducted experiments to clarify the cognitive characteristics and mea-sure brain activities related to the sensation depth caused by high-resolution video such as Super Hi-Vision. We also continued with our re-search on the analysis of psychological states while watching video by measuring the brain activities of viewers and research on advanced technologies to prevent negative effects of video on the human body.


4 User-friendly information presentation

Figure 1. Multi-point force presentation method

of finger tracks as an evaluation indicator, in order to iden-tify how the number of stimulus points affects cognition of the ridge line. The results showed that using four stimulus points is preferable (1). We also evaluated the effect of the degrees of freedom of the stimulus points when force is applied in the x, y and z directions on tactile cognition of curved surfaces. The results showed that at least one stimulus point needs to have three directional degrees of freedom. This result can be used as a design guideline for building presentation equipment. Part of this research was conducted in cooperation with the Univer-sity of Tokyo.

■ Sign-language CG translation technology for weather informationWe are researching technology to automatically translate

Japanese weather reports (text) into sign language computer graphics.

We continued our work from FY 2012 on improving the ac-curacy of translation by using example-based and statistical machine translation of clauses or phrases and upgraded our Japanese sign-language CG translation system. We prototyped a translation system in which the presentation speed of the sign-language CGs is adjustable so that the CG can be played in synchronization with the original weather report. The sys-tem was exhibited at the NHK STRL Open House. We also de-veloped a system to release the Japanese and sign-language dictionary used in our translation technologies on NHK Online to allow anyone to access the service (Figure 2). This system will make the results of our studies helpful for people learning sign language and will make it easier for us to gather opinions on the quality of sign-language CGs and use them to make improvements to them (2). Part of this study was conducted in cooperation with Kogakuin University.

[References](1) T. Handa, T. Sakai, T. Shimizu: “Recognition of Three-dimensional

Geometry using Multipoint Force feedback on a Fingertip,” IEICE Technical Report, vol. 113, no. 347, WIT2013-61, (2013)

(2) NHK Sign language CG Website, http://www.nhk.or.jp/signlan-guage/

4.2 Speech recognition for closed captioning

We are researching speech recognition for expanding the range of programs covered by real-time closed captioning so that more people including the elderly and those with hearing difficulties can enjoy TV programs.

■ Closed captioning for information programs

The re-speak method, in which automatic speech recogni-tion is used on a speaker in a quiet environment who is re-peating the speech of a program, is used to produce closed captions for sports programs such as sumo wrestling as well as information programs such as “Asaichi”. Closed captioning for information programs covering a wide range of topics re-quires a language model having word-sequence probabilities that change daily with each new topic and an acoustic model compliant with the changing frequency distributions of vowels and consonants of the re-speaker. These frequency distribu-tions may change when the speaker becomes fatigued during the re-speak task.

In FY 2013, we trained the language model by adding about 150 million words including closed caption texts for informa-tion programs with a broad range of topics. We also incor-porated a discriminative acoustic model trained to take ad-vantage of complementary searches with the language model and optimized the search method. We developed a method of training the language model by adapting it to topics (1). In an experiment, these efforts reduced the recognition error rate of re-spoken speech selected from the “Asaichi” program from 7.3% to 4.1%.

■ Closed-captioning for news programsWe helped to install speech-recognition closed-captioning

systems for news programs (2) in the NHK Osaka, Nagoya, Fu-kuoka and Sendai stations. These systems are being used for closed captioning of local programs. We also developed an al-


4　User-friendly information presentation4.1 User-friendly information presentation　｜　4.2 Speech recognition for closed captioning

Figure 2. Sign-language CG website

7.36.1

5.34.5 4.4 4.1

0

2

4

6

8

Recognition error rate (%)

At th

e en

d of

M

arch

201

3

+ Ad

ditio

n of

15

0 m

illion

wor

ds

+ D

iscr

imin

ativ

e a

cous

tic m

odel

+ Se

arch

opt

imiz

atio

n

+ To

pic

adap

ting

lang

uage

mod

el

+ El

imin

atio

n of

rest

atem

ent

Figure 1. Reduction of recognition errors in re-spoken speech for informa-tion program

gorithm to automatically correct recognition errors by using news scripts. This algorithm will be implemented in a closed captioning system at smaller local stations (3).

[References](1) Y. Fujita, T. Oku, A. Kobayashi, S. Sato: “Language Model Adap-

tation by Constrained Non-negative Matrix Factorization,” Spring Meeting of the Acoustical Society of Japan, 2-9-7 (2013) (in Japa-nese)

(2) A. Kobayashi, Y. Fujita, T. Oku, S. Sato, S. Homma, T. Arai, T. Imai: “Live Closed-Captioning System Using Hybrid Automatic Speech Recognition for Broadcast News,” NAB Proceedings, pp. 227-283 (2013)

(3) S. Sato, K. Onoe, A. Kobayashi, T. Oku, Y. Fujita, M. Ichiki: “An er-ror correction algorithm by using a WFST built from news scripts,” Spring Meeting of the Acoustical Society of Japan, 2-4-8 (2014) (in Japanese)

4.3 Speech and audio signal processing for the elderlyTo provide easy-to-hear audio services for everyone in-

cluding the elderly, we are researching technologies to adjust program sound at the receiver and to synthesize good-quality speech.

Earlier, we had developed a technology to adjust the balance of speech and background sound arbitrarily (1). In FY 2013, we made the algorithm less computationally intensive in order to make it more practical and developed equipment that incor-porates the algorithm and speech rate conversion technology. Experimental evaluations showed the effectiveness of this equipment (Figure 1).

In our research on high-quality speech synthesis and pro-cess technologies, we had previously developed a method to synthesize high-quality speech of arbitrary sentences by using massive amounts of speech data recorded from news programs. We also had developed automatic reading equip-ment and used it in broadcasts of the “Stock market report” program on NHK Radio 2. In FY 2013, we improved the speech synthesis technology and began study on speech process tech-nology to convey emotional impressions and thereby expand the range of expression of synthesized speech. To improve the performance of speech synthesis technology, we modified the previous database method, which directly handles acoustic and linguistic features of speech data. By changing the search unit to strengthen the connection of unvoiced consonants and silences that have less auditory influence, we increased the sequentiality of phoneme strings segmented from text (2). To stabilize the synthesized sound, we studied a way to switch to statistical models built by using machine learning of features when no phoneme string in the database.

In our work on speech process technology, we prepared an emotional voice database for adding emotional expressions to individual synthesized speakers. We collected speech judged to convey certain emotions and devised rules for adding emo-tional expressions to speech by analyzing the differences be-tween the acoustic features of speech conveying no emotion and the features of emotional speech (3). We also developed

a speech conversion technology that applies these rules. We conducted experimental evaluations that verified the emotion-al expression rules could correctly alter unemotional speech to a particular emotion 60% of the time.

We incorporated our results in a system for automatically reading stock market reports. This resulted in lowering the minimum unit of nominal prices to sen (1/100 of one yen) from yen, extending the maximum readable number from less than 10 million to less than 100 million, and allowing a market overview to be read. We also worked on an automatic weather report reading system. We manually checked all the waveform extension and contraction units obtained from a prior analysis of speech data and corrected errors in order to avoid reduc-tions in sound quality due to the speech rate conversion. The improved automatic stock market reading system began op-erations at the end of March (Figure 2).

[References](1) Komori, Imai, Seiyama, Takou, Takagi, Oikawa: “Development

of volume balance adjustment device for voices and background sounds within programs, for Elderly people,” AES 135th Convention Paper 9010 (2013)

(2) Segi: “Search-unit selection for concatenative speech synthesis systems,” Proceedings of 2014 Spring Meeting of Acoustical Society of Japan

(3) Seiyama, Segi, Imai, Takagi: “Evaluation of speech database for emo-tional speech synthesis,” ITE Winter Annual Convention 13-4, 2013


4　User-friendly information presentation4.2 Speech recognition for closed captioning　｜　4.3 Speech and audio signal processing for the elderly

Figure 1. Evaluation experiments

Figure 2. Automatic reading system

4.4 Language processing for barrier-free services

■ Japanese translation/conversion assistance tech-nology

We are researching technology to assist in the task of con-verting news program scripts into easy Japanese for non-na-tive Japanese speakers in Japan. The “NEWSWEB EASY” test site started in FY 2012, which provides regular news text re-written into easy Japanese, began its full-scale operation in FY 2013. Five news articles are released per day. To shorten the rewriting time, we developed a method to automatically assigning detailed information used for showing readings in furigana on top of kanji, displaying dictionary data for difficult words and highlighting proper names, achieving the accuracy of 95% or higher (1).

We also began research in FY 2013 on methods to divide a news script sentence into several short sentences and convert modifier phrases in independent sentences (2) in order to au-tomatically convert news scripts into easy Japanese. We pre-pared text data required for this study, such as examples of rewritten news scripts.

■ Opinion analysis technology We are researching technology to analyze opinions of view-

ers of a program in order to utilize them in program produc-tion. In FY 2013, we developed an algorithm to automatically detect program names in Twitter messages about programs and built a system to analyze tweets in a structured and ex-haustive way using the algorism (Figure 1).

The official name of a program or a hashtag indicating a pro-gram name is not explicitly mentioned in most of the tweets about TV programs. So, we developed a method to identify which program is mentioned in a tweet. We focused on three elements: a program name including its abbreviation and the names spelled in different ways, program description, and ex-pressions indicating viewing of the program, and devised a method to combine those three elements. This enabled detec-tion of the program mentioned in tweets with 77% accuracy, even if the official program name or hashtag is not included (3).

The program name detection algorism requires handmade rules to cover program names in abbreviation and different spellings. As these rules are used in combination with the algorisms to detect program description and expressions of viewing status, setting rules for the names of regular program serial will be enough to detect individual program name. This made the maintenance of rules easy and saved workload for continuous feedback analysis.

[References](1) H.Kumano, H.Tanaka: “Online learning of tagging to Japanese doc-

uments with dependent Dirichlet process,” 20th annual convention of ANLP, pp.1075-1078, (2014) (in Japanese)

(2) I.Goto, H.Kumano, H.Tanaka: “Analysis on manual rewrite from normal Japanese news to easy Japanese news,” 20th annual con-vention of ANLP, pp.15-18, (2014) (in Japanese)

(3) M.Hirano, K.Kanbe, T.Kobayakawa: “Automatic detection of TV program mentioned in tweets-For the overall and continuous de-tection-,” ITE Annual Convention 2013, 3-7, (2013) (in Japanese)

Various tweets about TV programs

Tweet Tweet

Tweet

Tweet

Tweet

Tweet

Tweet

Tweet

Tweet

Tweet

Information / Tabloid show

Documentary / Culture

Cartoon / Special effects

Hobby / Education

Theater / Performance

News / Report

Drama

Music

Movie Welfare

Variety show Kids

Language

Sports

Variety of programs

Results of structured and exhaustive analysis

Positive Neutral Request RecommendNegativeStand-by Plan OthersViewing

Program A 421 tweets

Total number of tweets (1,322)

Program B 81 tweets

Program C 87 tweets

Program D 733 tweets

0% 50% 100%

Figure 1. Tweet analysis system


4　User-friendly information presentation　｜　4.4 Language processing for barrier-free services

4.5 Content retrieval and recommendation technology

■ Content utilization technology using language infor-mation We are researching technology that uses text information to

manage, retrieve, and recommend broadcasting content. In our study on managing large amounts of content, we de-

veloped a method to identify the topics of a program and in-fer the relations between the topics and the program by using program descriptions in the Electronic Program Guide (EPG) (1). Using the obtained topics and relations and the semantic relations between words we had developed earlier, we built a content map to connect programs (Figure 1). The content map can manage programs with relation names, for example, “treatment” and “prevention” for a program with the topic of “high blood pressure.” We also developed a technology to ex-tract relations between programs from viewer feedback about programs. Experiments with actual feedback demonstrated that the method can identify many useful relations that are not included in the program descriptions; for example, it can de-termine if the same singer’s songs are used in two different programs.

In our studies on retrieving and recommending content, we devised a method to extract words indicating the reason for retrieving and recommending a program from the program description (2). The effectiveness was experimentally demon-strated in tests on a week’s worth of programs to be broadcast.

The results of these studies are being utilized by the Hybrid-cast application “Minogashi, natsukashi” launched in Decem-ber 2013 (Figure 2).

■ “Video bank” for utilizing raw video footageTo make use of raw video footage stored in video archives,

we are researching a video asset management system, called “Video bank,” which uses physical sensor and video analysis technologies to automatically add metadata useful for video production.

We developed a hybrid sensor consisting of small attitude and distance sensors created using semiconductor manu-facturing technology to acquire camera motion information (camera parameters) required for video synthesis. We also im-proved its operability by adding error compensation function-ality using pattern recognition. We also developed a method to acquire more robustand precise camera parameters that uses the hybrid sensor and a camera parameter estimation method that works by analyzing video footage.

In our work on video processing technology, we developed a method to interactively extract a specified object region in video. We improved the usability of the object extraction func-tion by preprocessing the region segmentation that requires a high calculation cost and by devising an extraction algorithm that takes into account the size of the pre-segmented regions to make sure an appropriate region matching the user’s direc-tions is extracted.

Regarding our work on video retrieval technology, we made the discrimination of objects more precise by analyzing nu-merous feature points, the relations between these points, and their positions in a region of the image (3). We devised a technology for extracting key frame images containing an en-tire object by using the region segmentation method and the image features of individual regions. We also studied video retrieval technology using the similarity of images. Here, we improved the retrieval accuracy of cuts, which is the minimum shooting unit, by evaluation of the similarity of the main ob-ject regions. After that, we developed a technology to extract a series of cuts shot at the same place (i.e., scenes) by using the appearance frequency of similar pictures together with a technology to retrieve scenes.

We used these video retrieval and object extraction tech-nologies to improve our earthquake disaster metadata system. We also built an experimental system with functions for get-ting and processing video from archive tapes by adding and correcting metadata. The system has been used at the NHK Fukushima station since the end of September 2013, and it automatically added metadata that would enable retrieval-by-shot to more than 10,000 video tapes related to earthquake disasters in only three months (Figure 3).


4　User-friendly information presentation　｜　4.5 Content retrieval and recommendation technology

Related programs

Figure 2. The “Minogashi, natsukashi” Hybridcast application

PC for retrieval Data storage disk

PC for processing LTO drive

Figure 3. Earthquake disaster metadata complementary system being tested at NHK Fukushima station

Figure 1. Content map

Part of this research was commissioned to NHK Engineering System, Inc. The research was conducted in cooperation with the Shimizu Corporation.

[References](1) K. Miura, I. Yamada, T. Miyazaki, N. Kato, H. Tanaka: “Generating

a TV Program Map Using Semantic Relations between Words,” In Proceedings of the 76th National Convention of IPSJ, 5C-4 (2014) (in Japanese)

(2) T. Miyazaki, A. Matsui, I. Yamada, N. Kato, M. Naemura, H. Sumiyo-shi: “ICA based keywords extraction for TV program recommenda-tion,” In Proceedings of ITE Annual Conference 2013, 12-4(2013) (in Japanese)

(3) Y. Kawai, M. Fujii: “Semantic Concept Detection based on Spatial Pyramid Matching and Semi-supervised Learning,” ITE Transac-tions on Media Technology and Applications, vol. 1, no. 2, pp. 190-198 (2013)

4.6 Viewers’ mental state estimation technology

To understand how the viewers watch TV programs and how they are psychologically influenced by them, we are re-searching ways of estimating their mental state from subjec-tive evaluations and brain activity.

In FY 2013, we evaluated the sensation of the depth of objects in natural (i.e., not artificially shaded) scenes with varying res-olution to explore the psychological effects of high-resolution images such as those of Super Hi-Vision. The results showed that the depth sensation increases as the resolution becomes higher, and this finding corroborates the results of our previ-ous experiments in FY 2012 on artificially shaded images (1). To see what neural mechanism underlies the depth sensation, we measured brain activities when viewers make a judgment about the depths of objects in images of varying resolution. The results demonstrated that a specific brain area that is well-known to be specialized for spatial and motion perception can be associated with the judgment of monocular depth (2).

In our research on analysis of mental states of persons while they are watching video, we measured the brain activities of subjects while they viewed a program by using functional magnetic resonance imaging (fMRI) equipment and analyzed them using machine learning in order to estimate changes in mental state through time. The results of an experiment on persons watching comedy TV programs showed that viewers are in a mental state of expecting the development of the pro-gram before they actually find it funny and that such a mental state is measurable (3) (Figure 1).

To prevent video from having undesirable effects on the hu-man body, we developed a method to estimate the degree of unpleasantness of shakiness, flicker and striped patterns in video and experimentally demonstrated its effectiveness.

[References](1) K. Komine, Y. Tsushima, N. Hiruma: “Higher-resolution image en-

hances subjective depth sensation in natural scenes,” Perception, Vol. 42, supplement, p. 119 (2013)

(2) Y. Tsushima, K. Komine, N. Hiruma: “Cortical area MT+ plays a role in monocular depth perception,” Perception, Vol. 42, supplement, p. 149 (2013)

(3) Y. Sawahata, K. Komine, T. Morita, N. Hiruma: “Decoding humor experiences from brain activity of people viewing comedy movies,” PLoS One 8: e81009 (2013)


4　User-friendly information presentation4.5 Content retrieval and recommendation technology　｜　4.6 Viewersʼ mental state estimation technology

Program video

Intensity of percieving humor

Brain activities

Continuous evaluation by slider

fMRI

Perception of Humor

-10 -8 -6 -4 -2 0 2 4 6 8 10

-10 -8 -6 -4 -2 0 2 4 6 8 10

Time (sec)

Predict a future humor experience

Figure 1. Prediction of humor experiences from brain activity analysis

5.1 Contribution transmission technology

5.1.1 120-GHz-band FPU for uncompressed SHV

We are continuing our research on a 120-GHz-band Field Pick-up Unit (FPU) for wireless transmission of uncompressed 8K Super Hi-Vision (SHV) signals using polarization multiplex-ing. In FY2013, we conducted wireless transmission experi-ments and improved the FPU.

We evaluated the performance of the compact 120-GHz-band radio frequency (RF) unit (No. 1) prototyped in FY 2012 and confirmed that it could transmit outdoors (1). We exhib-ited wireless transmission of uncompressed dual-green format SHV signals with the No. 1 unit and the FY 2011 prototype at

the NHK STRL Open House. Based on the results of performance evaluations of the No.

1 unit, we prototyped a compact 120-GHz-band RF unit (No. 2) with improved temperature and frequency characteristics (2). We also built a baseband (BB) signal processing unit that mul-tiplexes 16 HD-SDI signals transmitted with forward error cor-rection; the unit outputs two 11-Gbps-class signals. The result-ing FPU (Tx/Rx) consists of compact RF units (No. 1 and No. 2) and the BB signal processing unit (Figure 1) and can transmit uncompressed SHV signals.

In preparation for the launch of 8K Super Hi-Vision (SHV) broadcasting in 2020, we are devel-oping program contribution transmission equip-ment for producing high-quality content and performing high-capacity transmissions. We are also researching technologies to build contribu-tion networks using Internet Protocol (IP) and enhancing Integrated Services Digital Broad-casting-Terrestrial (ISDB-T).

In our research on wireless transmission of un-compressed SHV signals, we improved the tem-perature characteristics of 120-GHz-band Field Pick-up Units (FPUs) with polarization multiplex-ing. Field transmission experiments conducted under various conditions demonstrated the FPUs could operate in outdoor environments.

In our research on bidirectional FPUs for high-speed wireless transmission of file-based video, we implemented a time-division duplex (TDD) scheme in hardware and experimentally evalu-ated the transmission characteristics. We also studied the packet structure of a hybrid auto-matic repeat request (HARQ) scheme combining forward error correction and automatic repeat request.

The Japanese government has required that FPUs operating in the 700-MHz band be phased out in favor of ones that operate in the 1.2-GHz and 2.3-GHz bands. For this migration to the new bands, we have devised a multiple-input multi-ple-output (MIMO) transmission scheme using space-time trellis code that can stably transmit video at 35 Mbps rate in a mobile environment. We submitted a proposal on this scheme to the Information and Communications Council, and the Council later issued it as a technical standard for these frequency bands.

In our research on radio microphones, which the government has also required to be migrated to “white” space between channels or the 1.2-GHz band, we conducted transmission experi-

ments indoors and outdoors on transmitting un-compressed linear pulse code modulation (PCM) audio and a multi-channel operation method. The Association of Radio Industries and Busi-nesses (ARIB) subsequently issued the technical specifications of these methods as a standard.

In our work on millimeter-wave mobile cam-eras for studios and on-location reporting, we studied ways to reduce the amount of compu-tations of the multiple-input multiple–output (MIMO) transmission scheme and to equalize channel characteristics in the frequency domain by using a single carrier method. To improve the reception characteristics of the millimeter-wave mobile camera, we developed an elliptical horn antenna with improved horizontal gain by reduc-ing the antenna’s vertical half-value angle from 50° to 20°. The camera was used in golf relay programs.

We are researching technologies for Internet Protocol (IP) based contribution networks to en-able faster and more reliable contribution trans-missions. We developed a technology to transmit video stably even with wireless fading channels, such as wireless LAN, along with technologies to mitigate latency and simultaneously transmit multiple raw video footage using IP networks.

In our research on enhanced ISDB-T, we are studying an Area One-Seg system that can pro-vide urgent information to specific reception ar-eas during disasters and other emergencies. In September, we conducted field experiments on a system that allows local governments and other organizations to provide urgent information in times of disaster by using service-area-limited broadcasting. The study was conducted at an ex-perimental transmission station in Yatomi City, Aichi Prefecture. The results showed that the service area could be limited as designed, and the demonstration showed the effectiveness of the system.


5 Advanced content production technology

We conducted wireless transmission experiments in the field using this FPU. We confirmed stable wireless transmis-sion of an uncompressed SHV signal over a distance of 250 m under heavy rainfall conditions (60 mm/h) (3).

[References](1) J. Tsumochi, F. Suginoshita, S. Okabe: “120-GHz-band FPU for SHV

Signal Transmission,” ITE Technical Report, Vol. 37, No. 34, pp. 45-48, (2013) (in Japanese)

(2) S. Okabe, J. Tsumochi, F. Suginoshita: “Temperature Characteris-tics of 120-GHz-band Wireless Link,” IEICE General Conference, No. B-5-119, (2014) (in Japanese)

(3) J. Tsumochi, S. Okabe, F. Suginoshita, J. Takeuchi, H. Takahashi, A. Hirata: “Field Experiments on Super Hi-Vision Signal Transmission using 120-GHz-band FPU,” IEICE General Conference, No. C-2-111, (2014) (in Japanese)

5.1.2 Bidirectional Field Pick-up Unit (FPU) transmission technology

We are researching bidirectional FPUs for high-speed wire-less transmission of file-based video. In FY 2013, we studied the time-division duplex (TDD) scheme, implemented it in hard-ware, and experimentally evaluated its transmission character-istics.

■ Bidirectional FPU experimental equipment proto-type and evaluationWe evaluated the performance of the experimental micro-

wave-band (6 to 7 GHz) bidirectional FPU prototyped in FY 2012 (Figure 1) and improved its functions. Using a propagation delay generator, we verified that the proposed TDD method operates with the maximum time use efficiency without being affected by fluctuations in the transmission delay. We also conducted wire-less transmission tests using a dual-polarized parabolic anten-na with a splash plate in a radio wave anechoic chamber and confirmed the feasibility of one-way transmission at up to 180 Mbps over a distance of 50 km (1). We improved the transmission and reception characteristics of the radio frequency (RF) unit and implemented a high-precision automatic frequency control (AFC) function and an adaptive modulation function.

■ Media Access Control (MAC) layer packet structureWe have been studying Type-II hybrid automatic repeat re-

quest (HARQ) since FY 2012. HARQ is an automatic repeat re-quest scheme combining with forward error correction codes. “Type-II” means that a transmitter retransmits different par-ity bits from the first transmission. In FY 2013, we confirmed through computer calculations that transmitting efficiency does not decrease significantly even if the retransmission packet size is 1/4 of the first transmission and the packet con-trol field is included (2). We also implemented the HARQ scheme in hardware.

[References](1) K. Mitsuyama, N. Kogo, F. Uzawa, N. Iai: “Prototyping and perfor-

mance evaluation of TDD-based 2x2 MIMO-OFDM transceiver,” IEEE RWW2014, MO3D-1, pp. 43-45 (Jan. 2014)

(2) F. Uzawa, K. Mitsuyama, K. Aoki, T. Hiraguri: “An Evaluation on HARQ Scheme for Bi-directional FPU with Consideration of Pack-et Control Information,” IEICE General Conference 2014, No. B-5-134(2014) (in Japanese)

5.1.3 700-MHz-band frequency migration

In response to the Ministry of Internal Affairs and Commu-nications action plan for spectrum reallocation, NHK STRL is preparing to change the frequencies used by our FPUs and

wireless microphones. In particular, we want to achieve a smooth migration away from the 700-MHz band to the 1.2- and 2.3-GHz bands.


5　Advanced content production technology　｜　5.1 Contribution transmission technology

Baseband signal processing unit

Compact 120-GHz-band RF unit set on a camera tripod

Figure 1. 120-GHz-band FPU structure for uncompressed SHV (Receiver side)

Transmission/reception control unit 1

Transmission/reception control unit 2

Transmission/reception RF unit 1

Transmission/reception RF unit 2

VH

VH

TRC1 TRC2TRH1 TRH2

Dual-polarized transmission/reception antenna

Bidirectional 2×2 MIMO using

dual-polarization

Dual-polarized parabolic antenna with a splash plate (60cm in diameter, 30 dBi gain)

Transmission/reception control unit (TRC1)

Transmission/reception RF unit (TRH1)

Figure 1. Bidirectional FPU experimental equipment prototype

■ FPU transmission technology

We are researching multiple-input multiple-output (MIMO) transmission schemes using space-time trellis codes (STTCs) for FPUs used in mobile reporting to increase the transmission capacity beyond that of current schemes. In FY 2013, we de-veloped 2×4 STTC-MIMO transmission equipment capable of diversity reception using four reception antennas to deal with long delay waves and propagation loss in the 1.2- and 2.3-GHz bands. We conducted transmission experiments using the new frequencies in Kyoto, Hiroshima, and on the Lake Biwa road race course, and obtained data to make preparations for the frequency migration, such as the MIMO reception characteris-tics for a dual-polarized Yagi-Uda antenna. We also contributed to standardization activities at the Association of Radio Indus-tries and Businesses (ARIB); the STTC-MIMO scheme we have been studying was incorporated in one of its standards (1). This standard enables stable transmission at the 35-Mbps video bit rate recommended by ITU, even in a mobile environment.

To support large-scale reporting using many reception base stations, such as is done in covering road races, we devel-oped transmission equipment using the 2×(16,4) STTC-MIMO scheme, which selects for demodulation the four most appro-priate reception signals from a maximum of 16 reception sig-nals (Figure 1).

■ Radio microphone transmission technologyThe frequencies used by specified radio microphones, i.e.,

those subject to the Radio Law in Japan, will be migrated to the 1.2-GHz band or the white space of terrestrial TV broadcasting. The 1.2-GHz band is available throughout Japan, but it is used for radiolocation (radar). In order to share this band, the new microphones will need a digital method resistant to interfer-ence.

Current radio microphones use a single carrier QPSK scheme with audio latency of 1 to 3 ms (occupied bandwidth: 288 kHz

or 192 kHz). We are researching the use of orthogonal frequen-cy division multiplexing as a way to provide stable transmis-sions resistant to interference and multipath as well as low latency. An OFDM digital system with an occupied bandwidth of 600 kHz can transmit uncompressed linear PCM audio at a latency of under 1 ms.

In FY 2013, we performed transmission experiments in the studio and outdoors, on OFDM digital radio microphones with a 600-kHz occupied bandwidth. We also built hand-held and two-piece models (Figure 2). These prototypes can operate for about four hours on two AA batteries, and they show the fea-sibility of reducing the power consumption of the OFDM digital system.

At the same time, we verified the operation of prototype OFDM digital systems with occupied bandwidths of 288 kHz and 192 kHz and using various transmission parameters as a way to enable multi-channel operation. The OFDM digital sys-tems with occupied bandwidths of 600 kHz, 288 kHz, and 192 kHz were incorporated in an ARIB standard (2) on transmission schemes for specified radio microphones.

[References](1) ARIB: “1.2GHz / 2.3GHz-band Portable OFDM Digital Transmission

System for Television Program Contribution”, STD-B57 ver. 2.0 (in Japanese)

(2) ARIB: “Specified Radio Microphone for Land Mobile Radio Station (TV White Space Band, 1.2GHz Band,” STD-T112 ver.1.2 (in Japa-nese)

5.1.4 Wireless contribution link technology

For the next generation of contribution links supporting 8K Super Hi-Vision (SHV), we are developing reliable large-ca-pacity wireless cameras and FPUs. To this end, in FY2013, we studied multiple-input multiple-output (MIMO) transmission technology and single carrier-frequency domain equalization (SC-FDE) and improved the performance of our millimeter-wave mobile camera.

■ MIMO transmission technologyTo reduce the volume of computations used to detect MIMO

signals, we devised a maximum-likelihood decoding method that suppresses the number of computations as the number of MIMO transmission antennas and modulation order increase (1). This method orthogonalizes the propagation channel matrix for the transmission and reception antennas of smaller block matrices and selects transmission signal point candidates by using Manhattan metric in each block matrix. It reduces the

computation volume by 96.1% from that of the conventional maximum-likelihood decoding method when quadrature phase shift keying (QPSK) modulation is used for 4×4 MIMO multiplex transmission. Computer simulations also showed that the method achieves the same level of transmission char-acteristics as ordinary maximum-likelihood decoding even with high-correlation MIMO channels (channel correlation val-ue: 0.8). We also conducted 42-GHz transmission experiments based on software defined decoding and demonstrated the ef-fectiveness of our method.

■ Single Carrier-Frequency Domain Equalization (SC-FDE)We studied technologies for the SC-FDE method in order to

identify the transmission characteristics of the SC-FDE method when using the amplifier in the non-linear range and compared the level of improvement relative to the OFDM method. We op-

Figure 1. 2×(16,4) STTC-MIMO transmission equipment

Digital radio microphone (Two-piece model)

Digital radio microphone (Hand-held model)

Digital radio microphone receiver

Figure 2. OFDM digital radio microphone prototype


5　Advanced content production technology　｜　5.1 Contribution transmission technology

timized the noise parameter of minimum mean square error (MMSE) equalization in the frequency domain and implement-ed technologies including 16 amplitude and phase shift key-ing (APSK) modulation that is unaffected by the non-linearity of amplifier, error correction, and interleaving. The results of laboratory experiments showed the carrier to noise ratio (CNR) versus bit error rate (BER) characteristics were similar to those of computer simulations. The results will be the basis for imple-menting the SC-FDE method in hardware. We also found that the implementation complexity of the SC-FDE method is less than half that of the OFDM method for the modulator and twice that for the demodulator and verified the transmission charac-teristics of the SC-FDE method when using the amplifier in the non-linear range.

■ Improving the performance of millimeter-wave mobile cameraFor more stable Hi-Vision transmissions, we developed an

elliptical horn antenna which improved the horizontal (in the 0 degree direction) gain by reducing the antenna’s vertical half-value angle from 50° to 20° (Figure 1). As a result, the gain of this antenna in the 0 degree direction was improved by 3.4 dB compared with that of conventional conical horn antennas (2). We also developed a function to avoid the degradation in trans-mission performance caused by delayed reception of interme-diate frequency (IF) signals beyond the guard interval due to differences in camera cable length and multiplexed tally signals with camera control signals to support various cameras. We developed a practical demodulator with a link-quality monitor-ing function compliant with 2×4 MIMO transmission and began a study on compact and low-cost millimeter-wave transmis-sion modules. This millimeter-wave mobile camera was used for shooting programs such as the NHK Trophy and All Japan Tournament in gymnastics, the Japan Women’s Open and the Japan Open Golf Championship, and Music Japan, the Kouhaku year end music show, and charity concerts (Figure 2).

[References](1) S. Suzuki, N. Kogo, H. Hamazumi, K. Fukawa, H. Suzuki: “Complex-

ity-reduced MLD with Block QR Decomposition for Super Hi-Vision Television Wireless Transmission System,” ITE Journal, Vol. 67, No. 12, pp. J488-J496 (2013) (in Japanese)

(2) F. Ito, S. Suzuki, N. Kogo, H. Hamazumi: “Development and Propa-gation Experiments of Elliptical Horn Antenna for a Wireless Cam-era in 42-GHz Band,” ITE Technical Report Vol. 38, No. 8, pp. 43-46 (2014) (in Japanese)

5.2 IP technologies for IP based contribution networks

We are studying IP technologies for Internet Protocol (IP) based contribution networks to enable more flexible opera-tions and swift transmission of program contributions through the use of various IP networks.

■ Contribution transmission technology using vari-ous IP networkWireless IP networks, such as wireless Local Area Networks

(LANs) and mobile broadband networks, are prone to changes in link speed due to adaptive modulation and excessive delays due to buffers. To solve this problem affecting live transmis-sions, we studied a technology that can deal with link speed variations and file transmission technology that can mitigate excessive delays.

Stable live transmissions through a variable-speed network requires a transmission rate control function to prevent packet

losses due to excessive speed and a function for recovering lost packets. To deal with the speed changes affecting wire-less LANs, we developed a streaming device that controls the transmission rate by using the link speed information from the wireless LAN device (1).

We are also studying visible-light wireless IP transmission technology to make it possible to conduct underwater live broadcasts. Unlike the case of wireless LANs, the link speed of this system is constant but its communication is frequently interrupted by bubbles and particles suspended in the water. To deal with these problems, we studied the propagation charac-teristics of visible-light communication and designed a method to avoid such interruptions by using retransmission and redun-dant data transmission.

When large files are transferred over a mobile broadband network, packets can remain for a long time in large-capacity buffers of wireless devices. This can increase the transmission

<3D directivity calculation value>

<Antennas>

Conical horn antenna (conventional)

Elliptical horn antenna (proposed)

Aperture:9.4 mm

Long axis:30 mm

Short axis:8 mm

Figure 1. Conical horn antenna and elliptical horn antenna

Figure 2. In operation at the Japan Women's Open Golf Championship


5　Advanced content production technology5.1 Contribution transmission technology　｜　5.2 IP technologies for IP based contribution networks

delay to several seconds and increase the delays of other com-munications as well. We developed a new congestion control method for Transmission Control Protocol (TCP) so that a buf-fer will not be filled up with packets. The results of experiments using an actual mobile broadband network showed that the method can reduce delays while maintaining throughput per-formance (2).

■ Use of IP network for multiple contribution trans-missionsRaw video footage is transferred between broadcast stations

using a digital leased line reserved for each piece of raw video footage. If multiple pieces of such footage could be transmitted simultaneously, the total transmission time will be shortened and immediate transmission of additional raw video footage will be possible.

Structuring video transmission lines as an IP network in-

stead of digital leased line can enable this sort of transmis-sion, but it is necessary to manage the transmission rate for each piece of footage. We developed a system consisting of network traffic visualizer and transmission rate controller. For example, this system makes it possible to assign a high bit rate to high-priority raw video footage so that it can be delivered to the broadcast station as soon as possible. It is also possible to improve the picture quality of video being aired by decreasing the bit rate of the file transmission or monitoring video.

[References](1) S. Oda, M. Kurozumi, Y. Endo: “Development of mobile video trans-

mitter for wireless LAN network,” ITE Winter Annual Conference 2013, 12-8, (2013) (in Japanese)

(2) T. Koyama, K. Aoki, Y. Endo, “Delay-based TCP Considering the Latency by Data Link Layer Delay of Mobile Broadband Network,” 2013Globecom, CQRM-8-02, pp. 1573-1578, 2013

5.3 Enhanced ISDB-T

■ Field experiments on service-area-limited One-Seg service for disaster areas

The FY 2012 revision to Japanese government regulations has allowed municipal governments, universities and corpora-tions to provide broadcasting services to a limited service area by using unoccupied UHF-band frequencies in the area, called “white space,” on the condition of their not affecting reception of existing digital terrestrial television broadcasting. In FY 2012, we developed an Area One-Seg system with such limited cov-erage so that areas affected by disasters could receive urgent information from local governments and other organizations (Figure 1). The system consists of portable content production and information gathering equipment for One-Seg services and transmission equipment.

We conducted field experiments on the system in September of FY 2013 at a transmission station in Yatomi City, Aichi Pre-fecture (Figure 2) (1)(2). We recorded the reception status of mock area-limited broadcasting in case of disaster on multiple One-Seg receivers and confirmed that radio waves were delivered within the designed service area. The results of verification tests also showed that information directing people to evacu-ation centers and other lifelines would be received by One-Seg devices correctly, thereby demonstrating the system’s effective-ness at providing emergency information.

[References](1) H. Sanei, M. Nakamura, Y. Narikiyo, M. Takada: “Development of a

useful information collection and transmission system for disaster area via service-area-limited One-Seg broadcasting,” IEICE Gen-eral Conference, B-15-15, pp. 644 (Mar. 2013) (in Japanese)

(2) H. Sanei, H. Miyasaka, Y. Narikiyo, M. Nakamura, N. Nakamura, M.

Takada: “Field trial on service-area-limited broadcasting services for disaster relief,” ITE Winter Annual Convention, 12-6 (Dec. 2013) (in Japanese)

Reception display

Video

Data broadcasting

Transmission antenna

Figure 2. Verification experiments

CameraOne-Seg transmission

equipmentInformation gathering

system

One-Seg transmitterDisaster related information input terminal

H.264 encoderDirect input

<Antennas>

Info

rmat

ion

gath

erin

g

Mul

tiple

xing

Mod

ulat

ion

and

outp

ut

Dat

a br

oadc

astin

g co

nten

t gen

erat

ion

Figure 1. Service-area-limited One-Seg system for disasters


5　Advanced content production technology　｜　5.2 IP technologies for IP based contribution networks　｜　5.3 Enhanced ISDB-T

6 Devices for next-generation broadcasting

6.1 Advanced image sensors

6.1.1 Super-high-sensitivity image sensors

■ Low-voltage multiplier film technology

The sensitivity of solid-state image sensors decreases as the number of pixels and frame rate increase. To overcome this problem, we are developing photoconductive films able to multiply electric charges by applying a low voltage. In FY 2013, we improved the sensitivity to visible light of avalanche-type multiplication films using chalcopyrite semiconductors and re-duced the dark current of injection-type multiplication films us-ing crystalline selenium.

The avalanche multiplication film we prototyped in FY 2012 had a problem in that the carrier concentration of gallium oxide is low at the p-n junction formed by the gallium oxide (n-type material) and chalcopyrite material (p-type material) and a car-rier depletion layer forms only on that side. This decreases the film’s sensitivity to visible light. On the basis of experiments conducted in FY 2012, we increased the carrier concentration with a new film formation method that enables the gallium ox-ide to be doped with tin. This increased the film’s sensitivity to visible light (1) (2).

We are researching the next generation of im-age capture, recording, and display devices for new broadcast services such as 8K Super Hi-Vi-sion (SHV).

In our research related to image capture devic-es, we made progress in developing sensors with ultra-high sensitivity, organic image sensors, and 3D-structured imaging devices. Regarding our work on high-sensitivity devices, we inves-tigated a photoconductive film able to multiply electric charges by applying a low voltage in or-der to increase the sensitivity of solid-state im-age sensors. We improved our field-emitter-array image sensor with HARP film that is used in Hi-Vision cameras for reporting at nighttime and in other low-light situations. In our work on high-resolution organic image sensors, we continued developing technology to stack organic photo-conductive films and transparent thin-film tran-sistor circuits for reading electric charges from the photoconductive films on glass substrates. In our work on 3D-structured imaging devices, we made progress on 3D-structured image sensor technologies capable of pixel-parallel signal pro-cessing, for achieving both ultra-high resolution and a high frame rate. We also verified operating principle of these technologies.

In our research on recording devices, we con-tinued with our development of magnetic and ho-lographic recording devices. In particular, regard-ing our work on magnetic recording devices, we investigated the use of magnetic nano-domains that move along magnetic nano-wires to increase the speed of current-driven magnetic domain motion and developed reproduction technology

using a tunneling magnetoresistive magnetic-field sensor. In our work on holographic record-ing devices, we investigated a multiplex record-ing technology to increase the recording density and technology to speed up reproduction. We developed a two-dimensional angle multiplexing method that doubled the multiplexing number to 600. For the high-speed reproduction technol-ogy, we improved the performance of the record-ing media and improved the signal processing to speed up the data reproduction rate to 500 Mbps.

In our work on sheet-type displays for the SHV system, we fabricated an eight-inch VGA display that incorporates a power-efficient, long-lifetime red phosphorescent OLED that we had previously developed. We confirmed that the device did not deteriorate for 250 days after being sealed in a plastic substrate. We also clarified the light-emit-ting mechanism of OLEDs and fabricated an effi-cient, long-lifetime green phosphorescent OLED. We improved the performance of the TFTs used as driving elements for sheet-type displays by in-creasing the mobility of their charge carriers and reducing the size of the oxide semiconductors and organic TFTs.

We also worked on ultra-high-resolution, high-speed spatial light modulator (SLMs) as spatial imaging technology for 3D television with a wide viewing zone. To expand the viewing zone and increase the pixel count, we improved the perfor-mance of an SLM driven by spin-transfer switch-ing using tunnel magneto-resistance, designed a MOS-transistor-array circuit for the device, and developed an active-matrix driving SLM.

Voltage (V)0 5 10 15 2010-11

10-10

10 -9

10 -8

10 -7

10 -6

Dar

k cu

rrent

den

sity

(A/c

m2 )

Transparent electrode

p-type layern-type layer

Crystalline selenium (1μm)

Zinc oxide(50nm)

Glass electrode

Crystalline selenium (1μm)

With zinc oxide

Without zinc oxide

Figure 1. Structure of prototype sandwich cell and dark current comparison


Regarding the injection multiplication film, our sandwich cell with crystalline selenium between a transparent electrode and a metal electrode (FY 2012 prototype) achieved a quantum ef-ficiency of over 100 at an applied voltage of 10 V by exploiting the electron injection multiplication phenomenon. However, it also had a high dark current. In FY 2013, we found that the dark current can be reduced by forming a p-n junction combining crystalline selenium, which is a p-type material, and zinc ox-ide, which is an n-type material and also a hole blocking mate-rial (Figure 1) (3).

■ Compact super-high sensitivity imaging device for Hi-VisionWe are developing field emitter array image sensors with

HARP film that combine field emitters that emit electrons by simply applying a voltage to them with a sensitive High-gain Avalanche Rushing amorphous Photoconductor (HARP) film (Figure 2). This effort is part of our work on compact, super-high-sensitivity Hi-Vision cameras for reporting at nighttime and in other low-light situations. In FY 2013, we developed an electrostatic focusing field emitter to improve the image quality and designed an active-matrix drive circuit for Hi-Vision image sensors.

The image sensor with the electrostatic-focusing field-emit-ter array we prototyped in FY 2012 had many pixels that could not read the signal charges of the HARP film. This problem was due to the dispersion of the electron beam for each pixel, and it significantly degraded the image quality. In FY 2013, we found that the efficiency of getting electrons (as a beam) out of the cathode is about 30% and that increasing the amount of elec-trons in the beam can improve the image quality. Based on this finding, we developed a new electrostatic focusing field emit-ter that can efficiently produce an electron beam. In particular, we identified a structure that could double the amount of elec-trons over that of the previous method. For the active-matrix drive circuit, we designed a circuit structure with a lower drive voltage for the transistor that would enable the pixel size to

be reduced to one-quarter that (11μm×11μm) of the previous structure. We demonstrated the feasibility of high-speed driv-ing required for Hi-Vision (4).

[References](1) K. Kikuchi, S. Imura, K. Miyakawa, M. Kubota, E. Ohta: “Electrical

and optical properties of Ga2O3/CuGaSe2 heterojunction photocon-ductors,” Thin Solid Films, vol. 550, pp. 635-637 (2014)

(2) K. Kikuchi, S. Imura, K. Miyakawa, H. Ohtake, M. Kubota, E. Ohta: “Improved Electrical Properties of Ga2O3:Sn/CIGS Hetero-Junction Photoconductor,” MRS Proceedings, vol. 1635 (2014)

(3) S. Imura, K. Kikuchi, K. Miyakawa, H. Ootake, M. Kubota: “Photo-electric conversion properties of c-Se based p-n heterojunction pho-todiode,” ITE Winter Annual Convention, 11-4 (2013) (in Japanese)

(4) Y. Honda, M. Nanba, K. Miyakawa, M. Kubota, N. Egami: “Active-matrix drive circuit for image sensor consisting of field emitter array and avalanche photoconductor,” Proceedings of the 20th Interna-tional Display Workshops (IDW ‘13), FED1-2, pp. 806-809 (2013)

6.1.2 Organic image sensors

■ Continuously stacked organic image sensorsWe are developing organic image sensors with an image

quality comparable to that of three-chip color broadcast cam-eras for use in a compact single-chip color camera. These de-vices consist of alternating layers of three different organic photoconductive films (organic films) sensitive to each of the three primary colors of light and transparent thin film transistor (TFT) circuits for reading the signals from the photoconductive film on glass substrate (Figure 1).

Organic films are susceptible to high temperatures, but TFT circuits are formed at high temperatures (300°C or higher). Thus, to make this device, we had to increase the heat resis-tance of the organic films and decrease the temperature for forming the TFT circuits. In FY 2013, we improved the TFT char-acteristics by taking advantage of the increased heat resistance of the organic films we had previously developed and a tech-nology to form TFT at lower temperatures. We also prototyped “continuously stacked” devices.

To improve the TFT characteristics, we optimized the condi-tions of sputtering amorphous In-Ga-Zn-O (IGZO) film, reduced the resistance of transparent wiring for reading signals, and devised a process to reduce defects in IGZO film. This increased the mobility and achieved one-digit better on-off characteris-tics. We also prototyped a two-tiered continuously stacked device. The low-temperature formation technology is used as follows: a stacked film consisting of aluminum oxide and ti-tanium oxide and an interlayer insulator made of epoxy resin are formed on the first layer of the TFT reading circuit with an organic photoconductive film for red; a second layer consist-

HARP filmPixel

Field emitter array (incorporating drive circuit)

Insulating layer

Focusing electrode

Electron beam

Gate electrode Emitter

Electrostatic focusing field emitter (FY 2013)

Figure 2. Schematic diagram of image sensor and electrostatic focusing field emitter

Interlayer insulator

Organic film

Glass substrateZoom

10μm Transparent TFT circuit

Light

Figure 1. Cross section of continuously stacked device

2nd layer of TFT circuit +organic film for green

1st layer of TFT circuit+organic film for red

2層目TFT回路+緑色用有機膜

Interlayer insulatorapprox. 10 μm

2nd layer of TFT circuit +organic film for green

1st layer of TFT circuit+organic film for red

Figure 2. Prototype continuously stacked organic imaging device and im-aging example


6　Devices for next-generation broadcasting　｜　6.1 Advanced image sensors

ing of a TFT reading circuit with an organic fi lm for green is then formed. We verifi ed that the device could produce images (Figure 2) (1) and showed the feasibility of a high-resolution or-ganic image sensor. The organic photoconductive fi lms were developed in cooperation with Saitama University, and the TFT circuits were developed in cooperation with the Kochi Univer-sity of Technology.

[References](1) T. Sakai, H. Seo, S. Aihara, H. Ohtake, M. Kubota, M. Furuta: “Color

Image Sensor Using Stacked Organic Photoconductive Films with Transparent Readout Circuits Separated by Thin Interlayer Insula-tor,” To be published in Proc. SPIE (2014)

6.1.3 Core technology for 3D-structured imaging devices

We are researching imaging devices with a 3D structure to implement pixel-parallel signal processing as a way of improv-ing the resolution and increasing the frame rate. These devices have signal processing circuits for each pixel directly beneath the photoelectric conversion element. This enables the signals from all pixels to be read in parallel so that a high frame rate can be maintained even if the pixel count increases (Figure 1).

In FY 2013, we designed signal processing circuits, developed a technology to stack a photoelectric conversion elements and signal processing circuits, and verifi ed the operation principle of devices incorporating this technology. The signal process-ing circuits we designed have a wide dynamic range, and they convert the incident light intensity into electrical pulses (Figure 2). We also devised a simulation method considering the 3D wiring path (1) and prototyped a device with a signal process-ing circuit beneath the photoelectric conversion element by combining circuit wafer thinning technology we developed in FY2013 with technology to directly bond substrates with metal electrodes and insulators on them that we had developed in FY 2012. We confi rmed that the device outputs a number of pulse signals in proportion to the incident light intensity and demon-strated the basic operating principle of 3D-structured imaging devices capable of converting the light into digital signals in the pixels and reading them out in the depth direction of the device. This research was conducted in cooperation with the University of Tokyo.

[References](1) M. Goto, K. Hagiwara, Y. Iguchi, H. Ohtake, T. Saraya, E. Higurashi,

H. Toshiyoshi, T. Hiramoto: “Design of an In-pixel A/D Converter for 3D-structured Image Sensors,” The 30th Sensor Symposium on Sensors, Micromachines and Applied Systems, 6PM3-PSS-44 (2013) (in Japanese)

6.2 Advanced storage technology

6.2.1 Magnetic recording technology

■ Magnetic recording devices utilizing magnetic nano-domains

With the goal of realizing a high-speed magnetic recording device with no moving parts, we are developing recording de-vices that utilize the motion of nano-sized magnetic domains on magnetic nanowires. In addition to technology from FY 2012 that uses pulse currents to control the magnetic nano-domains, we worked on technology to detect the magnetization direc-tions (facing up or down) of the driven magnetic nano-domains by using a magnetic fi eld sensor attached to the nanowires. In other words, it is a reproduction technology to read out binary data corresponding to magnetic nano-domains from magnetic nanowires.

We re-examined the pulse current application system used to drive the magnetic nano-domains and succeeded in driving the nano-domains with pulses having temporal widths of only 50 ns, ten times faster than the 500 ns of our experiments in

FY 2012. Regarding the magnetic nano-domain reproduction technol-

ogy, we prototyped a magnetic domain scope that uses a tun-neling magnetoresistive (TMR) magnetic fi eld sensor to cover a wide area with nano-scale resolution (nano-MDS). We also developed a way to directly detect the density and spatial distri-bution of the magnetic fl ux generated from the magnetic nano-domains. The scope can quantitatively evaluate the magnetic domains, unlike magnetic force microscopy (Figure 1). We used the data we obtained to develop a principle for designing the materials and structures of magnetic nanowires.

We also modifi ed the electrode structure and method of pro-ducing magnetic nanowire elements to verify the motion and reproduction of current-driven magnetic domains simultane-ously on a single sample. We developed a process to make ultra-fl at nanowires with fewer trap sites in their magnetic do-mains by combining electron beam lithography using a nega-tive-type electron-beam resist and ion-beam milling.

Pixel

Pixel-parallel signal processing

Incident light Photoelectric conversion element

Signal processing circuit

Figure 1. 3D-structured imaging devices

Photoelectric conversion element (upper layer)

Signal processing circuit (lower layer)

Incident lightOutput

Pulse

Photoelectric conversion element (upper layer)

Figure 2. Circuit structure


6　Devices for next-generation broadcasting　｜　6.1 Advanced image sensors　｜　6.2 Advanced storage technology

The combination of these technologies of high-speed driv-ing of magnetic nano-domains, the magnetic domain scope covering a wide area with nano-order resolution (nano-MDS), and ultra-flat magnetic nanowires has enabled multiple mag-netic nano-domains to be driven on magnetic nanowires with a pulse current width of 50 ns and at the same time to detect the motion of these nano-domains in real time as continuous changes in the magnetization direction by using a magnetic field sensor directly attached to the magnetic nanowires. We demonstrated this method can be used as a magnetic-record-ing reproduction technology.

[References](1) M. Okuda, Y. Miyamoto, E. Miyashita, N. Hayashi: “Evaluation of

Magnetic Flux Distribution from Magnetic Domains in [Co/Pd] Nanowires by Magnetic Domain Scope Method using Contact-scan-ning of Tunneling Magnetoresistive Sensor,” J. Appl. Phys., Vol. 115, No. 17, pp. 17D113.1-17D113.3 (2014)

(2) Y. Miyamoto, M. Okuda, E. Miyashita: “Ultra-fast Recording Device Utilizing Current-driven Domain Wall Motion in Magnetic Nanow-ires,” ITE Journal, Vol. 68, No. 1, pp. J34-J40 (2014) (in Japanese)

6.2.2 Holographic memory

■ High-speed and high-density holographic memoryArchives for 8K Super Hi-Vision will need to be very large

capacity and have high transfer rates. We are researching holo-graphic memory to meet these needs, because it can reproduce and record a whole page of two-dimensional data at once and is capable of high-density storage with multiplexed recording. In FY 2013, we worked on multiplex recording technology to in-crease the recording density and further accelerated the high-speed reproduction technology.

Our previous multiplex recording technology overwrites multiple data on the same part of a recording device by varying the incident angle of light in one dimension. With this method, the multiplexing number is limited by the available range of the incident angle. The maximum multiplexing number that we previously achieved with this method was 300. To increase that number, we expanded the one-dimensional angle multiplexing to a two dimensions, which records data by changing the angle of recording media in the in-plane direction as well as the inci-dent angle. We confirmed that the average bit error rate could be limited to a practical level of 10-4 even when the multiplexing number was 600.

Regarding the high-speed reproduction technology, we im-plemented the reproduction signal processing algorithms we had developed for graphics processing units (GPUs) in a field programmable gate array (FPGA) and confirmed a throughput of 500 Mbps. We refined the algorithms to verify the required number of bits and reduce the number of multipliers by using GPUs before applying them to the FPGA. The resulting FPGA

could reproduce ultra-high resolution video with more pixels than Hi-Vision in real-time(1) (Figure 1).

[References](1) T. Muroi, N. Kinoshita, N. Ishii, K. Kamijo, H. Kikuchi: “Holographic

Data Storage with Wavefront Compensation and Parallel Signal Processing for Readout of Beyond HD Video Signal,” International Workshop on Holography and Related Technologies (IWH) 2013 Technical Digest, 16c-5 (2013)

6.3 Next-generation display technologies

6.3.1 Flexible displays

We are researching large, lightweight, and flexible sheet-type displays that can be rolled up and used in the home for showing 8K Super Hi-Vision. In FY 2013, we developed fabrication tech-nologies, thin-film transistors (TFTs) for driving active matrix displays, and organic materials and devices.

■ Display panel fabrication technologies

We fabricated an active-matrix driving organic light-emit-ting diode (OLED) display panel using plastic film substrates. It is necessary to reduce the panel’s power consumption, in addition to developing processing technology suited for large screens. In FY 2013, we fabricated an eight-inch VGA (640×480 pixels) panel at a temperature below 200℃ by using red phos-phorescent OLEDs featuring low power consumption and long

Figure 1. Real-time reproduction of ultra-high resolution video

Up-facing magnetic domains

Down-facing magnetic domains

Up-facing magnetic domains(+30mT)

Down-facing magnetic domains(−30mT) Magnetic

flux density＋30 (mT)

0

−302μm 2μm

(a) Result of conventional evaluation using magnetic force microscopy shows the directions of magnetic domains only.

(b) Result of evaluation using magnetic domain scope with nano-MDS shows the magnetic flux density as well as the directions of magnetic domains.

Figure 1. Comparison of evaluation methods of magnetic nano-domain structure


6　Devices for next-generation broadcasting　｜　6.2 Advanced storage technology　｜　6.3 Next-generation display technologies

lifetime. The red OLEDs use a platinum complex as an emitting mate-

rial and a beryllium complex as a host material to transfer en-ergy to the emitting material. We found that these OLEDs had excellent light-emitting characteristics, wherein almost 100% of the injected charge was emitted as photons (1).

The fabricated eight-inch panel was very thin (about 0.3 mm thick), very light (about 15 g), and bendable. The oxide TFT ar-ray, which is the driving device, had a mobility of approximately 8cm2/Vs, and the panel maintained the same level of driving performance over its whole area. The red, green and blue pixels emitted light equally with a peak luminance of around 100cd/m2, and the panel could display moving images with almost uniform quality over the entire screen.

We developed a panel fabrication technology based on print-ing technologies to make large flexible displays and increase productivity. In FY 2013, we formed patterns of OLED materials by using ink-jet printing, developed a coatable gate insulator, and formed OLED electrodes by screen printing. This effort re-sulted in a prototype panel built using printing technologies for not only the OLEDs but also the wiring and electrodes.

■ TFT materials and devicesWe are researching oxide semiconductor TFTs (IGZO-TFT)

with high field-effect mobility that can be used for large screens and organic TFTs with a flexible structure that can be formed at low temperatures as driving elements of large-sized sheet-type OLED displays. We developed a method to produce self-aligned IGZO-TFTs that have less parasitic capacitance and a shorter channel length. We fabricated a self-aligned IGZO-TFTs with channel lengths of 9 μm and lowered the IGZO’s resistance by irradiating it with excimer laser light. We showed that this technology has a sufficient process-margin for the irradiation intensity of the excimer laser (2). As it can suppress the influ-ences of uneven laser intensity and the intensity dispersion of each irradiation, it can uniformly form self-aligned IGZO-TFTs on a large area. We also developed a new high-speed method of evaluating the trap density, to analyze the TFT characteris-tics. A combination of capacity measurements and high-speed simulations enabled rapid measurement of the trap density in one tenth the time required for conventional methods using a two-dimensional simulator. This advance will improve the pro-duction process.

For organic TFTs, we are studying a fine patterning method for solution-processable semiconductors using selective modi-fication of the solvent wettability on the gate insulator. We de-veloped a technology to improve the uniformity of crystalline film by controlling the direction of solvent evaporation. We also fabricated and evaluated a TFT array with transparent paper substrates as a technology for future displays with a high level of flexibility. We experimentally showed that its electrical char-acteristics are equivalent to those of conventional plastic sub-

strates and that it can operate even when bent with a curvature radius of 2 mm.

■ Display materials and devicesWe are researching new device structures, light emitting ma-

terials, and charge transport materials to reduce power con-sumption and extend the operating/storage lifetime of flexible OLEDs. The greatest challenge in applying OLEDs to flexible displays is that OLEDs are sensitive to oxygen and moisture and deteriorate on a flexible plastic substrate. In FY 2013, we devel-oped an inverted OLED (iOLED) whose electrode structure is opposite to that of conventional OLEDs, and we confirmed that it did not deteriorate after being sealed in plastic substrate for 250 days (Figure 2). The inverted structure increased the selec-tivity of materials for the cathode and electron injection layers, which are the main factors of deterioration. The development of a new electron injection layer especially helped to make the OLED resistant to oxygen and moisture. We fabricated a five-inch monochromatic display using the iOLED (3). The iOLED was developed in cooperation with Nippon Shokubai Co., Ltd.

We elucidated the light-emitting mechanism in phosphores-cent OLEDs. It had been considered difficult to clarify the light-emitting mechanism of OLED devices because it has many processes. By simplifying the processes related to light emis-sion, we identified the excited state behaviors that contribute to the efficiency and lifetime of these devices. We also investi-gated how the device performance depends on charge trans-port materials and elucidated the molecular structure of charge transport materials suitable for efficient and stable green phos-phorescent OLEDs. Part of this research was conducted as a government-commissioned project from the Ministry of Inter-nal Affairs and Communications, titled “R&D on highly efficient and stable organic light-emitting device toward the realization of ultimate power-saving display.”

■ Thermoacoustic device

A flexible sound generator based on the thermoacoustic ef-fect was developed for use with flexible devices. The device was composed of a very thin electrode and several flexible films. It generates sound with no mechanical moving parts. We measured various acoustic characteristics of this device and found that it had a wide frequency band, from the audible range up to 100 kHz.

[References](1) G. Motomura, Y. Nakajima, T. Takei, Y. Fujisaki, H. Fukagawa, H.

Tsuji, M. Nakata, T. Shimizu, T. Yamamoto: “Oxide-TFT-Driven Flex-ible Display Using Highly Efficient Phosphorescent OLED,” Proceed-ings of 33rd International Display Research Conference, 10.4 (2013)

(2) M. Nakata, H. Tsuji, Y. Fujisaki, H. Sato, Y. Nakajima, T. Takei, T. Yamamoto, T. Kurita: “Fabrication method for self-aligned bot-tom-gate oxide thin-film transistors by utilizing backside excimer-laser irradiation through substrate,” Appl. Phys. Lett., Vol. 103, pp. 142111.1-142111.4 (2013)

OLED structure

OLED

Preservation period

1 day 15 days 51 days 103 days

1 day 8 days 58 days 104 days 250 daysiOLED

Figure 2. extending the lifetime of inverted OLEDs (iOLEDs)

Figure 1. Fabricated eight-inch VGA flexible OLED display


6　Devices for next-generation broadcasting　｜　6.3 Next-generation display technologies

(3) H. Fukagawa, K. Morii, Y. Arimoto, M. Nakata, Y. Nakajima, T. Shi-mizu, T. Yamamoto: “Highly Efficient Inverted OLED with Air Stable Electron Injection Layer,” Technical Digest Paper of SID Sympo-sium, P-140L, pp. 1466-1469 (2013)

6.3.2 Advanced display devices

■ Spatial light modulator driven by spin-transfer switching

We are researching electronic holography to realize a spa-tial imaging three-dimensional television that can provide fully natural 3D images. Displaying the 3D images with a wide view-ing zone requires a spatial light modulator (SLM) having a very small pixel pitch, a large number of pixels, and high driving speed. We are developing a spin-transfer SLM (spin-SLM) with a pixel pitch under 1 μm (1). This spin-SLM uses the magneto-optical Kerr effect, in which the polarization plane of reflected light rotates according to the magnetization direction of the magnetic materials in the pixel. The magnetization direction is controlled by the direction of the current in the pixel, and this is called spin-transfer switching. We previously developed a one-dimensional SLM with giant magneto resistance (GMR) structure, and confirmed its successful operation with passive-matrix driving by directional voltage application.

In FY 2013, we investigated tunnel magneto-resistance (TMR) for large-scale spin-SLMs. We fabricated an active-matrix (AM) driving TMR spin-SLM that can accurately turn the pixel on and off by using a transistor built in each pixel. The AM driv-ing TMR spin-SLM employs a structure in which a TMR light modulation element is stacked on the transistor that drives each pixel (Figure 1). The TMR light modulation element con-tains a pinned layer made of multi-layer film of terbium-iron-cobalt (Tb-Fe-Co) and cobalt-iron (Co-Fe) alloys, an insulating layer of magnesium oxide (MgO) film, and a light modulation layer of gadolinium-iron (Gd-Fe) alloy film with gadolinium and Co-Fe buffer layers. The magnetic tunnel junction formed with ultrathin (1-nm thick) MgO film between the Co-Fe films en-ables low-current operation. The gadolinium buffer layer inside the light modulation layer increased the Gd-Fe film’s tendency to magnetize in the vertical direction (a phenomenon called perpendicular magnetic anisotropy) and improved the light modulation performance. For the AM driving device, we built a circuit board with a two-dimensional array (10×5 pixels, 5-μm pixel pitch) of MOS transistors formed on a single crystal sili-con substrate. To make a two-dimensional array of submicron-sized TMR elements, we developed a nano-fabrication process-ing technology with a position adjustment accuracy of 20 nm or less, prototyped a two-dimensional spin-SLM, and verified that the magnetization of the Gd-Fe light modulation layer could be switched by controlling the external magnetic field.

We fabricated a pattern with a GMR structure (GMR holo-gram with a 1 μm pixel pitch and 3840×2160 pixels) and verified that the 3D image could be reproduced with a viewing-zone angle of 37 degrees. We also confirmed that the reproduced im-age could be turned on and off by applying an external mag-netic field to the GMR hologram (2). This research was supported by the National Institute of Information and Communications Technology (NICT) as part of the project titled “R&D on Ultra-Realistic Communication Technology with Innovative 3D Video

technology.”

■ Light beam directionality control deviceWe investigated optical devices for future integral 3D displays

without a micro-lens array. We fabricated a device consisting sub-micron-sized dielectric structures that has capability of deflecting a light beam in a designated direction. In FY 2013, we devised a device structure to reduce sidelobe (unintended light beam emission) and stray light (light leaking on unintend-ed paths) and verified its operation. Simulations showed that sidelobe can be reduced to 20% by increasing the number of structures from three to six (3). We designed a device with sub-micron sized structures, and we fabricated a device with light-shielding metal film to prevent stray light and confirmed that it had reduced sidelobe and stray light.

[References](1) K. Aoshima, N. Funabashi, K. Machida, Y. Miyamoto, K. Kuga, T.

Ishibashi, N. Shimidzu, F. Sato: “Submicron Magneto-Optical Spatial Light Modulation Device for Holographic Displays Driven by Spin-Polarized Electrons,”J. Display Tech., Vol . 6, No. 9, pp. 374-380 (2010)

(2) K. Machida, D. Kato, T. Mishina, H. Kinjo, K. Aoshima, A. Kuga, H. Kikuchi, N. Shimidzu, “Three-dimensional image reconstruction with a wide viewing-zone-angle using a GMR- based hologram,” OSA Topical Meeting Digital Holography and 3-D Imaging, DTh2A.5 (2013)

(3) Y. Hirano, K. Tanaka, Y. Motoyama, N. Saito, H. Kikuchi, N. Shimid-zu, “Directionality Control of Light-emitting Devices Through Sub-micron Dielectric Structures,” 2nd Int. Conf. on PHOTOPICS 2014, pp. 163-169 (2014)

Driving element (Si backplane)

TMR light modulation

element

Transparent electrode

Insulating layer 1

Gate

MOS transistorSi substrate

Source Drain

Electrode

Light modulation layerInsulating layer

Pinned layer

Insulating layer 2

Figure 1. Schematic illustration of AM driving TMR spin-SLM structure


6　Devices for next-generation broadcasting　｜　6.3 Next-generation display technologies

7.1 Joint activities with other organizations

7.1.1 Participation in standardization organizations

NHK STRL is participating in standardization activities, within and outside of Japan, mainly related to broadcasting. We are contributing to the creation of technical standards that incorporate our research results.

Following its approval of Recommendation BT.2020 for 8K Super Hi-Vision (SHV) video formats, ITU-R SG6, the study group that handles standardization for broadcasting, estab-lished Recommendation BS.2051 in February 2014 for hybrid sound systems combining channel-based and object-based methods, including the 22.2 multichannel sound loudspeaker arrangement,.

At MPEG, the essential part of MPEG Media Transport (MMT) multiplexing technology for broadcast-broadband hy-brid transmissions, including the packet structures and control signals for hybrid distribution, was approved as an interna-tional standard. We proposed the addition of 22.2 multichan-nel sound to MPEG-4 Advanced Audio Coding (AAC), and the standard was revised accordingly. Also in FY 2013, we initiated in standardization activities on coding technologies for inte-gral 3D television.

At SMPTE, we added a 120-Hz frame rate compliant with ITU-R BT.2020 and a wide color gamut to the S2036-1 UHDTV video standard.

The Advanced Television Systems Committee (ATSC), an or-ganization standardizing TV broadcasting systems in the U.S., began its work on next-generation digital terrestrial broad-casting in FY 2013. NHK presented our research results on transmission and multiplexing technologies for SHV terrestrial broadcasting at the ATSC 3.0 conference and participated in their meetings.

At the World Wide Web Consortium (W3C), which oversees the HTML5 standard for describing content delivered through broadcasting and telecommunications, NHK presented the Hy-bridcast specifications that were established at IPTV Forum in March 2013 and reported on the services launched in Septem-ber.

The technical committee and general meetings of the Asia-Pacific Broadcasting Union (ABU) were held in Hanoi, Viet-nam. NHK presented a keynote lecture and submitted nine project-manager and group-chairman reports on topics such as production, transmission, and frequency use. We also gave lectures at a digital broadcasting symposium held in March in Kuala Lumpur.

Besides the above activities, we promoted standardization activities through participation in international and domestic standardization organizations, including the European Broad-casting Union (EBU), International Organization for Stan-dardization (ISO), International Electrotechnical Commission (IEC), Audio Engineering Society (AES), the Japan Electronics and Information Technology Association (JEITA), and the Tele-communication Technology Committee (TTC) of Japan.

NHK STRL actively promotes the use of our research results on 8K Super Hi-Vision and Hy-bridcast in various ways, including through the NHK STRL Open House and other exhibits and reports. It also helps to develop technologies by forging links with other organizations and col-laborating in the production of programs.

The essential part of the MPEG Media Trans-port (MMT) multiplexing technology to transmit data using radio waves and telecommunications networks was approved as an international stan-dard by MPEG, and the 22.2 multichannel sound system was approved as an ITU-R international standard (BS.2051). In addition, we contributed to domestic and international standardization activities at ABU, the Information and Commu-nications Council, Association of Radio Indus-tries and Businesses, and various organizations around the world. We also promoted the ISDB-T digital broadcasting format by giving lectures about it.

The theme of the FY2013 NHK STRL Open House was “Expect, Explore, Experience.” It fea-tured 37 exhibits, centering on Super Hi-Vision technologies, such as the world’s first 8K HEVC real-time encoder, and Hybridcast. The event had two special lectures and three research presen-tations, and it was attended by 20,607 visitors. Besides the Open House, we held 31 exhibitions

in Japan and overseas.We also gave 136 tours of our laboratories to a

total of 1,771 visitors. Of these, 41 tours were for visitors (387) from overseas.

We published 610 articles (including 13 press releases) describing NHK STRL research results at conferences and in journals within and outside Japan. We continued to consolidate our intellec-tual property rights, with 488 patent applications submitted and 320 patents obtained in FY2013. As of the end of FY2013, NHK held 2,017 patents.

We are also cooperating with outside organiza-tions. Last year, we participated in 27 instances of collaborative research and seven instances of commissioned research. We hosted 28 visiting re-searchers (including 24 trainees), and dispatched two of our researchers overseas.

Our research results, such as on multi-view-point robotic cameras, virtual studios with hybrid sensors, and macro-diversity reception systems for marathon relays were used in the produc-tion of NHK television programs. In FY2013, NHK STRL collaborated with the parent organization in making 53 programs. Finally, in recognition of our research achievements, NHK STRL received a total of 36 awards in FY2013, including the Meri-torious Award on Radio and the Takayanagi Me-morial Award.


7 Research-related work

7.1.2　Collaboration with overseas research facilities

Encouraged by the selection of Japan’s ISDB-T format for di-Cooperative relationships with local broadcasters and carriers will be essential to facilitating transmission and production of 8K Super Hi-Vision (SHV) video on site. In preparation for the 2014 FIFA World Cup and Rio 2016 Olympics in Brazil, we cooperated with Rede Nacional de Ensino e Pesquisa (RNP), Brazil’s academic network of research institutions, in a series

of IP transmission tests on sending SHV video from Brazil to Japan. We also held exhibits to promote SHV in cooperation with a local broadcaster, TVGlobo.

We collaborated with many research institutions, including the German Fraunhofer Institutes, ABU and TVGlobo, on the standardization of 22.2 multichannel sound.

In October, the Digital Group of Broadcasting Technology

■ Leadership activities at major standardization organizations■ International Telecommunication Union (ITU)

Committee name Leadership role

International Telecommunication Union, Radio communi-cation Sector (ITU-R)

Study Group 6 (SG 6, Broadcasting services) Vice-chairman

Working Party 6B (WP6B, Broadcast ser-vice assembly and access)

Chairman

■ Asia-Pacific Broadcasting Union (ABU)


Technical committee Vice-chairman

■ Information and Communications Council


Information and communications technology subcommittee

Spectrum management and planning committee Expert member

Radio-wave propagation committee Expert member

Satellite and scientific services committee Expert member

Broadcast services committee Expert member

Terrestrial wireless communications committee Expert member

■ Telecommunication Technology Committee (TTC)


IPTV Working Group Committee chairman

■ Association of Radio Industries and Businesses (ARIB)


Technical committee

Broadcasting international standardization working group

Chairman

Quality assessment methods investigative re-search group

Committee chairman

Monitoring and evaluation methods working group

Manager

Flat-panel display image quality evalu-ation working group

Manager

Sound quality evaluation working group

Manager

New broadcast technology investigative re-search group

Future 3D television formats investiga-tive working group

Manager

Next-generation digital broadcast transmission technology investigative working group

Manager

Digital broadcast systems development section Committee chairman

SBTVD-T standardization task force Leader

Copyright protection working group Manager

Video coding methods working group Manager

Data coding methods working group Manager

Advanced data imaging (H.264) TG Leader

Multiplexing working group Manager

Download methods TG Leader

Data broadcasting methods working groupApplication control ad hoc group LeaderBroadcast service application meta-data working group

Manager

Digital receivers working group Manager

Download TG Leader

Emergency broadcast transmission TG

Leader

AC Security AHG Leader

Ultra-high-definition TV broadcast receivers TG

Leader

Digital satellite broadcasting working group

Manager

Digital terrestrial broadcasting trans-mission path coding working group

Manager

3D television broadcast formats inves-tigative working group

Manager

Contribution transmission development section

Terrestrial wireless contribution trans-mission working group

Manager

Millimeter-wave contribution trans-mission TG

Leader

New Frequency FPU Study TG LeaderUltra-high-definition TV studio facilities develop-ment section

Committee chairman

Acoustic system study working group Manager

Device interface study working group Manager

Standard Assembly

Low power radio station working group

Radio microphone WG

New Digital Transmission Format Study TG

Leader


7　Research-related work　｜　7.1 Joint activities with other organizations

7.1.3　Collaborative research and cooperating institutes

In FY2013, we conducted a total of 27 collaborative research projects and 13 cooperative research projects on topics rang-ing from system development to materials and basic research.

We collaborated with graduate schools in eight universi-ties (Chiba University, University of Electro-Communications,

Tokyo Institute of Technology, Tokyo Denki University, Tokyo University of Science, Toho University, Tohoku University, and Waseda University) on education and research in activities such as sending part-time lecturers and accepting trainees.

7.1.5　Commissioned research

We are participating in research and development projects with national and public facilities in order to conduct our re-search on broadcast technology more efficiently and effective-ly. In FY 2013, we took on seven projects from national and other public organizations including the Ministry of Internal Affairs and Communications, NICT*, and SCOPE**. The topics were as follows:

• R&D on efficient use of frequency resources for next-gener-ation satellite broadcasting systems

• R&D on expansion of radio spectrum resources• R&D on ultra-realistic communication technology with in-

novative 3D video technology

▶Device technology for innovative 3D video displays▶Recognition and transmission of sensitivity information

• R&D on systems to capture spatial information using com-posite image planes

• R&D on high-efficiency, long-life organic EL devices for low-power displays

• R&D on video content evaluation methods using viewer gaze in the living space

* NICT : National Institute of Information and Communications Technology**SCOPE : Strategic Information and Communications R&D Promotion Programme

7.1.6　Committee members, research advisers, guest researchers

We held two meetings of the broadcast technology research committee and received input from academic and experienced external committee members. We held 15 sessions to gain in-

put from research advisers. We also invited researchers from other organizations to work on five topics with us.

7.1.4　Visiting researchers, trainees, STRL staff overseas

We hosted one researcher from Vietnam, as part of our pro-gram to invite young researchers from facilities affiliated with the Asia-Pacific Broadcasting Union (ABU). We also hosted two researchers from other Japanese broadcasters and took on one post-doctoral research project (Table 1).

We provided guidance to a total of 24 trainees from nine uni-versities (University of Electro-Communications, Tokai Uni-

versity, Tokyo Denki University, Tokyo University of Science, Nagaoka University of Technology, Waseda University, Tokyo Institute of Technology, Tokyo Woman’s Christian University, and Keio University) in their work towards their Bachelor’s and Master’s degrees.

Two STRL researchers were dispatched to research institu-tions in the United States (Table 2).

and Research (DGBTR) convened in Beijing, China. Attended by ABS, CCTV (China), and KBS (Korea), we exchanged infor-mation on the latest broadcast technology and research in Ja-pan, China, and Korea.

Brazil adopted Japan’s ISDB-T standard as a basis for their digital terrestrial broadcasting in June 2006. Since then, the

public and private sectors in Japan have worked together in promoting ISDB-T to the world. This effort has so far resulted in 15 countries adopting ISDB-T. We dispatched lecturers to seminars on ISDB-T in the Republic of Botswana, who adopted ISDB-T in February 2013, and the Philippines.


7　Research-related work　｜　7.1 Joint activities with other organizations

Table 1. Accepted researchers details

Type Term Research topic

Visiting researcher 2013/2/1 to 2013/6/30 Service models for smart television

Visiting researcher From 2014/1/14 Advanced Hybridcast services

Visiting ABU researcher 2012/12/1 to 2013/5/31 Cable transmission technology for Super Hi-Vision

Post-doctorate From 2012/5/1 Acceptance characteristics for sensation of depth from high-resolution video

Table 2. NHK STRL staff overseas

Location Term Research topic

Duke University, USA. 2013/8/1 to 2014/1/31 Development of ultra-high-pixel camera system using multiple cameras

University of Michigan, USA. 2013/8/29 to 2014/2/28 Pixel structure suitable for large-sized, multi-pixel sheet-type displays

7.2 Publication of research results

7.2.1　STRL Open House

The STRL Open House’s theme for 2013 was “Expect, Ex-plore, Experience”, and the event included 37 exhibits, three interactive exhibits and 13 poster displays of the laboratories’ latest research results, centering on 8K Super Hi-Vision (SHV) and Hybridcast. The Hybridcast exhibits on services NHK launched in September 2013, service examples of commercial broadcasters and pay TV service providers, and the world’s first 8K HEVC real-time encoder attracted much attention from visitors. A SHV screening of the London Olympics and the

Rio Carnival was presented in the auditorium, and there were events, such as the “Hands-On Construction Experience”, for school children on the weekend. The open house attracted a total of 20,607 visitors.

ScheduleMay 28 (Tuesday) Opening ceremonyMay 29 (Wednesday) Open to inviteesMay 30 - June 2 (Thursday to Sunday) Open to the public


7　Research-related work　｜　7.1 Joint activities with other organizations　｜　7.2 Publication of research results

Name Affiliation

** Kiyoharu Aizawa Professor, University of Tokyo

Toshihiko Kanayama Vice President, National Institute of Advanced Industrial Science and Technology (AIST)

Yasuhiro Koike Professor, Keio University

Tetsunori Kobayashi Professor, Waseda University

Hiromichi Shinohara Executive VP, NTT Group; Director, Research and De-velopment Planning Department

Yoichi Suzuki Professor, Tohoku University

Junichi Takada Professor, Tokyo Institute of Technology

Atsushi Takahara Director, Institute for Materials Chemistry and Engineer-ing, Kyushu University

* Fumihiko Tomita Vice President, National Institute of Information and Communications Technology (NICT)

Toshiei Toyonaka Managing Director, Tokyo Broadcasting System Televi-sion Inc.

Yasuyuki Nakajima President/CEO, KDDI R&D Laboratories

Yasumasa Nakata Executive Director, Technology Research and Develop-ment Dept., Fuji Television Network, Inc.

Masatoshi NozakiSection Manager, Broadcast Technology Division, Infor-mation and Communications Bureau, Ministry of Inter-nal Affairs and Communications

Michihiko Minoh Director-General, Institute for Information Management and Communication, Kyoto University

Masayuki Murata Professor, Osaka University

■ Broadcast Technology Research Committee Members(Titles omitted) March 2014

** Committee chair * Committee vice-chairName Affiliation

Makoto Ando Professor, Tokyo Institute of Technology

Koichi Ito Professor, Chiba University

Susumu Itoh Professor, Tokyo University of Science

Tohru Ifukube Project Researcher (Honorary Professor), Institute of Ger-ontology, University of Tokyo

Hideki Imai Professor, Chuo University

Tatsuo Uchida President, Sendai National College of Technology

Juro Ohga Emeritus Professor, Shibaura Institute of Technology

Tomoaki Ohtsuki Professor, Keio University

Wataru Kameyama Professor, Waseda University

Jiro Katto Professor, Waseda University

Satoshi Shioiri Professor, Tohoku University

Takao Someya Professor, University of Tokyo

Fumio Takahata Professor, Waseda University

Takashi Taniguchi Professor, Osaka Gakuin University

Katsumi Tokumaru Emeritus Professor, University of Tsukuba

Yoshihisa Nakamura President, Iwate Prefectural University

Mitsutoshi Hatori Emeritus Professor, University of Tokyo

Takayuki Hamamoto Professor, Tokyo University of Science

Hiroshi Harashima Emeritus Professor, University of Tokyo

Philip Nelson Professor, University of Southampton, U.K.

Takehiko Bando Emeritus Professor, Niigata University

Kuniyoshi Yokoo Emeritus Professor, Tohoku University

Akihiko Yoshikawa Professor, Organization for Academic-industrial Collabo-ration and Intellectual Property, Chiba University

■ Research Advisers (Titles omitted) March 2014

Name AffiliationKenji Ozawa Professor, University of Yamanashi

Yoshimasa Kawata Professor, Shizuoka University

Takefumi Hiraguri Associate Professor, Nippon Institute of Technology

Tokio Nakada Professor, Tokyo University of Science

Kazuhiko Fukawa Professor, Tokyo Institute of Technology

■ Guest Researchers (Titles omitted) March 2014

Hybridcast service exhibits Character greeting

■ Lectures

Title SpeakerA New Broadcasting System: the immersive, pervasive, data-rich services of the future Matthew Postgate, BBC Research & Development, Controller

Toward Next-Generation High-definition Video Broadcasting Susumu Itoh, Tokyo University of Science, Professor

7.2.2　Overseas exhibitions

The world’s largest broadcast equipment exhibition, the National Association of Broadcasters Show 2013 (NAB Show 2013), was held in April. We exhibited the latest 8K Super Hi-Vision (SHV) technologies, such as a 300-inch theater screen, cameras, and terrestrial transmission equipment, as well as our research results on Hybridcast. We demonstrated to at-tendees that broadcast technologies are advancing rapidly. The show attracted about 92,000 visitors from around the world.

International Broadcasting Convention 2013 (IBC 2013), the

largest broadcast equipment exhibition in Europe, was held in September. We exhibited an 8K HEVC real-time encoder for the first time at an overseas exhibition, along with SHV and Hy-bridcast. IBC was attended by about 53,000 visitors.

We also exhibited SHV and Hybridcast technologies at Asia-pacific Broadcasting Union 2013 (ABU 2013) conference held in October to discuss broadcast-related issues in the Asia-pacific region. Our exhibitions gained a great deal of interest from broadcast industry of Asian countries.


7　Research-related work　｜　7.2 Publication of research results

Title Speaker

Media Transport Technologies for Next Generation Broadcasting Systems Shuichi Aoki, Advanced Broadcasting Platforms Research Division

Super Hi-Vision HEVC Real-time Encoding Equipment Yasuko Sugito, Advanced Television Systems Research Division

A Flexible OLED Display for Large-screen Ultra-thin Televisions Yoshiki Nakajima, Display & Functional Devices Research Division

■ Research Presentations

1 Hybridcast 14 Twitter Analysis Technology for Opinion Mining 27 SNS Style CG Content Creation System with Content Re-creation

2 Advanced Hybridcast 15 Sign Animation Synthesis System for Japanese Weather News 28 Multi Device Link Service with Mobile Camera

3 Super Hi-Vision Production Devices for Mobile 16 Tactile and Haptic Presentation Technology of 2&3-dimensional Information 29 Wide-Band Transmission Technologies for

12-/21-GHz-band Satellite Broadcasting

4 120-fps Full-specification Super Hi-Vision Image Sensor 17 120-GHz-band FPU for Uncompressed SHV Sig-

nal Transmission 30 Video Asset Processing and Management Sys-tem “Video Bank”

5 Compact Video Recorder for Super Hi-Vision Camera 18 Media Transport Technologies for Next Genera-

tion Broadcasting Systems 31 NHK Broadcast Museum

6 Super Hi-Vision HEVC Real-time Encoder 19 Adjustment System of Sound Levels in TV Pro-grams for Elderly Listeners 32 Utilization and Development of NHK’s Technol-

ogy

7 12-GHz-band Satellite Broadcasting System for Super Hi-Vision 20 Wide-color-gamut Super Hi-Vision System 33 Technologies in the Broadcasting Station

8 Large-capacity Transmission Technologies for Next Generation Terrestrial Broadcasting 21 Multi-Viewpoint Robotic Cameras 34 Flexible Organic Light-emitting Diode Display

9 Long Haul Optical Transmission for Super Hi-Vision 22 Low Delay and High Quality Digital Wireless Mi-

crophone System 35 High-speed Data Readout in Holographic Mem-ory

10 145-inch Super Hi-Vision Display Integrated with Loudspeaker Array 23 Mobile Video Transmission Technologies on

Wireless LAN IP 36 Elemental Technologies for Advanced Imaging Devices

11 Integral Three-dimensional Television 24Transmission Technologies of Next Generation Digital Terrestrial Broadcasting for Mobile Re-ception

37 Data Broadcasting Service to Select Sub-chan-nel Using Color Buttons on Remote Controller

12 Real-time Video Coding with Super-resolution Techniques 25 P2P Live Streaming Technology J Consultation Booth on Digital Broadcasting Re-

ception

13 Hybridcast Application Production and Distribu-tion System 26 Integrated Downloadable CAS for Broadcasting

and Communications

■ Research exhibits

1 A Parallel Distributed Processing System for Broadcasting Contents 6 Carrier Multiplication at Low Voltage in Photo-

conductor 11 Display Driving Technology Using Optical Wave-length Multiplexing

2 Spherical Microphone for Super Hi-Vision 22.2 Multichannel Sound 7 Low Current Driving of Multiple Magnetic Do-

mains in Magnetic Nanowires 12 Technologies for Fabrication of Self-aligned Ox-ide Semiconductor Thin-film Transistors

3 Accommodation Responses in Viewing Integral 3DTV 8 High-density Recording Technologies on Thin

Optical Disk 13 Far-field Pattern Control of Light-emitting De-vices through Sub-micron Dielectric Structures

4 Improvement on Speech Recognition for Live Captioning of Emergency Disaster Broadcasts 9 Spin Spatial Light Modulator Using Tunnel Mag-

netoresistive Effect

5 TV Program Recommendation Based on a Per-sonal Preference Model 10 Flexible Sound Generator Based on Thermoa-

coustic Effect

■ Poster exhibits

1 Experiencing News Service in “Easy Japanese” 2 Hide and Seek Over a Lattice 3 Easier Visual Recognition of Moving Objects

■ Interactive exhibits

■ Three overseas exhibitions

Event name Dates Speaker

NAB Show 2013 (Las Vegas, USA) 4/8 to 4/11 Super Hi-Vision, Hybridcast

IBC2013 (Amsterdam, Netherlands) 9/13 to 9/17 Super Hi-Vision, Hybridcast

ABU2013 (Hanoi, Vietnam) 10/23 to 10/29 Super Hi-Vision, Hybridcast

7.2.3　Exhibitions in Japan

Throughout the year, NHK broadcasting stations all over Ja-pan hosted exhibits on the latest broadcast technology result-ing from our R&D. We gave a new presentation of a Hybridcast

application that provides program-related content in synchro-nization with the progress of the program and presented it at many events.

7.2.4　Academic Conferences, etc.

We presented research results at many conferences in Japan, such as the ITE and IEICE conferences, and had papers pub-lished in international publications such as IEEE Transactions, Applied Physics Letters, and Advanced Functional Materials.

7.2.5　Press releases

We issued 13 press releases on our research results and other issues.

7.2.6　Visits, tours, and event news coverage

To promote R&D on Hybridcast and Super Hi-Vision, we held tours for people in broadcasting-related fields. We welcomed visitors from around the world, including officials of standard-ization and international broadcasting conference organiza-tions, such as ITU and IBC, and from broadcasters in various

countries.


7　Research-related work　｜　7.2 Publication of research results

■ 28 exhibitions in Japan

Event name (Major events) Dates Exhibits

Thanks in Shibuya (Shibuya de domo) 5/3 to 5/5 "TV no shikumi", Hybridcast

Cable Tech Show 7/30 to 7/31 Cable TV transmission technology for Super Hi-Vision

NHK Science Stadium 9/28 to 9/29 Super HARP camera, Ultra-high-speed camera

CEATEC JAPAN 2013 10/1 to 10/5 Super Hi-Vision, Hybridcast

EXPO Hiroshima, IT's a solution 10/23 to 10/25 Hybridcast

NHK Osaka Station Open House 11/8 to 11/10 Super Hi-Vision, Hybridcast

NHK Matsuyama Station Open House 11/22 to 11/24 Super Hi-Vision, Hybridcast

NHK Olympic Park, Sochi Winter Olympics memorial event (in Shibuya) 2/8 to 2/23 "Gurutto Vision", Ultra-high-speed camera

NHK Nagoya Station 60th anniversary event 2/28 to 3/2 Super Hi-Vision, Hybridcast

Academic journals in Japan 58 papers

Overseas Journals 34 papers

Academic and research conferences in Japan 265 papers

Overseas / international conferences, etc. 130 papers

Contributions to general periodicals 51 articles

Lectures at other organizations 72 events

Total 610 instances

Dates Press release content

2013/5/9Development of a multi-viewpoint robotic camera system- Dynamic video expression by serial control of multiple cameras -

5/9Development of a device to reproduce the “sense of tracing the object outline”- Conveying information with tactile displays -

5/9NHK STRL Open House 2013, Expect, Explore, Experience from 5/30

5/9World’s first 8K Super Hi-Vision- Development of an HEVC encoder -

5/16Development of an oxygen- and moisture-resistant OLED device- For thin and light sheet-type displays -

5/16

8K Super Hi-Vision: Terrestrial transmission tests using a single frequency network is successful- Spatio-temporal coding technology to enable more stable large-capacity transmission -

5/23Development of a transmission scheme for low latency digital radio microphones- For smooth frequency migration to the 800-MHz band -

Dates Press release content

5/23Handy camera for real-time CG synthesis- Development of a compact sensor to measure camera’s positions and postures -

5/27Development of compact 8K Super Hi-Vision cube-style camera head

6/18NHK STRL received the “Walter Kosonocky Award”- Research on 8K Super Hi-Vision image sensors -

9/5Launch of sign-language CG evaluation website- For enhanced sign language services -

2014/1/208K Super Hi-Vision: Long-distance transmission test is successful

2/13NHK STRL Open House 2014 schedule decided- To be held in May, on the theme, “Hear t to Heart Technology” -

Inspections, tours136 instances (41 from overseas)

1771 visitors (387 from overseas)

News media 35 events

7.2.7　Bulletins

NHK STRL publishes bulletins describing its research activi-ties and achievements.

NHK STRL R&D publishes special features on topics such as natural language processing, high frame rate video technol-ogy, and 3D video technology.

The Broadcast Technology journal is oriented to overseas readers. It features in-depth articles about our latest research, and this year covered topics such as “Large-capacity Trans-mission Technologies for Next-generation Terrestrial Broad-casting”, “Trial Service of Easy Japanese News on the Web”, and “Progress on Super Hi-Vision Image Sensors.”

Domestic PublicationsSTRL Dayori (Japanese, monthly) No.97 to No.108NHK STRL R&D (Japanese, bimonthly) No.139 to No.144Annual Report (Japanese, annually) FY2012 Edition

Publications for overseas readersBroadcast Technology (English, quarterly) No.52 to No.55Annual Report (English, annually) FY2012 Edition

7.2.8　Website

The NHK STRL website describes our laboratories and their research and posts reports and announcements on events such as the Open House and the organization’s journals. We have redesigned the website with a new layout that makes it easy to find information one is looking for. For this year’s Open House, the website encouraged visitors to participate in a lab-oratory expedition event held on the weekend, in collabora-tion with NHK Netclub. It also celebrated the release of theme music “One step ahead”.

7.3 Applications of research results

7.3.1　Cooperation with program producers

Our R&D results have been used in many programs. For in-stance, our multi-viewpoint robotic camera system that uses multiple cameras to display three-dimensional video was used for capturing dynamic images in gymnastics and figure skat-ing competitions. The Super HARP camera that enables shoot-ing in darkness, the ultra-high-speed camera, and the insect microphone performed well in nature and science programs. In FY 2013, we collaborated in the production of 53 programs.

■ Use of multi-viewpoint robotic camera in sport programsOur robotic camera system that can capture multi-view-

point video following the motion of objects was used in the 52nd Artistic Gymnastics NHK Trophy. The system makes time slice video, in which the subject is frozen in time while the viewpoint rotates around it, by displaying different video im-ages captured simultaneously by multiple cameras positioned around an athlete in rapid succession. The camera system made it easier to see the gymnasts’ motions and postures in parallel bars and horizontal bar events by displaying images

from various viewpoints.

■ Virtual studio with hybrid sensorsWe used compact hybrid sensors that can obtain informa-

tion on the camera’s motion and position for shooting in virtu-al studios, which synthesize live-action video and CG images. The hybrid sensors built into hand-held cameras and crane cameras enabled a wide variety of camerawork conveying a strong sense of reality. They were also used in the production of various programs at local broadcasters, as they can display spatial views even in small studio sets by synthesizing CG im-ages over the whole screen.

■ Use of macrodiversity receiver system for mara-thon coverageTo cover sports events such as marathons, video signals sent

wirelessly from mobile motorbikes or relay trucks are received by receivers at multiple locations along the route. Our macro-

STRL website


7　Research-related work　｜　7.2 Publication of research results　｜　7.3 Applications of research results

STRL Dayori NHK STRL R&D BROADCAST TECHNOLOGY

diversity receiver system combines video signals received at widely separated locations (dozens of kilometers). The system expands the area over which stable mobile coverage can be

provided. It was used to cover marathons, golf tournaments, and other events.

7.3.2　Patents

NHK participates in the Digital Broadcasting Patent Pool, which bundles licenses of patents required by digital broad-casting standards under reasonable conditions. The pool es-pecially promotes the use of patents held by NHK to help with the switch to digital broadcasting. We are also protecting the rights to our broadcasting and communications-related R&D as part of our intellectual property management efforts, and

we are actively promoting transfers of patented NHK technolo-gies using the “Technology Catalogue” summarizing NHK’s technologies and at events such as, “Program Technology Exhibition,” held at the NHK Broadcast Center, and “Technical Show Yokohama 2014,” hosted by the City of Yokohama.


7　Research-related work　｜　7.3 Applications of research results

Time slice video of parallel bars event Macrodiversity receiver systemHandy camera with built-in hybrid sensor

Type New Total at end of FY

Domestic Patents 447 1299

Utility models 0 0

Designs 0 0

Overseas 41 232

Total 488 1531

Patents and utility model applications submitted


Contracts 38 289

Licences 29 426

(breakdown)

Patents 17 217

Expertise 12 209

Patents and utility models in use (NHK Total)

Type Total

Technical cooperation projects 28

Commissioned research projects 7

■ Technical cooperation (NHK Total)


Domestic Patents 307 1815

Utility models 0 0

Designs 0 8

Overseas 13 194

Total 320 2017

Patents and utility models granted

■ Patents

Award Winner Award Name Awarded by In recognition of Date

Hokuto Seo Funai Research Promotion AwardFunai Foundation for Information Tech-nology

Research on stacked image sensor with organic photoconduc-tive films

2013/4/13

Hiroyuki Segi, Nobumasa SeiyamaMinisterial Commendation Prize for Science and Technology, Develop-ment Category

Ministry of Education, Culture, Sports, Science and Technology

Development of auto broadcasting system using sound synthesis and speech rate conversion

2013/4/16

Masaki TakahashiMinisterial Commendation The Young Scientists’ Prize

Ministry of Education, Culture, Sports, Science and Technology

R&D on moving object tracing algorism based on video analysis 2013/4/16

Takehiro Sugimoto, Kazuho Ono45th Ichimura Science Award, Con-tribution Award

The New Technology Development Foundation

Development of lightweight, thin loudspeakers using electric field driven high molecules

2013/4/25

Yoichi Suzuki, Akinori Hashimoto, Yoshifumi Matsusaki, Shoji Tanaka, Takeshi Kimura

Satellite Communications Research Award 2012

IEICE Technical Committee on Satellite Communications

A study on Multi-Level Coded Modulation using LDPC Code and Set Partitioning for 8PSK, presented on 2013/2/21

2013/5/8

Masayuki Sugawara ITU-AJ Award ITU Association of JapanContributed research results on ultra-high resolution video to ITU-R and promoted international standardization

2013/5/17

Hiroyuki OkuboITU-AJ International Activity Award, Award for achievement in field

ITU Association of JapanContribution to establishment of requirements for advanced mul-tichannel sound

2013/5/17

Keiji IshiiSpecial Recognition Award(2013 SID)

Society for Information Display (SID) Development of diagonal 145-inch SHV (8K x 4K) PDP 2013/5/20

Takehiro Sugimoto, Masakazu Iwaki, Kazuho Ono, Mit-suo Okano

The Acoustical Society of Japan, 21st Technical Development Award

The Acoustical Society of JapanDevelopment of narrow-angle directivity microphone with sup-pressed backward sensitivity

2013/5/21

Toshiki Arai, Kazuya Kitamura, Jun Yonai, Hiroshi Oo-take, Tetsuya Hayashida, Hirotaka Maruyama

Tanba–Takayanagi Award, Re-search Paper Award

Institute of Image Information and Televi-sion Engineers (ITE)

Improvement of characteristics of ultra-high speed CCD dynamic range

2013/5/29

Yasuyoshi Miyamoto, Mitsunobu Okuda, Eiichi Miyas-hita

Fujio Frontier AwardInstitute of Image Information and Televi-sion Engineers (ITE)

Research on ultra-high speed recording device using magnetic nanowire

2013/5/29

NHK & Panasonic SHV PDP development teamHi-Vision, next-generation television technology award

Institute of Image Information and Televi-sion Engineers (ITE)

Development of PDP for Super Hi-Vision 2013/5/29

Taiichiro Kurita FellowInstitute of Image Information and Televi-sion Engineers (ITE)

Achievement in image information and television field and con-tribution to ITE

2013/5/29

Ultra-high resolution spin-SLM R&D groupHODIC Suzuki-Okada Memorial Award

JSAP Holographic Display Artists and Engineers Club

Development of fast-response spatial light modulator driven by spin-transfer switching with 1μm pixel pitch

2013/5/31

Kazuya Kitamura Walter Kosonocky Award International Image Sensor Society, Inc.Papers on SHV image sensors selected for the Best Paper for last two years

2013/6/15

700-MHz frequency migration R&D group24th Meritorious Award on Radio, Ministerial Commendation

Association of Radio Industries and Busi-nesses (ARIB)

Contribution to frequency migration of 700 MHz broadcasting systems

2013/6/19

Daiichi Koide, Takeshi Kajiyama, Yoshimichi Takano, Ryuji Sato

ISOM'13 The Best Technical Paper Award

ISOM Technical Program CommitteeRecording Characteristics of High- Density Thin Optical Disk Us-ing Near-Field Optical Recording

2013/8/22

Daisuke Kato Suzuki Memorial AwardInstitute of Image Information and Televi-sion Engineers (ITE)

Development of a Spatial Light Modulator Driven by Spin-trans-fer-switching for Electronic Holography Applications and Recon-struction of 3D Images with Wide Viewing-Zone Angle from GMR Hologram

2013/8/29

Takeshi Kajiyama Suzuki Memorial AwardInstitute of Image Information and Televi-sion Engineers (ITE)

Real-time protocol conversion system for file-based storage of Ultra High Definition Television and Real-time Transmission via a Network for File-based Storage of Ultra High Definition Television

2013//8/29

Yoshihiko Kawai12th Forum on Information Technol-ogy Encouragement Award

Information Processing Society of Ja-pan, Forum on Information Technology Operations Committee

Object detection method using geometric phrase pooling and large-area characteristics

2013/9/6

Nobumasa Seiyama Testimonial for Editorial ActivitiesThe Institute of Electronics, Information, and Communication Engineers (IEICE), Engineering Sciences Society

Devoted activities as editorial member of English journals 2013/9/18

Hiroshi TsujiTechnical Committee Encourage-ment Award

The Institute of Electrical Engineers of Japan (IEEJ), Electronics, Information and Systems

Development of Simulation Model for Oxide Semiconductor Thin-Film Transistors

2013/9/27

Masayuki SugawaraAward for Person of Cultural Merits of Tokyo Citizen, Technology Devel-opment

Tokyo Metropolitan Government Research and standardization of Super Hi-Vision 2013/10/1

Tetsuhiko Muroi, Nobuhiro Kinoshita, Norihiko Ishii, Koji Kamijo, Hiroshi Kikuchi

IWH2013 Best Paper AwardIWH (International Workshop on Holog-raphy and Related Technologies)

Holographic data storage with wavefront compensation and par-allel signal processing for readout of beyond HD video signal

2013/10/17

Millimeter-wave mobile camera development group66th MPTE Technology Develop-ment Award

Motion Picture and Television Engineer-ing Society of Japan, Inc.

Development of millimeter-wave mobile camera 2013/10/24

Yasutaka MatsuoImage Coding Symposium, Best Poster Award

The Institute of Electronics, Information, and Communication Engineers (IEICE), Image Engineering Technical Group

Research on ultra-high resolution video coding using spatial bit-depth reduction and reconstruction

2013/11/6

Masahide Goto, Kei Hagiwara, Yoshinori Iguchi, Hiroshi Ohtake

Outstanding Poster AwardThe Institute of Electrical Engineers of Japan (IEEJ), Sensors and Microma-chines

Design of an In-pixel A/D Converter for 3D-structured Image Sensors

2013/11/7

Kazuhiko Shibuya, Koichiro Imamura, Hiroyuki Hama-zumi, Tomoaki Takeuchi

Kanto Region Invention Award, In-vention Encouragement Award

Japan Institute of Invention and Innova-tion

Excess guard interval multipath equalization device 2013/11/8

Ultra-Realistic Communications Forum, 3D image evaluation WG

I3DS Good Practice Award 2013 International 3D Society Japan Commit-tee (I3DS-J)

Evaluation of fatigue caused by watching 3DTV 2013/11/20

Tomoyasu Komori, Atsushi Imai, Nobumasa Seiyama, Reiko Takou, Tohru Takagi

Human Communications Award IEICE Human Communications GroupDevelopment of an Adjusting Device for Elderly Person's TV : Aim to make broadcasting sounds easier to listen to

2013/12/19

Kazuhisa Iguchi, Atsuro Ichigaya, Yasuko Sugito, Shin-ichi Sakaida, Yoshiaki Shishikui

Best Academia Paper Award, 2014 FIRST PLACE ACADEMIA

ICCE(IEEE International Conference on Consumer Electronics)

HEVC Encoder for Super Hi-Vision – the world’s first hardware encoder for SHV

2014/1/12

Kinji Matsumura Kenjiro Takayanagi Memorial Award Kenjiro Takayanagi FoundationResearch on service platform technologies for convergence of broadcasting and telecommunications

2014/1/20

Masanori SanoInternational Standard Development Award

Information Processing Society of Japan, Information Standards Study Group

Contributions as project editor on two international standards (ISO and IEC)

2014/1/27

Shingo Asakura Research encouragement awardInstitute of Image Information and Televi-sion Engineers (ITE)

No. of lectures in FY2013: 3 2014/1/30

Shinichi Suzuki Research encouragement awardInstitute of Image Information and Televi-sion Engineers (ITE)

No. of lectures in FY2013: 3 2014/1/30

Automatic closed-captioning system using speech rec-ognition development group

Maejima Award Tsushinbunka AssociationDevelopment of auto closed-captioning system using speech recognition technology

2014/3/25

7.3.3　Prizes and degrees

NHK STRL received 36 prizes, including the Meritorious Award on Radio and the Takayanagi Memorial Award. Six re-searchers obtained new doctoral degrees in FY 2013, and at

the end of FY 2013, 84 STRL members held doctoral degrees.


7　Research-related work　｜　7.3 Applications of research results

Head of STRL Shuichi Fujisawa Executive Research Engineer Tadashi Nakayama

Deputy Head of STRL Toru Kuroda

Planning and management of research, public relations, international support, domestic / international liaison, etc.

Ryo Ikezawa

Director

Takashi KatoPatent rights management, technology transfers, etc.

Hybridcast, security, production and utilization of metadata, content recommendation, etc. Makoto Yamamoto

Satellite/terrestrial transmission technology, millimeter wave and optical SHV contribution technology, multiplexing technology, IP transmission technology, etc.

Tetsuomi Ikeda

SHV program production equipment, video coding for efficient transmission, highly realistic audio systems, etc.

Masayuki Sugawara

Speech recognition and synthesis, advanced language processing, such as simple Japanese and sign language CG creation, tactile and haptic presentation of information

Mahito Fujii

Spatial 3D video system technologies (integral 3D, etc.), 3D display device technology, cognitive science and technology, etc.

Yuichi Iwadate

Naoto HayashiUltra-high resolution and ultra-sensitive imaging devices, high-capacity fast-write technology, sheet-type displays, etc.

Harue MatsumotoStaff administration, accounting, facilities management, etc.

Planning and Coordination Division

Patents Division

Integrated Broadcast-Broadband Systems Research Division

Advanced Transmission Systems Research Division

Advanced Television Systems Research Division

Human Interface Research Division

Three-Dimensional Image Research Division

Advanced Functional Devices Research Division

General Affairs Division

B-SAT3a（Provided by: B-SAT/LMCSS）

1925 : Radio broadcasting begins

U.S.-made television purchased for the home of the first subscriber

1953 : Television broadcasting begins

1930 : NHK Technical Research Laboratories established

1964 : Hi-Vision research begins

1966 : Satellite broadcasting research begins

1989 : BS Analog broadcasting begins

1982 : Digital broadcasting research begins

1995 : Super Hi-Vision research begins

2020 : Full service broadcasting of Super Hi-Vision scheduled to begin

2016 : Super Hi-Vision test broadcasting scheduled to begin

2011 : Switchover to all-digital television broadcasting

2006 : One-Seg broadcasting begins

2003 : Digital terrestrial broadcasting begins

2000 : BS Digital broadcasting begins

1991 : Hi-Vision test broadcasting begins（analog）

The NHK Science & Technology Research Laboratories（NHK STRL）is the sole research facility in Japan specializing in broadcastingtechnology, and as part of the public broadcaster NHK, its role is to lead Japan in developing new broadcasting technology and contrib-uting to a rich broadcasting culture.

■ History of broadcasting development and STRL

■ NHK STRL organization

■ STRL by the numbers

■ STRL Open House

■ Current research building

（at end of FY2013）

Patents held Domestic（NHK Total） International 194

Degree-holding personnel 84 Qualified patent agents 1

Employees （including 225 researchers）

Established in June 1930June 1930 - January 1965 Technical Research LaboratoriesJanuary 1965 - July 1984 Technical Research Laboratories,

Broadcast Science Research LaboratoriesJuly 1984 - Present Science & Technology Research Laboratories

Completed March 2002High－rise building :

14 floors above ground, 2 below groundMid－height building :

6 floors above ground, 2 below groundTotal floor space : Approx. 46,000m2

Includes research area : Approx. 16,000m 2

Total land area : Approx. 33,000m2

NHK Science & Technology Research LaboratoriesOutline

（at end of FY2013）

48 ｜ NHK STRL ANNUAL REPORT 2013

252

1,815

The STRL Open House is held every year in May to introduce the public to our R&D.



To ShinjukuNHK STRL

Bus stopBus stop

Bus stopToho


and Development

TomeiExpressway



Setagaya Dori

Yoga

Tokyu

Den-en-toshi line

To Shibuya

Shuto

Expressway

Yoga I. C.

Access to NHK STRL



Directions




Table of Contents

Greetings …………………………………………1


……14

…

5noisiV-iHrepuS1.1


5saremaC2.1.1

6syalpsiD3.1.1


16Hybridcast1.1.3



20

24




6.1.1


11

13




…………16











technology …………………………………28





























14


2.1.2

16


To ShinjukuNHK STRL

Bus stopBus stop

Bus stopToho


and Development

TomeiExpressway



Setagaya Dori

Yoga

Tokyu

Den-en-toshi line

To Shibuya

Shuto

Expressway

Yoga I. C.

Access to NHK STRL



Directions




Table of Contents

Greetings …………………………………………1


……14

…

5noisiV-iHrepuS1.1


5saremaC2.1.1

6syalpsiD3.1.1


16Hybridcast1.1.3



20

24




6.1.1


11

13




…………16











technology …………………………………28





























14


2.1.2

16

20

13

Nippon Hoso Kyokai [Japan Broadcasting Corporation]

August 2014

2013


Annual Report

2013


Annual Report

Annual R

eport

Date post:	28-May-2020
Category:	Documents
Upload:	others
View:	5 times
Download:	0 times

ANNUAL REPORT 2013 - NHKAccomplishments in 2013 2 ｜ NHK STRL ANNUAL REPORT 2013 Television...

Documents