Texas Instrument - TMS320TCI6488 W-CDMA DSP SoC

8/8/2019 Texas Instrument - TMS320TCI6488 W-CDMA DSP SoC

1/22

TMS320TCI6488 W-CDMA DSP SoC

Introduction This white paper will highlight the Digital

Signal Processor (DSP) based System-on-

Chip (SoC) approach for high-performance

wireless Base Stations. TIs

TMS320TCI6488 device will be used as an

example to highlight the benefits of a flex-

ible DSP SoC approach for W-CDMA base-

band processing. A DSP SoC can provide asingle-chip solution, lower power usage,

better performance, more frugal use of

board real estate, simpler integration, and

lower part counts. Compared to multi-chip

solutions, the DSP SoC has several advan-

tages that will be discussed in this paper.

Introduction to TMS320TCI6488 The TMS320TCI6488, as shown in Figure 1, is a very high-performance base

DSP SoC designed specifically for W-CDMA base stations.

This three-core DSP, running at 1GHz per core, supports all of the necessar

band functions required for a macro base station on a single chip. Designed

ically to solve problems at a system level, this baseband on a chip eliminate

need for Field Programmable Gate Arrays (FPGA) and Application-Specific

Circuits (ASIC) and other bridging devices, reducing the total bill of material

Original Equipment Manufacturers (OEMs) by up to a factor of five, resulting

ered equipment costs for service providers.

The TCI6488 is built on the latest cutting edge technology, the new 65-nm

node, allowing an unprecedented level of functional integration leading to a v

high-performance, high-density modem solution. This allows the device to pe

a level that is an order of magnitude higher than the previous process node at

tion of the power consumption.

W H I T E P A P E R

By Sandeep Kumar

Strategic Marketing Manager,Communications Infrastructure;

Alan Gatherer Chief Technical Officer,Communications Infrastructure

(Continued next column)

Enhanced

DM

A

Controller

L2

Cache

Memory

512

K Bytes

Enhanced

DM

A

Controller

L2

Cache

Memory

512

K Bytes

GPIO PLL I C2

Timers Others BootROM

McBSP AntennaInterface

EDMA 3.0 with Switch Fabric

DDR-2I/F

C64x+Core

C64x+Core

C64x+Core

L1 Data L1 Data L1 DataL1 Prog L1 Prog L1 Prog

L2 Memory L2 Memory L2 Memory

RSA RSA RSA

10/100/1GEthernet

SerialRapidIO

VCP2

TCP2

RAC

Figure 1: TMS320TCI6488 Block Diagram


2/22

With a high level of functional integration and a high channel density supported on a

single device, this DSP offers a modular and scalable design with a small footprint.TCI6488 is a software programmable solution that is upward code-compatible with prev

ous devices belonging to the C6000 DSP family, and allows for the reuse of existing DS

software.

Time-to-Market

The time-to-market benefits of using a standard DSP SoC are clear. Eliminating ASIC

design from scratch can lead to significant time savings. In addition, not having to re-sp

one or multiple of the various functional devices in a legacy system can be the differencbetween an OEM getting a majority of the market share and missing the market window

As illustrated in Figure 2 above, being slightly late to market, ensures missing a signific

portion of the revenue and virtually ensures not capturing the entire market when the

product is finally released.

The total revenue lost in this scenario can be calculated as the difference in area

between the two triangles the red shaded area.

Assuming a constant market ramp, on-time revenue is half (base) (height)

= 0.5 Product Lifetime (L) Maximum Revenue (R) = 0.5 L R

And delayed revenue is half (base) (height)

= 0.5 Delayed Product lifetime (LLD) Delayed Maximum Revenue (RD)

= 0.5 (LLD) RD

Total Revenue Loss is = 0.5 L R 0.5 (LLD) RD =0.5 [L R(LLD) RD]

Texas Instruments

TMS320TCI6488 W-CDMA DSP SoC White Paper May 2007

0.5L L

On-Time SlopeMarketRamp

Late to MarketSlope

Loss Due to

Late to Market

R

RD

LDFigure 2: Revenue Lost Due to Delayed Market Entry

General Benefits of

DSP SoC


3/22TMS320TCI6488 W-CDMA DSP SoC White Paper May 2007

Texas Instrumen

Example of Late to Market Revenue Loss

For example, if the total product lifecycle (L) is 48 months, the Max Revenue is $10M p

month, the project delay (LD) = 6 months and the delayed maximum revenue is $7M,

The Total Lost Revenue = 0.5 (1048 (42)7) =$93M!!(Almost40%of potential

total revenue from the project)As can be seen, the revenue lost due to delayed market entry has components of both

delayed market start and inability to hit 100% market share as components of the lost re

enue. This does not yet comprehend the additional development expense for the duratio

of the delay as well as the lost margin in having to compete for market share when com-

ing from a late-to-market position.

Total Cost of Ownership

Lower Development and Fixed Cost

As can be seen in Figure 3 below, the SoC approach considerably lowers the cost of

development.

The NRE costs associated with developing IP blocks, generating and re-spinning ASI

well as testing and qualifying various devices can run in the 10s of millions of dollars.

There is also another component of the initial investment the additional cost of havi

to develop custom prototyping boards for ASIC, as off-the-shelf solutions are not availa

with the custom pin-out. DSP SoCs, on the other hand, have a variety of off-the-shelf st

dardized board solutions available. These are generally also lower in cost as they are so

across multiple OEMs and application areas.

Total$Am

ount

NRE & Fixed Cost Custom

NRE & Fixed Cost DSP SoC

Total Cost DSP SoC

Volume

Total Cost Custom

Total Revenue

Variable Cost Custom

Breakeven Custom

Variable Cost DSP SoC

Breakeven DSP SoC

Figure 3: Standardized DSP SoCs Lower NRE and Fixed Cost


4/22

Texas Instruments


Lower Bill of Materials

In W-CDMA applications, the TCI6488 DSP eliminates the need for ASIC or FPGAs fo

chip-rate processing. This results in significant unit cost savings. In addition, through th

integration of standardized system-level interfaces such as Serial RapidIO and OBSAI, it

also eliminates the need for bridges or glue logic, allowing the device to seamlessly com

municate with the backplane and a host of other devices. As such, the TCI6488 can lowthe total Bill of Materials (BOM) for a system by up to one-fifth of legacy systems as se

in Figure 4.

Compared to an all ASIC approach, the DSP SoC approach provides a good balance o

flexibility and cost efficiency. DSP SoCs are used across multiple air-interface standards

due to their support of multiple standards, and across multiple generations of systems,

due their ability to support emerging features and requirements. Typically, this leads to

DSP SoCs shipping in volumes that exceed what any single custom device would be abl

to achieve in isolation. These volumes are leveraged to lower silicon costs and are used

offer a very attractive price point to help OEMs meet their BOM targets for established

markets, as well as niche and emerging applications.

There are several scenarios where the TCI6488 DSP SoC further demonstrates its abil

to fit into various partitions and resource allocation approaches. In one such example, as

shown previously in Figure 4, the device can be used to construct a 192-user 3GPP

Release 99 system quite easily using four devices. The same device can also be used to

Rx SRDSP

Rx SRDSP

Rx CRDSP

Rx CRDSP

Rx CRDSP

Rx CRDSP

Traditional Rx Card

10

Rx CR

192 DL & UL User Card based on Faraday

EthernetSwitch

(Optional)Traditional Rx CardTraditional Rx Card 64 Users

Tx SRDSP

Traditional Tx Card 192 Users

4 5 Devices

HostProcessor

HostProcessor

FPGAs/ Bridges?

FPGAs/ Bridges?

Serial RIO

192 DL & UL User Card Based on Faraday

Backplane to Radio

OBSAI / CPRI

OBSAI / CPRI

TxASIC

Rx SRDSP

10 18 Devices Total

Rx CRDSP Rx CR

ASIC

OBSAI / CPRI

RapidIOSwitch

(Optional)

Backplane to Network

HostProcessor

TCI6488

DSP

TCI6488

DSP

DSP

TCI6488TCI6488

DSP

Figure 4: TCI6488 Can Reduce BOM Cost by a Factor of 5



Texas Instrumen

create a High-Speed Downlink Packet Access (HSDPA) system. A virtually complete ba

band system with 192-User, 3-Sector RACH, HSDPA / HSUPA (additional small device

needed for transmit chip rate functionality) can be implemented using just four devices a

seen in Figure 5. This simple example demonstrates how, through software modification

the same architecture (if not the same channel card) can be used for voice-only users,

HSDPA/HSUPA users, dedicated data users or various combinations of three.

Software Programmable Solution

The move from modems dominated by lots of low data rate voice calls to modems with

few large data rate data channels that are time multiplexed has actually decreased the

front end signal processing complexity. This is seen conceptually in Figure 6 on the foll

ing page.

As the data portion of the processing mix on the modem increases, there has been a

steady increase in modem complexity because this processing is correlated with the data

bit rate. On the other hand, the signal processing complexity actually decreased as we

moved away from voice dominated standards and towards time-multiplexed, shared cha

nels, as used in HSDPA. The addition of antenna arrays and more complicated receiver

structures with HSUPA has increased the complexity of the signal processing. But never

theless, the role of data processing has increased significantly and this type of processin

has always been done in software. Hence, modems are becoming increasingly soft.

TCI6488

DSP1

TCI6488

DSP2

TCI6488

DSP3

TCI6488

DSP4

GigabitEthernet toNetwork

Backplane

Ethernet

Switch

SRIO

SRIO

SRIO

DSP SoC 1, 2, 3:64 UL DCH +

1 cell RACH PD +8 RACH m +64 R.99 SR

DSP SoC 4:3 cell Mac-HS +

3 cell HSDPA SR +192 R.99 SR

AIF to AntennaBackplane

AIF

AIF

AIF

Figure 5: 192-User, 3-Sector RACH, HSDPA/HSUPA on TCI6488


6/22

Texas Instruments

Software programmability on the baseband, such as that available on the TCI6488, isbecoming more desirable.

By using the TCI6488, OEMs can accelerate their channel card development since it

offers a software-programmable solution and allows for the reuse of existing DSP soft-

ware developed for previous generations of TI infrastructure DSPs. TCI6488 is upward

code-compatible from previous devices that are part of the TMS320C6000 DSP platfo

Customers can re-use their existing software code base and evolve their system architec

tures ensuring they get to market more quickly.

Easier Higher Layer Software Provisioning

The main benefit of a system allowing easily provisioned software is quick time to mark

A system using the TCI6488 can provide this benefit in three ways:

Single-Chip Physical Layer Functionality

From the higher layer software perspective, each DSP SoC looks identical and operates

a user level. This is because a complete user can reside on a single SoC. Therefore the

board-level resource manager does not have to be cognisant of the resource mapping on

the hardware, simplifying resource management. The resource manager on the SoC is in

charge of load balancing the users it is given.

Aggregated MAC Functionality

Traditionally, higher-layer software also has to deal with the task of partitioning physica

control and signalling as well as MAC layer functionality across separate devices.

SignalProcessing

Dedicated ChannelsMany Fingers

C o m p l e x

i t y

Standards evolution

Scarce SpectrumSingle, Shared chanelHSDPA, OFDM

HSUPA, OFDMAAntenna Arrays, MIMO

Higher DL Data RateMAC Functionality

Higher ULData Rate

DataProcessing

Figure 6: Signal Processing Complexity vs. Standards Evolution



Texas Instrumen

Typically, the software layers of the modem protocol were separated onto different CPU

with the Layer 2 protocol and the board control being implemented on a control proce

sor, usually a PowerPC. One reason for this split was that the control processor devices

had a more suitable peripheral set for communicating with upstream components and ha

better software development tools for this type of code.However, over the last few years, the software development support for Texas

Instruments DSPs has improved significantly with industry competitive C, C++ and mi

Linux support. The TMS320C64x+ CPU used on the TCI6488 also supports memory

agement and a user/supervisor mode. TCI6488 supports both Ethernet and Serial RapidI

and therefore does not suffer from being interface poor for board level and Layer 2

processing.

Greater DSP Horse Power Allows System Homogeneity

With 1-GHz performance per DSP core, the TCI6488 has sufficient processing power o

the three CPUs to absorb the L2 processing and the board control functions on to the D

SoC. This gives system designers greater flexibility as to how the software is partitioned

This has the dual-purpose benefit of simplifying upper-layer software provisioning as w

as reducing the modem board BOM.

Easier Software Replication and System Testing

A direct consequence of the system homogeneity discussed above is that the same physi

cal layer (and potentially L2) software would be written once and then replicated across

multiple TCI6488 devices making the software development easier and enabling shorter

time-to-market.

Also, since every SoC is self contained, when designing a higher-performance system

adding more SoCs, the channel card would require little or no testing at the SoC level.

This would mean minimal additional testing at a system level when scaling a system.

Investment Protection via Flexibility

Feature additions and standards evolution are now routinely expected in the wirelessdevelopment world. For example, the difference between GPRS and W-CDMA maturity

deployment is shown in Figure 7 on the following page. A non-editorial change request

(CR) is essentially an engineering bug in the standard. Increasingly, standards are

launched early and equipment is deployed with bugs and other uncertainties still present


8/22

Texas Instruments


in the standards. Wireless operators demand that this early equipment be fixed in thefield after deployment. Replacement of equipment is considered an ugly (and expensive

fix for the problem and is only available as a last resort option.

Add to this every carriers requirement to differentiate themselves via a slightly differ

ent twist on the standards specification and it is the recipe for frequent changes to sys-

tem designs that OEMs must make, sometimes mid-stream in a design. Also, in the early

stages of any system development project, it is often not clear which functions should be

implemented with hardware and which in software. Consequently, the ability to make

changes during development without being too disruptive can be valuable. Given the

recent history of tighter development budgets, its no wonder that more of the OEMs are

looking at flexible implementations such as DSP SoC. Also, the cost of ASIC developm

is rising and becoming harder to justify with typical infrastructure volumes. This is more

clearly seen with emerging wireless markets where the uncertainty of gaining critical

market share is even greater.

The DSP SoC, like the TCI6488, allows OEMs to have the flexibility to deal with feat

creep. Often they can begin development without nailing down the complete architectur

of the system knowing that, through software, the TCI6488 is able to support a variety o

system partitions and configurations.

Improved Power Efficiency

The increased processing demands of todays advanced wireless networks have also

increased power consumption. The ever-growing demand for high-capacity, high-data-ra

signal processing in base stations raises a chronic technology challenge. The challenge i

to squeeze higher functionality and performance within increasingly tighter power and

System-Level Benefits of

DSP SoC

0

200

400

600

800

1000

1200

Source:

Non-Editorial CRs

J u n 9

7

J u n 9

8

J u n 9

9

J u n 0

0

J u n 0

1

J u n 0

2

J u n 0

3

J u n 0

4

D e c 9

7

D e c 9

8

D e c 9

9

D e c 0

0

D e c 0

1

D e c 0

2

D e c 0

3

D e c 0

4

ETSI CR Database 11 Jan 2005

W-CDMA Total

GPRS Total

W-CDMA Test

Source: Agilent Technologies

Early GPRSDeployment

Global GPRS

GPRS SubscriberGrowth

Early R99 W-CDMADeployment

Figure 7: Open Change Requests Comparison Between GSM/GPRS and W-CDMA



Texas Instrumen

space constraints. As a result, power-performance metrics are now a central concern in

infrastructure system design. New methods have been devised enabling designers toaddress the two main areas of power consumptionnamely leakage power and dynamic

powerto significantly improve performance compared to conventional techniques.

To solve this problem, the TCI6488 DSP SoC sports a new power and performance m

agement technology called SmartReflex. Texas Instruments SmartReflex technology

used to decrease both static and dynamic power consumption while maintaining the spe

ified device performance.

This technology takes advantage of the fact that a devices leakage and associated

power can vary significantly across a process distribution as shown in Figure 8.SmartReflex technology make use of this phenomenon by adjusting the core voltage of

devices to suit the process type of the material.

Another element of SmartReflex technology is enhancing device performance and opt

mizing power efficiency by considering thermal parameters. To achieve improved powe

efficiency, the device implements temperature-dependent dynamic voltage scaling.

The final solution allows full performance utilization of highly integrated silicon desig

such as TCI6488, with optimized power consumption. A comparison of the 65-nm node

(with and without SmartReflex technology) with the 90-nm node is shown in Figure 9 o

the following page. This allows designers the ability to add multiple TCI6488 devices o

single card, while still meeting the board power budget.

Another primary component of dynamic power is the chip-to-chip interface between

multiple devices on legacy systems. Through system integration, the DSP SoC approach

ensures that device-to-device communication is kept at a minimum, reducing the overal

system power consumption.

Distributionof Devices

Power

1 GHz

Weaker StrongerProcessStrength

Performance

Figure 8: Device Leakage and Power Can Vary Significantly Across Process Distribution


10/22

0 Texas Instruments


Smaller Form Factor

The efficient integration of the various system-level components on the DSP SoC also

allows for a smaller number of input and output signals (I/Os) and package ball count,which in turn reduces the size of the device. This has the dual benefit of reducing packa

ing cost as well as allowing a greater device density on a single board, thus allowing sy

tem designers to pack more functionality on their board with every generation. The

TCI6488 fits into a small 23 mm 23 mm flip chip BGA package.

With fewer signals leaving the device, DSP SoCs such as the TCI6488 also have the

added benefit of simplifying board layout and reducing board complexity associated boa

cost.

Greater Scalability

According to some analysts, the market for small form factor base stations is on the rise

(see Figure 10).

Relative Power per Function Relative Power per Unit Area

90 nm 90 nm65 nm 65 nm

Process Node Process Node

65 nm withSmartReflex

65 nm withSmartReflex

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

3.0

2.5

2.0

1.5

1.0

0.5

0.0

Relative Dynamic Power

Relative Leakage (Static) Power

Relative Dynamic Power

Relative Leakage (Static) Power

Figure 9: 65-nm Node (With and Without SmartReflex) Compared with 90-nm Node

Percentage of BTS Shipments

12%

W-CDMA

GSM/GPRS/EDGE

CDMA

9%

6%

3%

0%

2 0 0 1

2 0 0 2

2 0 0 3

2 0 0 4

2 0 0 5

2 0 0 6

2 0 0 7

2 0 0 8

2 0 0 9

2 0 1 0

Figure 10: Forecast of Pico Base Station as a Percentage of Total Base Stations Shipped

Source: Dell OroMobility ForecastReport 3Q06



Texas Instrumen

However, the uncertainty around the timing and slope of the pico base station market

ramp make this a questionable market for a custom chip development from the very beg

ning. As such, a phased approach starting with a standard DSP SoC moving to a custom

SoC is recommended for such markets.

The beauty of a DSP SoC approach is that it allows designers to pick an appropriate

granularity of system performance and then replicate that to match the end system as

shown in Figure 11.

A single TCI6488 can support a complete HSPA pico base station. Thus, software can

be, written, verified and qualified once for such a single DSP SoC to support a pico imp

mentation and then replicated on top of multiple devices to create larger system configu

ration, such as micro or macro, in the future. In addition, the availability of high band-

width, peer-to-peer interconnect like Serial RapidIO, enables OEMs to reduce overall

R&D by allowing it be re-used across form-factor product lines ranging from small ente

prise class pico base station, all the way to a super macro covering over a 100-Km cell.

Conversely, if the modem designer wishes to dedicate a chip to only one function, sucas Random Access Channel (RACH) messaging, this is also possible with the TCI6488.

designer can therefore trade off the advantages of code that is repeatable across multiple

devices with the efficiency that might come from concentrating a single function in one

device.

TCI6488

DSP

TCI6488

DSP

TCI6488

DSPMicro

Macro &Super Macro

TCI6488 to TCI6488 data transport via: Serial RapidIOAntennae data transport via: AIF with daisy chain aggregation

Network data transport via: Integrated Ethernet

PicoTCI6488

DSP

TCI6488

DSP

TCI6488

DSP

TCI6488

DSP

TCI6488

DSP

TCI6488

DSP

TCI6488

DSP

Figure 11: Standard DSP SoCs Allow for Easy System Scalability


12/22

2 Texas Instruments


Simpler Redundancy and Load Balancing

Inherent in DSP SoC architecture is the built-in multi-function redundancy. Since all the

baseband functions are performed on the TCI6488, system redundancy can be implemen

ed by just adding an additional TCI6488 to the system.

With proper system design, it may be possible to have system redundancy withoutadding an additional device. In such cases, the system software would sense if a function

al unit or logical block on the DSP SoC was not performing and simply switch the task t

another DSP SoC via the Serial RapidIO subsystem. However, this approach will likely

require the system to drop to a fewer number of supported users.

Via the same approach, the presence of a Serial RapidIO-connected DSP subsystem al

makes it easy for the system to balance the processing load between various DSP SoCs

the system.

Multi-Standard SupportBack almost a decade ago at the start of the 3G standards development, industry pundits

predicted that the world would converge to one wireless air interface. The mess of stan-

dards that had been 1G had given way to a slightly better situation, at least in Europe,

with 2G, and the hope was that 3G would finish the job. Now, a decade later, this utopia

ideal has been replaced by a fragmented world where multiple standards not only surviv

but are being pushed to compete and inter-operate at the same time.

The explosion in the number of wireless air interfaces compounded by a simultaneous

cut back in capital expenditure is forcing OEMs to consider ways to address multiple maket opportunities while getting the most out of their architecture investments.

This is where TCI6488 comes to the rescue with its ability to support UMTS, GSM, T

SCDMA, WiMAX and cdma2000 applications as seen in Figure 12. The three 1-GHZ D

Figure 12: TCI6488 Enables Multiple Wireless Modem Standards



Texas Instrumen

cores go a long way in enabling this device to support multiple standards, while the acce

eration of the mature, compute intensive W-CDMA functionalities makes sure it is able

meet the aggressive W-CDMA cost efficiency targets. This flexibility in a small form fa

tor, scaleable solution provides OEMs a unique solution to support established markets

and extend these into new and emerging infrastructure applications.

Primary W-CDMA Functional Processing

Symbol Rate and MAC-hs Processing

Uplink and downlink symbol rate processing, as well as MAC-hs processing, is perform

on one or more of the three TMS320C64x+ 1-GHz DSP cores available on the TCI64

TCI6488 is upward code-compatible from previous devices that are part of the

TMS320C6000 DSP platform, with the TCI6488 allowing system designers to re-use

existing software code base.

In addition, twenty eight (28) new instructions have been added to this DSP core with

large number of these targeted at improving performance in wireless infrastructure appli

cations. As such, designers can improve their code performance on a per cycle basis sim

ply by taking advantage of these new instructions.

FEC Decoding for Voice and Data

The device has two high-performance embedded coprocessors [enhanced Viterbi Decod

Coprocessor (VCP2) and enhanced Turbo Decoder Coprocessor (TCP2)] that significan

speed up channel-decoding operations on chip. The VCP2 can decode over 763 12.2-Kbadaptive multi-rate (AMR) [K=9, R=1/3] voice channels. The TCP2 can decode up to 4

384-Kbps or 8 2-Mbps turbo encoded channels [assuming 6 iterations].

Both TCP2 and VCP2 are loosely coupled into a DSP subsystem and are fully parame

ized to flexibly support various options within the 3GPP specifications. The presence of

VCP2 and TCP2 frees up DSP resources for other processing and enables basebands wi

much higher voice and data density than would have been possible with a DSP-only

approach.

Chip-Rate Processing

Transmit Chip-Rate Acceleration Using RSA

Transmit chip-rate processing on the TCI6488 is implemented by a DSP subsystem and

associated Rake Search/Spread Accelerator (RSA) extensions. These RSA extensions

accelerate CDMA transmit processing by performing the spreading and scrambling func

tions. The RSA extensions are also capable of carrying out the stream aggregation

W-CDMA-Specific Benefits of a

DSP SoC Approach


14/22

4 Texas Instruments


functionality Also, in conjunction with the DSP cores, they can also perform search func

tionality that can be used to augment the Preamble Detect (PD) and Path Monitoring (PM

functions typically performed on the Receive Accelerator.

The RSA extensions allow the TCI6488 to truly shine in supporting very high user de

ty, multiple antennas, variety of data formats as well as an array of system configuration

Thus a base station using the TCI6488 can flexibly support multiple voice and data user

as well as cell sizes over 100 miles in size. All of this is accomplished without sacrificin

cost or power efficiency, in large part due to the RAC.

Figure 13 shows the functional split of transmit chip-rate processing between the DSP

subsystem and the RSA extensions. The DSP core generates both OVSF and PN codes a

provides the multiplied result of these two codes as input to the RSA. The modulated us

symbols are also provided as input to the RSA. The RSA applies the code values to themodulated symbols to achieve spreading and scrambling.

Receive Chip-Rate Acceleration Using RAC

Receive chip-rate processing on the TCI6488 is implemented via the Receive DSP core

the Receive Accelerator (RAC). The RAC is comprised of highly flexible and programm

correlation engines, as shown in Figure 14. These can be configured to carry out various

receive chip-rate functions, including Rake finger de-spreading (FD), EOL finger tracki

(FT), search or path monitoring (PM) operations and RACH preamble detection (PD) options. These blocks receive 2 over sampled chip-rate antenna streams and provide the R

accelerator DSP with either de-spread symbols or correlation energies. These blocks sup

port a very large amount of correlation resources and can support up to 6,144 32-chip co

relations per 32-chip period. The DSP associated with the RAC serves to control and co

figure the two correlation engines.

DSPSystem

OVSF and PNCode Generation

384 channels24 antennastreamsMIMO and

beamformingenabled

1-Bit I,1-Bit Q

EFI

EFI 32-Bit I,32-Bit Q

L2 Memory(User Symbols)

RSAMultiply

and

Accumulate

ModulationApply Closed-Loop Gain

Apply Power-Control Gain

Figure 13: Functional DSP and RSA Split for Transmit Chip-Rate Processing



Texas Instrumen

Advanced and Emerging Features Support

The TCI6488 has support for various Multiple Input Multiple Output (MIMO) antenna cfigurations. Fixed and adaptive beam forming can be enabled on both uplink and downl

via the RSA and RAC. Parallel Interference Cancellation (PIC) is enabled with re-sprea

possible via the RSA. Thus, as the market demand for advanced features materializes,

OEMs using TCI6488 will be able to quickly support advanced features without having

redesign their existing baseband platforms. The soft nature of the DSP SoC also makes

an ideal candidate for OEMs to support emerging features and capabilities on their curre

platforms.

W-CDMA Software Library Elements

The TCI6488 platform is supported by a hardware adaptation software layer, which

abstracts the hardware implementation details and provides a well-defined API to the

upper software layers. A comprehensive set of optimized hardware-dependent functions

are provided as apart of this hardware adaptation layer. In addition, a robust set of

W-CDMA software library elements are also provided to give system designers a head

start in developing a complete physical layer. Alternately, OEMs can build their own pro

cessing elements on top of the hardware adaptation software to complement or replace

the ones provided along with the TCI6488. An API layer on top of the hardware adapta

software and the W-CDMA software library elements allows OEMs the freedom to choo

the level at which they want to develop their differentiated software component.

CorrelationEngines

Receive Accelerator (RAC)

CorrelationEngines

Host InterfaceAntennaInterface

Receive Chip-Rate Processing Using RAC

Flexible support ofFinger De-Spread (FD),Path Monitoring (PM)& Preamble Detect (PD)

48 Antenna Streams

Continuous SearchSupport

Large Cell Size Support(160 km)

Beamforming Enabled

HostInterruptInterface

System (VBUS)Bridge

Figure 14: Receive Chip-Rate Processing Using RAC


16/22

6 Texas Instruments


W-CDMA on DSP SoC Architecture Considerations

Introduction

A base station modem is a real-time system with multiple users, all with their own real-

time constraints, being supported simultaneously. Recent modems that support data tran

fer have the added complexity of having to support many different users with different

data rates and QoS requirements leading to differing I/O, MIPS, and memory require-ments. In order to understand the benefits of the DSP SoC approach for W-CDMA base

stations, it is pertinent to consider the trade-offs made by designers when designing a

multi-user modem.

For the purposes of this discussion, a simplified receiver (demodulation) modem data

flow is considered, as shown in Figure 15, which includes both a low latency control pat

and a higher latency data path.

Additionally, every system is assumed to have at least a basic Real-Time Operating

System (RTOS) to schedule tasks and users.

Task Division

One of the basic decisions concerning whether a basic software task can be performed

relates to a user or a function, as seen in Figure 16. This decision impacts the way

filter

demod FEC unpack

demod FEC unpack

data

control

Figure 15: Simplified Receiver Modem Data Flow

Filt1

D_slow1

Filt2

F_slow1

F_slow2

U_slow1

U_slow2

D_fast1 D_fast1

D_fast2 D_fast2

U_fast1

U_fast2 U_fast2

F_fast1

F_fast2 F_fast2

Tasks Split by User Tasks Split by Function

D_slow2 F_slow2 U_slow2D_slow2

Filt1

Filt2

D_slow1 U_slow1F_slow1

U_fast1F_fast1

Figure 16: Processing Tasks Split by User and Function



Texas Instrumen

interrupts are generated and how often tasks switch. It also affects the way software

interacts with any peripherals and hardware acceleration on the DSP SoC.

Dividing Tasks by Users

If the tasks are divided by users, the RTOS will not know how many tasks will be presen

at any given time. The main issue with dividing the tasks by user is that the number of

tasks grows with the number of users. For instance, on a macro base station, there may b

up to 64 users running on a TCI6488 and these users may require multiple tasks each.

Also, as the number of tasks grows, so generally will the number of task switches per se

ond. Not only is there a crushing number of task structures to manage, but also more tim

will be spent in interrupt routines and in the kernel, and less time doing useful work.

Typically, this can lead to an unmanageable system above a few tens of users.

Dividing Tasks by Functions

If the tasks are divided by functions, the RTOS does not have to know how many users

present in the system. It only has to know how many unique functions are to be per-

formed. As the number of users increases, the time it takes a task to complete will

increase as that task will run for each user that needs that task at that point in time. If the

task is called immediately, when there is data available for each task, then each task will

be called for each user and the number of task switches will increase with the number of

users. This can again lead to a crushing number of task switches.

A Hybrid Approach

A better way to manage this is to assign each task to a linked list. When a task runs to

completion, for each user, it will add an item to the linked list of the task associated with

the next function to be performed on that user. This will not cause an interrupt to be gen

erated and the users will accumulate on the linked list. At some point the task will be

activated and it will process its linked list to completion, or until it is pre-empted.This

method of task definition and processing is generally preferred on the TCI6488.

Key Take-Away From the User and Task Grouping Analysis

In order to keep the number of tasks, and the task pre-emption overhead, to a manageab

level, they must not scale with the number of users. Tasks should be associated with fun

tions and not users. This can be done by setting up a number of queues. Tasks are


18/22

8 Texas Instruments


associated with a queue (usually several tasks per queue) and requests for operation of a

task on a particular user are linked to these queues. The kernel will manage the draining

of these queues and can do this in several ways, using periodic or data-driven interrupts

to achieve real-time operation. As mentioned earlier, this is the recommended approach

when using the TCI6488 for W-CDMA applications.

Scheduling on a Multi-Core DSP SoC with Acceleration

Having considered how to group tasks and users on the TCI6488, the question naturally

arises of how to map users and task on a DSP SoC. The TCI6488 is a multi-CPU DSP S

and composed of independent IP blocks that interoperate and synchronize to achieve a s

gle complete modem function. As such, mapping the priority queue into a multi-CPU en

ronment is an important consideration when designing W-CDMA systems using it.

Mapping Users Across DSP SoC CPUs

One simple way to do this is to divide the users amongst the CPUs, so that each CPU

maintains its own queues. However, some functions, such as filtering and demodulation

may be shared amongst all users. Also, some functions may be required to share

coprocessors or peripherals, and are therefore interdependent.

In this case, the interaction between the sets of priority queues can get quite complicat

ed and it gets difficult to ensure real-time performance. Also, the complexity of the

coprocessors and peripherals increases because they have to support multiple CPUs. Thi

involves making decisions about priority of tasks from different CPUs. All this adds com

plexity to hardware and also to software drivers. It also makes testing of the final system

more complex.

Mapping Tasks across DSP SoC CPUs

Another approach is one of assigning a functional task to a single CPU so that each CPU

in charge of a unique group of functions. Each coprocessor, which generally accelerates

specific type of function, is associated with a single CPU and control of the order of task

performed on that coprocessor is significantly simplified. In many cases peripherals will

only communicate with a single CPU as well. This reduces the testing required to verify

that tasks will not be starved of data.

Synchronization between CPUs can be achieved by system-wide synchronization sig

that align the CPUs to frame, slot and symbol boundaries. Communication between CPU

is in the form of blocks of data generated by one task and destined for another via Direc



Texas Instrumen

Memory Access (DMA) approach between L2 memories. This is the recommended

approach when using the TCI6488.

Recommended Resource Mapping with TCI6488

The architecture of the TCI6488 has been kept as symmetrical as possible so that it may

be used with a variety of functional splits and system partitions. All DSP cores have

access to the Receive accelerator, for instance. It is therefore possible to run the same

functions on all CPUs, and have all CPUs access all coprocessor and peripheral resource

Yet, it is noted that simplicity of software architecture, along with the nature of many

modem algorithms, leads the smart system designer to partition the tasks so that the soft

ware is not symmetric across CPUs. This is why despite the symmetric architecture, the

is still a recommended usage model, the outcome of extensive study of code cycle esti-

mates, spreadsheet analysis and transactional level models.Based on this analysis, software partition recommended for W-CDMA allows the sim

plicity of having only one CPU controlling the RAC, one CPU controlling the TCP and

and one CPU performing transmit chip-rate function as well as communication with the

antenna array interface for output. Each CPU is also equipped with its own L2 memory

is appropriate in an implementation where each CPU has a unique function. This simpli

the operation of the device and allows the system designer to extract maximum applica-

tion level efficiency from the device.

For other standards, such as those based on OFDM, the natural inclination may be to

use a symmetrical software architecture. But even in this case, it is recommended to

divide the problem so that certain functions, such as FFT/IFFT and some modulation an

demodulation is performed by one CPU and the results communicated to another CPU f

symbol-rate processing. In fact, in the case of OFDMA, the modulation is jointly perfor

for all users and users cannot be completely separated onto different CPUs. This certain

simplifies communication between the antenna interface (or Serial RapidIO if this is

used for antenna data) and the CPU processing the front end. This also has the added be

efit of simplifying the back-end symbol-rate processing and its communication with the

network interface.

Balancing Resources on a Multi-DSP SoC System

As it was recommended in the previous section that each CPU perform different tasks

from each other, it is worthwhile considering if the same approach can be extended to

each DSP SoC having a different task in a multi-DSP system. For instance, a scenario c


20/22

0 Texas Instruments


be imagined in which there would be one DSP SoC performing nothing but symbol-rate

decode and one DSP SoC performing nothing but chip-rate modulation.

In a purely soft implementation, this makes sense. However, in such a scenario, any co

processing elements present on chip will not be used efficiently. For instance on a

TCI6488, performing only symbol-rate processing needs a powerful set of VCP2 and TCengines, as present. However, on another DSP SoC performing only chip-rate tasks, the

VCP2 and TCP2 engines would be unused.

Thus, dedicating DSP SoCs to a particular subset of functions also does not make for

scalable system. Clearly, if one wishes to increase the channel density on a board, with

each SoC performing the same complete set of functions, one can simply add more SoC

to the board. The TCI6488 is designed to allow this to happen with the minimal extra h

ware. The Antenna Interface (AIF) and Serial RapidIO both can be connected in a daisy

chain configuration to enable multi-device systems. The gigabit Ethernet and Serial Rapi

interfaces can also be attached to a switch to create a scaleable fabric-based system.

Handling of W-CDMA Tighter Latency on DSP SoC

Compared to a design where each device performs a different task, DSP SoC has the

advantage of latency in transferring data from task to task, while executing the data path

of one user is smaller because the data is kept contained on chip.

System Architecture Conclusion

For a system design using DSP SoCs, such as TCI6488, the system architecture that is thmost scalable at the board level and leads to the simplest, most easily tested software is

one in which each CPU in the DSP SoC performs a unique subset of tasks, but each SoC

the system performs the same set of tasks as the other SoCs. TCI6488 has been optimize

for this scenario for W-CDMA/HSPA, but is flexible enough to also efficiently support

other modem standards and partitions.

The goal for any SoC is to have the perfect balance of on-chip resources. Therefore, pro

erly sizing memory, IO, CPU and other resources from the outset is critical. Comprehen

and balancing resource requirements on a DSP SoC requires a good grasp of the end sy

tem as well as a clear understanding of the real-time constraints. Failure to understand

real-time requirements can be disastrous and lead to resource underutilization, insufficie

capacity, bandwidth, memory or a combination of these. Sharing hardware resources and

multiple threads or services allocated to multiple processor cores increases complexity

and risk. Up-front understanding and avoidance of known resource sharing pitfalls is

Challenges to the DSP SoC

Approach


21/22

Texas Instrumen

strongly suggested. When multiple resources must be shared, interaction should be well

synchronized, brief and well tested.

How TCI6488 Guards Against Typical SoC Pitfalls

Though the TCI6488 was designed from the ground-up to be a very effective real timeW-CDMA baseband processor, it still incorporates features to help the system designer,

should one of the typical problems listed above occur.

Insufficient memory In addition to the hefty on-chip memory already included, the

TCI6488 also allows for a variety of external memory devices supported through its

DDR-2 memory interface.

Inadequate acceleration The presence of Serial RapidIO allows for a very-high band-

width, low-latency option for connectivity to external accelerators, including those

implemented on FPGAs.

Insufficient performance Multiple copies of the TCI6488 can be instantiated. Given th

re-use of the software and the ease of connectivity via Serial RapidIO, getting to the

right performance is easier than ever.

The value in a DSP SoC approach lies in its ability to provide time-to-market advantage

without sacrificing BOM savings. The ease of software provisioning and testing eases th

burden on system software. The software programmable nature of TCI6488 helps insula

OEMs and carriers against unanticipated changes, while allowing them to leverage their

existing code base. The ease of software replication and the enhanced device features,along with power efficiency and small form factor, allow system designers to easily scal

their designs to meet application requirements. DSP SoCs are also particularly well suit

to allow implementation of redundancy and load balancing. Via hardware and software

mechanisms, well-designed DSP SoCs, like the TCI6488, also avoid the typical pitfalls

can plague SoCs.

TCI6488 DSP offers a high-performance, power-efficient and cost-effective baseband

platform capable of supporting physical layer processing, including symbol-rate and chi

rate processing. Designed specifically for the W-CDMA base station, the TCI6488 help

designers achieve unprecedented system density and performance.

Summary and Conclusion

SPRAAM8 2007 Texas Instruments IncorporatedPrinted in the U.S.A.

Important Notice: The products and services of Texas Instruments Incorporated and its subsidiaries described herein are sold subject to TIs standard terms andconditions of sale. Customers are advised to obtain the most current and complete information about TI products and services before placing orders. TI assumeliability for applications assistance, customers applications or product designs, software performance, or infringement of patents. The publication of informatioregarding any other companys products or services does not constitute TIs approval, warranty or endorsement thereof.

All trademarks are the property of their respective owners.


22/22

I M P O R T A N T N O T I C E

T e x a s I n s t r u m e n t s I n c o r p o r a t e d a n d i t s s u b s i d i a r i e s ( T I ) r e s e r v e t h e r i g h t t o m a k e c o r r e c t i o n s , m o d i f i c a t i o n s , e n h a n c e m i m p r o v e m e n t s , a n d o t h e r c h a n g e s t o i t s p r o d u c t s a n d s e r v i c e s a t a n y t i m e a n d t o d i s c o n t i n u e a n y p r o d u c t C u s t o m e r s s h o u l d o b t a i n t h e l a t e s t r e l e v a n t i n f o r m a t i o n b e f o r e p l a c i n g o r d e r s a n d s h o u l d v e r c o m p l e t e . A l l p r o d u c t s a r e s o l d s u b j e c t t o T I s t e r m s a n d c o n d i t i o n s o f s a

T I w a r r a n t s p e r f o r m a n c e o f i t s h a r d w a r e p r o d u c t s t o t h e s p e c i f i c a t i o n s a t h T I s s t a n d a r d w a r r a n t y . T e s t i n g a n d o t h e r q u a l i t y c o n t r o l t e c h n i q u e s a r e u s e dx t e n t T I d e e m s n e c e s s a r y t o s u p p o r t t h iw a r r a n t y . E x c e p t w h e r e m a n d a t e d b y g o v e r n m e n t r e q u i r e m e n t s , t e s t i n g o f a l l p a r a m e t e r s o f e a c h p p e r f o r m e d .

T I a s s u m e s n o l i a b i l i t y f o r a p p l i c a t i o n s a s s i s t a n c e o r c u s t o m e r p r o d u c t d e s i g n . C u s t o m e r s a r e r e s p o n s i b l e f o r t h e i r p r o d u a p p l i c a t i o n s u s i n g T I c o m p o n e n t s . T o m i n i m i z e t h e r i s k s a s s o c i a t e d w i t h c u s t o m e r p r o d u c t s a n d a p p l i cp r o v i d e a d e q u a t e d e s i g n a n d o p e r a t i n g s a f e g u a r d s .

T I d o e s n o t w a r r a n t o r r e p r e s e n t t h a t a n y l i c e n s e , e i t h e r e x p r e s s o r i m p l i e d , i s g r a n t e d u n d e r a n y T I p a t e n t r i g h t , c o p y r i g h t , w o r k r i g h t , o r o t h e r T I i n t e l l e c t u a l p r o p e r t y r i g h t r e l a t i n g t o a n y c o m b i n a t i o n , m a c h i n e , o r p r o c e s s i n w h i c h T I p r o d u c t s o r sa r e u s e d . I n f o r m a t i o n p u b l i s h e d b y T I r e g a r d i n g t h i r d - p a r t y p r o d u c t s o r st u t e a l i c e n s e f r o m T I tp r o d u c t s o r s e r v i c e s o r a w a r r a n t y o r e n d o r s e m e n t t h e r e o f . U s e o f s u c h i n f o r m a t i o n m a y r e q u i r e a l i c e n s e f r o m a t h i r d p a r t t h e p a t e n t s o r o t h e r i n t e l l e c t u a l p r o p e r t y o f t h e t h i r d p a r t y , o r a l i c e n s e f r o m T I u n d e r t h e p a t e n t s o r o t h e r i n t e l l e c t u a l p r o p e

R e p r o d u c t i o n o f i n f o r m a t i o n i n T I d a t a b o o k s o r d a t a s h e e t s i s p e r m i s s i b l e o n l y i f r e p r o d u c t i o n i s w i t h o u t a l t e r a t i o n a n d i s a c c o m p a n i e d b y a l l a s s o c i a t e d w a r r a n t i e s , c o n d i t i o n s , l i m i t a t i o n s , a n d n o t i c e s . R e p r o d u c t i o n o f t h i s i n f o r m a t e r a t i o n i s a n u n f a i r a n d d e c e p t i v e b u s i n e s s p r a c t i c e . T I i s n o t r e s p o n s i b l e o r l i a b l e f o r s u c h a l t e r e d d o c u m e n t a t i o n .

R e s a l e o f T I p r o d u c t s o r s e r v i c e s w i t h s t a t e m e n t s d i f f e r e n t f r o m o r b e y o nv o i d s a l l e x p r e s s a n d a n y i m p l i e d w a r r a n t i e s f o r t h e a s s o c i a t e d T I p r o d u cn e s s p r a c t i c e . T I i s n o t r e s p o n s i b l e o r l i a b l e f o r a n y s u c h s t a t e m e n t s .

T I p r o d u c t s a r e n o t a u t h o r i z e d f o r u s e i n s a f e t y - c r i t i c a l a p p l i c a t i o n s ( s u c r e a s o n a b l y b e e x p e c t e d t o c a u s e s e v e r e p e r s o n a l i n j u r y o r d e a t h , u n l e s s s p e c i f i c a l l y g o v e r n i n g s u c h u s e . B u y e r s r e p r e s e n t t h a t t h e y h a v e a l l n e c e s s a r y e x po f t h e i r a p p l i c a t i o n s , a n d a c k n o w l e d g e a n d a g r e e t h a t t h e y a r e s o l e l y r e s p o n s i b l e f o r a l l l e g a l , r e g u l a t o r y a n d s a f e t y - r e l ar e q u i r e m e n t s c o n c e r n i n g t h e i r p r o d u c t s a n d a n y u s e o f T I p r o d u c t s i n s u c h sa p p l i c a t i o n s - r e l a t e d i n f o r m a t i o n o r s u p p o r t t h a t m a y b e p r o v i d e d b y T I . Fy i n d e m n i f y T I a n d i t s r e p r e s e n t a t i v e s a g a i n s t a n y d a m a g e s a r i s i n g o u t o f t h e u s e o f T I p r o d u c t t i c a l a p p l i c a t i o n s .

T I p r o d u c t s a r e n e i t h e r d e s i g n e d n o r i n t e n d e d f o r u s e i n m i l i t a r y / a e r o s p as p e c i f i c a l l y d e s i g n a t e d b y T I a s m i l i t a r y - g r a d e o r " e n h a n c e d p l a s t i c . " O n l y p r o d u c t s d e s i g n a t e d b y T I a s m i l i t a r y - g rs p e c i f i c a t i o n s . B u y e r s a c k n o w l e d g e a n d a g r e e t h a t a n y s u c h u s e o f T I p r o d u c t s w h i c h T Is o l e l y a t t h e B u y e r ' s r i s k , a n d t h a t t h e y a r e s o l e l y r e s p o n s i b l e f o r c o m p l i a n c e w i t h a l l l e g a l a n d r e g u l a t o r y r e q u i r e m e n t s i n c o n n e c t i o n w i t h s u c h u s e .

T I p r o d u c t s a r e n e i t h e r d e s i g n e d n o r i n t e n d e d f o r u s e i n a u t o m o t i v e a p p lf i c T I p r o d u ca r e d e s i g n a t e d b y T I a s c o m p l i a n t w i t h I S O / T S 1 6 9 4 9 r e q u i r e m e n t s . B u y e r s a c k n o w l e d g e a n d a g r e e n o n - d e s i g n a t e d p r o d u c t s i n a u t o m o t i v e a p p l i c a t i o n s , T I w i l l n o t b e r e s p ob l e f o r a n y f a i l u r e t o m e e t s u c h r e

F o l l o w i n g a r e U R L s w h e r e y o u c a n o b t a i n i n f o r m a t i o n o n o t h e r T e x a s I n s t r u m e n t s p r o d u c t so n s o l u t i o n s :

P r o d u c t s A p p l i c a t i o n s

A m p l i f i e r s a m p l i f i e r . t i . c o m A u d i o w w w . t i . c o m / a u d i o

D a t a C o n v e r t e r s d a t a c o n v e r t e r . t i . c o m A u t o m o t i v e w w w . t i . c o m / a u t o m o t i v e

D S P d s p . t i . c o m B r o a d b a n d w w w . t i . c o m / b r o a d b a n d

I n t e r f a c e i n t e r f a c e . t i . c o m D i g i t a l C o n t r o l w w w . t i . c o m / d i g i t a l c o n t r o l

L o g i c l o g i c . t i . c o m M i l i t a r y w w w . t i . c o m / m i l i t a r y

P o w e r M g m t p o w e r . t i . c o m O p t i c a l N e t w o r k i n g w w w . t i . c o m / o p t i c a l n e t w o r k

M i c r o c o n t r o l l e r s m i c r o c o n t r o l l e r . t i . c o m S e c u r i t y w w w . t i . c o m / s e c u r i t y

R F I D w w w . t i - r f i d . c o m T e l e p h o n y w w w . t i . c o m / t e l e p h o n y

L o w P o w e r w w w . t i . c o m / l p w V i d e o & I m a g i n g w w w . t i . c o m / v i d e o W i r e l e s s

W i r e l e s s w w w . t i . c o m / w i r e l e s s

M a i l i n g A d d r e s s : T e x a s I n s t r u m e n t s , P o s t O f f i c e B o x 6 5 5 3 a s , T e x a s 7 5 2 6 5 C o p y r i g h t 2 0 0 7 , T e x a s I n s t r u m e n t s I n c o r p o r a t e d
http://www.ti.com/wirelesshttp://www.ti-rfid.com/http://www.ti.com/telephonyhttp://microcontroller.ti.com/http://power.ti.com/http://www.ti.com/opticalnetworkhttp://logic.ti.com/http://www.ti.com/militaryhttp://interface.ti.com/http://www.ti.com/digitalcontrolhttp://www.ti.com/digitalcontrolhttp://dsp.ti.com/http://dsp.ti.com/http://www.ti.com/broadbandhttp://dataconverter.ti.com/http://www.ti.com/automotivehttp://amplifier.ti.com/http://www.ti.com/audiohttp://www.ti.com/wirelesshttp://www.ti.com/videohttp://www.ti.com/lpwhttp://www.ti.com/telephonyhttp://www.ti-rfid.com/http://www.ti.com/securityhttp://microcontroller.ti.com/http://www.ti.com/opticalnetworkhttp://power.ti.com/http://www.ti.com/militaryhttp://logic.ti.com/http://www.ti.com/digitalcontrolhttp://interface.ti.com/http://www.ti.com/broadbandhttp://dsp.ti.com/http://www.ti.com/automotivehttp://dataconverter.ti.com/http://www.ti.com/audiohttp://amplifier.ti.com/

Date post:	10-Apr-2018
Category:	Documents
Upload:	anovarebooks
View:	218 times
Download:	0 times

Texas Instrument - TMS320TCI6488 W-CDMA DSP SoC

Documents