hjhghjj

8/10/2019 hjhghjj

1/12

International Journal of Software Engineering and Its Applications

Vol.8, No.1 (2014), pp.235-246

http://dx.doi.org/10.14257/ijseia.2014.8.1.21

ISSN: 1738-9984 IJSEIA

Copyright2014 SERSC

Indoor Wireless Localization Using Kalman Filtering in

Fingerprinting-based Location Estimation System

Geon-Yeong Park, Min-Ho Jeon and Chang-Heon Oh

School of Electrical, Electronics, and Communication EngineeringKorea University of Technology and Education

[email protected],[email protected] and [email protected]

Abstract

As smartdevices such as smart phone and smart TVs become widely distributed, variousstudies on location-based services have been conducted. Such location-based services areuseless, however, unless the users location is known. A number of researchers haveexamined methods to trace and determine indoor locations for indoor location-based

services. In particular, WALN has been examined in various studies because of its advantage

to use a frequency band available without advanced settings. This study suggests a newindoor tracing method to reduce time delays upon location fingerprinting for point datacollection, which is a disadvantage of the existing Kalman filtering algorithm and

fingerprinting type location tracing algorithm. This study also compares its performance withthat of existing methods based on the collected data. As a result of the experiment, the fast

collection algorithm is presented as a solution to the problems of existing methods. It isproven that the fast collection algorithm presented in this study is applicable to a locationtracing system in an actual environment.

Keywords: fingerprinting, location estimation, WLAN, fast collection, NLOS

1. Introduction

In recognition of the usefulness of Location-based Service (LSB), large-sized companies,

plants, and colleges are demanding services for its utilization [1]. Since Global PositioningSystem (GPS) itself cannot provide sufficient data to determine the users location when he isin a building, indoor location tracking to determine the users location in a building,

particularly when utilizing wireless local area network (WLAN), has been actively studied[2]. This is because WLAN is being used in every place from college campuses to airports,hotels, large-sized companies, and even private homes. This study introduces a WLAN-based

indoor location estimation system. This determines the users location in reference to receivedsignal strength indication (RSSI) collected from access points (AP) [3-7]. Fingerprinting isthe most common method used to determine the users location because its method forlocation estimation is simple compared to angle of arrival (AOA), time difference of arrival

(TDAO), and time of arrival (TOA). Fingerprinting is used to link location-relatedcharacteristics such as the RSSI received from a number of APs to the location rather thanrelying on the accurate estimation of the angle or distance and to infer the location of such

characteristics. In this case, the wireless network interface card (NIC) and existing WLANinfrastructure could be readily reused with no need for special hardware at a mobile station(MS).

Fingerprinting consists of the offline step and real-time step. The offline step collects thedata used in determining location by using the fingerprinting method. In this step, the location
mailto:@koreatech.ac.krmailto:@koreatech.ac.kr

8/10/2019 hjhghjj

2/12


Vol.8, No.1 (2014)

236 Copyright2014 SERSC

fingerprint or radio map database is established [8]. Setting up the database involves themapping of each location point and location fingerprint data (RSSI). One of the disadvantagesof the offline step in fingerprinting is that it takes long to collect data and the values change

over time. Hence, this problem needs to be solved first of all. In this study, a location

fingerprint consists of vectors of various APs RSSI values at a certain location of

( ). The data extracted in real time to solve the problem above is used as thelocation fingerprint data. The real time step of fingerprinting estimates the location based onthe location information of the database collected in the offline step. A mobile node (MN)

acquires location information . The Hamming signal distance between

the and for each in the database is computed. The information collected in the realtime step and the route of the users movement are used to update location fingerprinting datain consideration of the surroundings.

Based on the points above, this study suggests a method to collect offline data effectively

in a fingerprinting-based indoor location estimation system. Chapter 2 points out theproblems of the existing methods and explains the fast collection algorithm in the offline stepto solve such problems. Chapter 3 explains the fast collection algorithm in the real-time step.Chapter 4 evaluates the RSSI collection and transfers software for the location fingerprint

database, experiment environment, and algorithm suggested in this study. Chapter 5 presentsthe conclusion.

2. Algorithm of Fast Collection for Offline Step Data

The fingerprinting-based indoor location estimation system estimates indoor locationin reference to the RSSI of WLAN. Hence, it involves less cost for the additionalsystem compared to other indoor location estimation systems. Since the collected datain the offline step and period for the establishment of the fingerprinting offline step

may be changed over time, there must be a system to address this problem. This studysuggests the fast collection (FC) algorithm to reduce the time of collecting locationfingerprints in the offline step in the indoor location estimation system and to

complement location fingerprints that might be changed over time.Location fingerprint data collected in a static status may be different at each time of

measurement due to the dense multiple channels of indoor radio waves and radio effects

such as reflection, diffraction, and dispersion. Multipath fading changes the receivedsignals around the medium values at a certain location. As such, received signals are

presented and they are influenced by the combined effects of large-scale fading and

small-scale fading [9].Kamol Kaemarungsi has studied RSSI characteristics such as the effect of a users

body, effect of users orientation, distribution of received signal strength, standard

deviation of the RSS, stationary condition of the RSS, and so forth [10]. When it comesto the RSSI characteristics that he studied, special attention must be paid to their effect

to the users body, the effect of the users orientation, the time dependency of received

signal strength, the interference from multiple APs, and the independence of multipleRSSs.

As to RSSI characteristics presented by Kamol Kaemarungsi, a location fingerprint

database needs to be established in consideration of the route of the users movementand number of surrounding APs to form the fingerprinting type location estimationsystem. When a location fingerprint database must be established in the offline step

based on the points above, it will take more time depending on the size of the building.On the other hand, the collection of location fingerprint data would be quite efficient.

8/10/2019 hjhghjj

3/12


Vol.8, No.1 (2014)

Copyright2014 SERSC 237

When the information of a moving object is collected and the database is establishedafter all possible routes of the users movement are calculated, the time of establishingthe location fingerprint database will be shortened. In addition, using the location

information collected every hour will further shorten the offline step, with the errors inthe real-time step reduced. This theory will be proven in the following section before

the fast collection algorithm is presented.

Figure 1. A Comparison of AP RSSI Depending on the Location

Figure 1 shows the RSSI values of APs collected during the movement. As shown inthis figure, the object moves at similar rates in the same place, and the measured RSSIvalues are in a similar scale. The RSSI values of APs collected in a static status,

however, are significantly different.To reduce the difference in the fingerprint method, it is important to collect as many

vector Rs as possible. Location vector division provides vector R, which forms -

in Table 2 of the movement coordinates, with a number for . For such location vector

division, as to collected initially in Expression 1 and Vector R of collected

thereafter, all the variables of Vector R included in are subtracted by vector R ofto calculate the absolute value.

(1)

When the value of calculated in Expression 1 is , there is no need for AP-

centric distribution; thus, a new Vector is generated as many as . Afterwards, k,

8/10/2019 hjhghjj

4/12


Vol.8, No.1 (2014)


which is the value of is added to the generated

Vector .

(2)

Table 1 shows the data after the location vector division of the - section in

Table 2 in application of Expressions 1 and 2.

Table 1. Location Vector Division Data of - Section

Vector (location) AP 1 AP2 AP3 AP4

(0m) , reference point -47 -45 -33 -53

(1m) -47.4 -44 -33.8 -51.8

(2m) -47.8 -43 -34.6 -50.6

(3m) -48.2 -42 -35.4 -49.4

(4m) -48.6 -41 -36.2 -48.2

(5m) -49 -40 -37 -47

(6m) -49.4 -39 -37.8 -45.8

(7m) -49.8 -38 -38.6 -44.6(8m) -50.2 -37 -39.4 -43.4

(9m) -50.6 -36 -40.2 -42.2

(10m) -51 -35 -41 -41

AP1 and AP3 are of almost a straight line, which indicates that when , the

value added to the variable of Vector k is low. Two interpretations are

possible: first , AP may exist in the range of - . In this case, , and this value

gets bigger as AP goes out of the center. Hence, when , the use of AP-centric

distribution increases the value of , and the added variable increases accordingly.

In the second case, when the AP from which the value of RSSI is collected is far, the

received signals may go through interference. The added variable is . Tosolve this problem, the fast collection algorithm suggested in this study calculates thevector in the way of location vector division, inserts it, and determines its location in

reference to Hamming distance in the section of - .Table 2 shows the RSSI data right under AP. This indicates that RSSI values may be

changed depending on the direction even right under AP. As for the AP-centricdistribution of the fast collection algorithm, the difference in location vector division is

complemented by the data above when . AP-centric distribution collects the datafrom four different directions with AP as the center as shown in Figure 3 . The directionis designated to a certain value. Once the direction is designated, the value of AP-

centric distribution is inserted between and , and then the value of is

generated through location vector division.

8/10/2019 hjhghjj

5/12


Vol.8, No.1 (2014)


Figure 2. Location Vector Division Data of - Section

Table 2. RSSI Value of the Movement Route Measured Right Under AP

Direction RSSI

Front -35Right -29

Back -41

-23

Figure 3. A Comparison of the RSSI of each AP on Li Depending on theMeasuring Methods

3. Algorithm of Fast Collection for Real-time Step Data

As stated earlier, the environmental change over time also needs to be considered.Hence, the way to complement it in the real-time step also will be discussed. Tocomplement the disadvantage of data change over time, the location fingerprint, which

is used in location estimation in the real-time step, is revised by using Expression 3.

(3)

8/10/2019 hjhghjj

6/12


Vol.8, No.1 (2014)


in Expression 1 is the variable of Vector R, and is the variable of Vector R

to which will be compared. The RSSI value for a certain AP among those that werecollected for 24 hours at one place may change from a minimum of -29dBm to amaximum of -33dBm in an adjacent area. It may change at every second irregularly.When Expression 1 is substituted for the changing amount, however, it is converged to

one variable at one time zone as in Figure 1. The distribution traces form a lineal curveover time. The lineal distribution may be described with the example in Figure 4.Although the change is irregular from 09:00 to 08:00 on the next day, the variable ofthe collected Vector R changes in a lineal form from -31dbm (09:00) to -33dBm (10:00)when it is observed each hour. When the changing data collected in the offline step isadjusted according to the timing, the difference of the fingerprinting type estimation

system can be reduced.

Figure 4. RSSI Information of an AP

4. Performance Analysis

4.1. Development of the RSSI Collection and Transfer Software

WLAN-based indoor location estimation methods are economical compared to otherindoor location estimation methods, and a number of studies have been conducted onthis area. WLAN-based indoor location estimation methods cannot be realized unless

RSSI from APs is read. Hence, this research has developed an RSSI collection andtransfer that can be used to measure RSSI.

The computer application is developed by using C# in application of a native Wi-FiAPI provided by Microsoft Developers Network (MSDN). As shown in Figure 5,

information on SSID, link quality, RSSI, frequency, MAC address, and beacon periodat a certain place is collected and transmitted to the server. One outstanding feature isthe beacon message must be transmitted after the program is initialized when AP

information is collected and then recollected. Otherwise, the information collectedearlier remains as it is. Since information on APs may be different depending on thetypes of NIC, it is vital to collect various types of NIC and information among common

APs so that it can be applied to various devices.

8/10/2019 hjhghjj

7/12


Vol.8, No.1 (2014)


Figure 5. RSSI Reader

4.2. Measurement Setup

A laptop computer equipped with a WLAN card of Ralink technology as well as theRSSI collection and transfer software developed to collect offline data and estimatelocation was used to collect the RSSI information of AP within the 1st engineering

building in the Korea University of Technology and Education. It is a four-storybuilding in which the 1st to 3rd floors have 18 common APs for each and the 4th floorhas 15. On each floor, there are 16 to 28 personal APs, and the data collected in the

offline step include about 38-50 AP signals at one place.

Figure 6. The 4th Floor of the 1st Engineering Building with the Locationsof Published APs

When all of the AP information is collected to establish a location fingerprint

database, it takes about 10 seconds per step to collect, transmit, and save APinformation. Hence, information on APs with the highest strength of RSSI and commonAPs installed at the Korea University of Technology and Education is used. This is

because common APs are arranged at the same intervals in a row as shown in Figure 6;thus, there are duplicated vector data even if only three data sets are used. Further,when signals from an adjacent corridor enter a room in the building, the RSSI signals

received become very weak. Based on the measuring method suggested by Kamol

8/10/2019 hjhghjj

8/12


Vol.8, No.1 (2014)


Kaemarungsi, the signals received by the WLAN card are limited from -93dBm to0dBm.

4.3. Experimental Methods

As shown in Figure 7, after the table to predict the route of movement in each

direction and the fingerprint location database in the offline are prepared, the userslocation information is received and the actual location is compared with the estimatedlocation of the estimation program in application of the fast collection algorithm. Toevaluate the fast collection algorithm suggested in this study , the time of collecting data

in the existing fingerprinting type offline step and the time of the algorithm suggestedin this study are compared.

Figure 7. Routes within the Building

4.4. Experimental Results

Figure 8 shows the time of collecting AP information when four APs are selected atone place and the fast collection algorithm of fingerprinting is applied in comparisonwith existing fingerprinting methods. In the case of existing fingerprinting methods, the

information is collected at intervals of 1m while in the case of the suggestedfingerprinting method the information is collected at intervals of 10 seconds. Foraccurate verification, the object moves 1m per second or 10m per 10 seconds. Theexperimental result indicates that the suggested fingerprinting method involves increasein data collecting time at consistent rates because of continuous movement whileexisting fingerprinting methods involve an increase at inconsistent rates due to such

variables as collected data, transmission initialization, distance calculation, etc. As thetime of measuring increases, the difference increases up to 10 times. As the range ofmeasurement becomes larger, the measuring time increases accordingly.

8/10/2019 hjhghjj

9/12


Vol.8, No.1 (2014)


Figure 8. A Comparison of the Data Collecting Time in the Offline Step

Figure 9 shows the range of signal errors at the moment of collecting RSSI from a

moving object in application of the fingerprinting method without AP-centricdistribution as a fast collection algorithm and an existing fingerprinting method. Sincethe existing fingerprinting method does not consider variables of mobility, the range ofsignal errors of the fast collection algorithm in consideration of mobility is low. SinceAP-centric distribution is not considered, however, the range of errors is relatively

similar.Figure 10 shows the range of signal errors at the moment of collecting the RSSI of a

moving object in application of fingerprinting with the AP-centric distribution applied

and an existing fingerprinting method. In comparison with Figure 9, the fingerprintingmethod with the AP-centric distribution applied involves a narrower range of errorsthan the existing fingerprinting method.

Figure 9. RSSI Errors when the AP-centric Distribution is Applied

8/10/2019 hjhghjj

10/12


Vol.8, No.1 (2014)


Figure 10. RSSI Errors when the AP-centric Distribution is Applied

Figure 11 shows the range of errors in comparison of or RSSI at the measuringlocation and Vector R in the location fingerprint database in application of thesuggested fingerprinting method. To verify the excellence of the suggested algorithm,the values of Vector R collected by means of an existing fingerprinting method arecompared to determine the range of errors, and then the results are illustrated. It turns

out that the performance of the suggested method is better than that of existingfingerprinting methods in terms of tracing a moving object.

Figure 11. The Difference from the Actual Distance

5. Conclusion

To highlight the necessity of the fast collection algorithm suggested in this study, theproblems of existing fingerprinting methods were addressed, and the data of indoorenvironments at various angles were collected and comparatively analyzed to verify theexcellence of the fast collection algorithm. The experimental result indicated that whilethe existing indoor location estimation systems in application of the fingerprinting

8/10/2019 hjhghjj

11/12


Vol.8, No.1 (2014)


method could be used to find an object in a fixed location, they did not prove to beuseful for tracing a moving object. In addition, it has been proven that the datacollecting time could be minimized in the offline step where the fingerprinting method

was applied to trace the location. This was because the performance of the existingfingerprinting method in collecting data and tracing location was inferior or the

experiment was conducted in a fixed condition. Hence, the fingerprinting method inapplication of the fast collection algorithm as suggested in this study to estimate thelocation of a moving object by means of actual wireless moving nodes proved to bemore efficient and accurate for moving objects. Currently, the excellence of the

suggested algorithm was evaluated based on the valid data of a lineal distance. Thestudy should be developed further in application of this algorithm and in reference tovalid data regarding mobility at various locations.

References

[1] K. Virrantaus, J. Markkula, A. Garmash and Y. V. Terziyan, Developing GIS-Supported location based

services, Proceedings of the Second International Conference on Web Information System Engineering,Kyoto, Japan, (2001)December 3-6.

[2] R. Want, A. Hopper, V. Falcao and J. Gibbons, International Journal of ACM Transactions on informationsystems, vol. 10, no. 1, (1992), pp. 91.

[3] D. Madigan, E. Einahrawy, R. P. Martin, W. Ju, P. Krishnan and A. S. Krishnakumar, Bayesian indoorpositioning system, Proceedings of 24th Annual Joint Conference of the IEEE Computer and

Communications Societies, Miami, USA, (2005)March 13-17.[4] C. Wann and M. Lin, Data fusion methods for accuracy improvement in wireless location system,

Proceedings of Wireless Communications and Networking Conference, LA, USA, (2004)March 21-25.[5] J. Yim, International Journal of Expert Systems with Applications, vol. 34, no. 2, (2008), pp. 1296.[6] H. Seo and H. Kim, Journal of information and communication convergence engineering, vol. 10, no. 4,

(2012), pp. 349.

[7] C.-H. Oh, Journal of information and communication convergence engineering, vol. 10, no. 3, (2012), pp.269.

[8] M. A. Youssef, A. Agrawala and A. U. hankar, WLAN Location Determination via Clustering andProbability Distributions,Proceeding of the First IEEE International Conference on Pervasice Computingand Communications, Texas, USA, (2003)March 26-26.

[9]

B. Sklar, International Journal of IEEE Communications Magazine, vol. 35, no. 7, (1997), pp. 90.[10] K. Kaemarungsi and P. Krishnamurthy, Properties of Indoor Received Signal Strength for WLAN LocationFingerprinting, Proceeding of the First Annual International Conference on Mobile and UbiquitousSystems: Networking and Services, Boston, USA, (2004)August 22-26.

Authors

Geon-Yeong Park, he received a B.S degree in the Department of

electrical engineering from Hnabat Nation university, Deaheon,Korea, in 2008. He is currently pursuing a M.S. degree in Electrical,Electronics and Communication Engineering at the Korea Universityof Technology and Education, Cheonan, Korea. His researchinterests are in the areas of wireless communications, wireless

sensor network and real-time operating system.

8/10/2019 hjhghjj

12/12


Vol.8, No.1 (2014)


Min-Ho Jeon, he received the B.S degree in the Department ofGame Digital Content from Far East University, Umsung, Korea, in2009, and an M.S. degree in Electrical, Electronics and

Communication Engineering at the Korea University of Technologyand Education, Cheonan, Korea, in 2011. He is currently pursuing a

Ph.D. degree in Electrical, Electronics and CommunicationEngineering at the Korea University of Technology and Education,Cheonan, Korea. His research interests are in the context-aware,wireless sensor network, wireless localization, channel coding and

M2M network.

Chang-Heon Oh, he received the B. S. and M.S.E. degrees intelecommunication and information engineering from KoreaAerospace Univ. in 1988 and 1990, respectively. He received thePh.D. degree in avionics engineering from Korea Aerospace Univ.,in 1996. From Feb. 1990 to Aug. 1993, he was with HanjinElectronics Co. From Oct. 1993 to Feb. 1999, he was with the

CDMA R&D center of Samsung Electronics Co. Since Mar. 1999, hehas been with the School of Electrical, Electronics and CommunicationEngineering, Korea University of Technology and Education, where he iscurrently a professor. His research interests are in the areas of wireless

communications, mobile communication, and wireless sensor networkswith particular emphasis on wireless localization.

Date post:	02-Jun-2018
Category:	Documents
Upload:	luong-dinh-thap
View:	218 times
Download:	0 times

hjhghjj

Documents