Study On Indoor Location Estimation Technology In WLAN

In the latest semicentennial，with the rapid development of wireless communicationtechnology，the location-based service has become increasingly important, and there aregrowing demand in timeliness and accuracy，especially in military，medical and otheremergency scenarios.Therefore，various location-based services have mushroomed.GPSand cellular network based positioning systems are two common localization system.GPSwith at least four satellites can achieve global location positioning in unobstructedenvironment.For strong sheltered indoor environment，GPS become useless.The cellularnetwork based positioning system make the use of the time information of the base stationto obtain the location of the mobile device.Generally synchronization between basestations is very poor，thus such kind of localization system will achieve very pooraccuracy.In recent years，with the rapid development of WLAN technology，WLAN deviceshave been widely deployment in densely populated areas，and WLAN devices is cheapand easy to deploy，so a lot of WLAN-based localization techniques have been developed.Though many achievements have gained，there are some drawbacks those technique hademerged，such as poor accuracy and heavy labor cost，especially in large scaleenvironment,workload will be significantly increased as the scene size increases. Maincontent of this paper is to discuss the problems emerged in large scale environment, maincontent is as following:This paper first gives a brief introduction to the research background and summed upthe inadequacies of some typical positioning technology and some of the advantages ofWLAN-based positioning system. Then introduce two kinds of topologies in WLAN andchannel characteristics in WLAN. A detailed and comprehensive introduction on theprincipals and positioning technology of WLAN positioning are also provided and thenleads to the discussing on fingerprinting based technology in large scale environment.Workload and fingerprinting matching number will increase significantly as thepositioning scene becomes bigger, response time will also increase as a result. In order tosettle the second problem, here we propose a new fingerprint grouping algorithm, and givea new definition of the fingerprint, and make improvements to the traditional locationalgorithm. The simulation and experimental results show that, compared to other groupingalgorithm, the proposed grouping algorithm and positioning algorithm can not only effectively reduces comparisons of fingerprints, but also enhances the positioningaccuracy and hit rate.