A STUDY ON RECEIVED SIGNAL STRENGTH BASED INDOOR LOCALIZATION

Abstract

WIRELESS LOCALIZATION TECHNOLOGIES HAVE BECOME VERY POPULAR WITH THE INCREASING DEMAND IN VARIOUS LOCATION-AWARE SERVICES. RECEIVED SIGNAL STRENGTH (RSS) BASED LOCALIZATION TECHNIQUES HAVE BEEN EXTENSIVELY EXPLORED DUE TO ITS LOW IMPLEMENTATION COST AND ROBUSTNESS UNDER NON-LINE-OF-SIGHT SITUATIONS. IN THIS THESIS, WE HAVE DIVIDED THE CONTRIBUTION INTO THREE PARTS BASED ON THE KEY PERFORMANCE METRICS IN RSS-BASED LOCALIZATION, I.E. ACCURACY IMPROVEMENT, TRAINING COST ELIMINATION, AND ROBUSTNESS ENHANCEMENT. FIRST, WE HAVE DEVELOPED TWO NEW APPROACHES TO ENHANCE THE ACCURACY, I.E. FUSION APPROACH AND STATISTICAL APPROACH. SECOND, WE HAVE DERIVED AN UNSUPERVISED TRAINING ALGORITHM TO REDUCE THE TRAINING COST IN RSS FINGERPRINT APPROACH. LAST, WE HAVE PROPOSED A POWER-GAP ELIMINATION (PE) ALGORITHM TO MITIGATE THE PERFORMANCE DEGRADATION BROUGHT BY THE INCONSISTENCIES OF MOBILE DEVICES. MOREOVER, WE HAVE FURTHER EXTENDED THE PE ALGORITHM TO MITIGATE THE DETERMINISTIC RSS DEVIATION.