Research On Ranging Localization And Error Mitigation Technology In UWB Systems

Indoor location awareness is drawing great attention recent years due to the development of location based services(LBS), wireless sensor networks(WSN) and a rising demand for indoor navigation. For ranging based localization, a precondition for precise localization is the high ranging accuracy. High-accurate indoor ranging can be enabled through ultra-wide band(UWB) transmission, due to its large bandwidth, good timedelay resolution and obstacle-penetration capabilities. The major challenge in UWB ranging systems is the harsh indoor environment with dense multipath and non-line-of-sight(NLOS) conditions, which affect the ranging performance as well as localization accuracy greatly.The key of UWB ranging is to detect the first path(FP) and determine its arrival time. To estimate the time of arrival(TOA) of the first path in UWB ranging systems and reduce the impact of the dense multipath and NLOS conditions of indoor environment, first, novel information entropy based TOA estimation is proposed. The entropy is used to measure the randomness of the received signals and the FP can be determined by choosing the sample which is followed by drastic entropy decreasing. The entropy based method provides great improvements compared to traditional threshold crossing, coherent detection, etc. Then, support vector machine(SVM) regression based ranging error mitigation is proposed in the research. The SVM regression is employed to mitigate the ranging errors directly by the features extracted from the received signals. Finally, LSE(Least Squared Error)-Taylor based localization algorithm is introduced in the research.The result of LSE is used as initial value of Taylor series expansion, which solve the initial value selecting problem as well as a better localization result can be obtained by a few iterations compared to conventional LSE method.The simulation results show that the proposed approaches provide significant improvements in channel CM1 to CM4 of IEEE 802.15.4a standard compared to conventional approaches. The ranging and localization error under most conditions is several centimeters size, which can meet the requirements of practical applications.