Research On Wireless Localization Technology Of Indoor Moving Target

Moving target positioning technology has been widely used in the area of therobot technology, pervasive computing, intellisense, safety monitoring and medicalimage processing and many other fields. It has a broad application prospect and playsa very important role in improving people’s living standards and social security. Thepresence of NLOS and multipath propagation presents a significant challenge in termsof localization. Therefore, experts and scholars have putted forward different methodsto improve the accuracy of target tracking, for example Bayesian theory is been usedin moving target location area to improve the precision of localization. Particle filteris effective in dealing with nonlinear non-Gaussian problem. But standard particlefilter heavily relies on model. Though resampling algorithm effectively solvesparticles degradation problem, it inevitably brings particles exhausted problem.In this thesis, an improved particle filter is being proposed to solve the problemsof standard particle filter based on the research of the moving target locationtechnology. The method introduces the current measured value in the process ofparticles prediction and resampling, so that we can address the problem brought byinaccuracy prediction model, at the same time the method proposed in this thesis canguarantee the diversity of particles to some extent.In addition, this thesis adopts the nanoLOC development kit to conduct rangingexperiments and use fitting method to get fitting range value. Then we use thesevalues to verify the performance of improved particle filter, standard particle filter andleast square method. By comparing the performance of three methods, we can draw aconclusion that using the improved particle filter can achieve higher localizationaccuracy than other two methods.