Location Tracking Sequential Monte Carlo Method

Wireless localization and tracking is a hot research topic, several theorestic and applied results have been in recent years.This paper addressed two problems:1. How to mitigate position error caused by non-line-of-sight (NLOS) propogation when we use range-based tracking algorithm in Cellular Networks?2. How to use Sequential Monte Carlo method for range-free localization scheme when we estimate the unknown node position in Wireles Sensor Networks?In this paper we focus on literature research about Sequential Monte Carlo method and primarily answer these questions by simulation approach. The main contents are as follows:1. Review the theory and application of Sequential Monte Carlo method, and discusse the filtering error counteraction problem by a nonlinear and non-Gaussian filtering example.2. Present a modified particle filter (PF) algorithm based on the LOS/NLOS binary state information of propagation environment using the numerical method of the probability density function (PDF) about the hybrid noise. The key idea is using numerical method to calculate the hybrid noise PDF, and integrating binary environment information to get filtering result.3. Discuss the Sequential Monte Carlo localization method for wireless sensor networks scheme and modify the basic algorithm to overcome the sample degeneracy problem in resampling stage.