The Study Of Particle Filter And Its Application In Cellular Wireless Location

In the past, many of the problems that have been studied were in Gauss environment. With the depth of knowing the nature, it is believed that the linear gauss problems are very few. To most of the non-linear and non-Gauss problems, the solutions were linearization using some methods, such as Taylor series expansion method, using only the first order derivative and extended Kalman method, using Jaccobi matrix. To the issue with one dimension variance and unapparent linearity, these solutions are efficient. When the deminensions of state variance and the nonlinearity increasing, the others are needed.Particle filter is a kind of approximately solutions of Bayes estimate based on sample theory. It is compounded by Monte Carlo Methods and Bayes Theory. The based manner is trying to find out a series of random samples from the state space and to approximate the posterior probability p ( x0 :n | y1 :n). The mean of samples takes the place of [ ]E g ( x0 :n ) |y1 :n, and the minimum variance estimate is obtained. The key of this method is finding the random samples according to p ( x0 :n | y1 :n). These samples are called particles vividly. As far as the meanings, Particle Filter Method is one of approximate Bayes estimate methods using adaptive method of lattice. The technology is adapted to any non-linear and non-Gauss systems that can be indicated by dynamic space model and traditional Kalman filter. The precision is close to the best.The main steps of particle filter are system initialization, sampling from the important particles, threshold decision, and resample. The system initialization aims to start the computation by setting an initialization. As far as the samples from important ones, the prior probability function is often used as the important densityfunction to get the needed ones. And then the weights are got. For improving the algorithm efficiency and the sample representation, the samples with bigger weight is reserved. The particle filter resamples from the bigger weight samples discretely, and uses these to approximate the posterior probability of state vectors. While, the regularization particle filter resamples from the samples continuously, and gets the posterior probability from them. This way conquers the degeneration of sample.Based on the characters of the non-linear and non-Gauss system, this paper talks about the TDOA location utilized the particle filter. The frequent methods were Taylor series expansion and extended Kalman filter. The same character of these methods is that they process the model after linearization of the non-linear. Especially for the system of poly-dimensional state variances, the efficiency is depressed. The particle filter is a method of statistic approximation. It does not completely extract the model. It avoids solving the complex non-linear system. So we can say that this method can get the estimation of the poly-dimensional state variances in TDOA model. The results of emulation prove it. In the emulation, some noises are Gauss and the others are not. In this way, the validity of particle filter is proved.Though the PF can estimate the system of non-linear and non-Gauss well, the shortages are the account is increasing according to the dimensions and the error rate is infected by the start location and the important density function. The large account will be solved along with the development of computer. The affect of the start location and the important density function need the future study.