Sequential Monte Carlo Methods With Applications To Communications

Recently, withthe rapid improvement of performance of digital processor,sequential monte carlo (SMC) method has a wide range of application in engineering,especially in signal processing, statistics, and econometrics etc. The time varyingsystems can be stated in the form of a dynamic state space model. For linear models andGaussian noise, the Kalman filter provides analytical expressions for posterior filtering.However, for non-linear models and non-Gaussian noise, such closed form expressionsare almost impossible to obtain, and sequtial monte carlo method provides itsapproximation. The basic idea of this method is to produce particles from the posteriordensities, and these weighted samples provide approximations to the densities. In this dissertation, sequential monte carlo method and its applications incommunication are investigated. Firstly, its basic idea, method and improved methodare introduced, then induce the bound of particles, which determines the performance ofthe system, and the more particles are chosen, the higher computation is done, finally, anew detector based on new importance function is proposed, which deals with the jointchannel estimation and detection in flat fading channels. Simulations show that theproposed detector is effective.