Mobility tracking in cellular networks with sequential Monte Carlo filters

This paper considers mobility tracking in wireless communication networks based on received signal strength indicator measurements. Mobility tracking involves on-line estimation of the position and speed of a mobile unit. Mobility tracking is formulated as an estimation problem of a hybrid system consisting of a base state vector and a modal state vector. The command is modeled as a first-order Markov process which can take values from a finite set of acceleration levels, in order to cover the wide range of acceleration changes, a set of acceleration values is pre-determined. Sequential Monte Carlo algorithms-a particle filter (PF) and a Rao-Blackwellised particle filter (RBPF) is proposed and their performance evaluated over a synthetic data example.