Diffusion Bias-Compensated RLS Estimation in Noisy Autoregressive Process Over Adaptive Networks

摘要

This paper studies the problem of distributed collaborative parameter estimation for autoregressive (AR) process in the presence of white observation noise. Distributed standard recursive least squares (RLS) algorithms are biased when considering the white noise in measurements. We adopt a bias-compensated method and develop a diffusion bias-compensated RLS (diff BC-RLS) algorithm that can obtain a consistent estimate of the unknown parameter in noisy AR models over the network. Simulation results are given to illustrate the good performance of the proposed algorithm.