Networked strong tracking filters with noise correlations and bits quantization

Abstract

We study the design of quantized Kalman filters with strong tracking ability for the single sensor system with the correlation between process and measurement noises and adaptive bits quantization in this paper. Firstly, we perfect the problem formulation for the quantized tracking system about the correlation between original process and measurement noises and the correlation matrixes between quantized error and original process and measurement noises. Both are clear innovation in our study. Secondly, based on this problem formulation, two direct quantized Kalman filters are presented by use of statistical modeling and augmented state modeling ways respectively. Finally, the strong tracking method which can deal with noise correlation is used to propose two quantized strong tracking filters, which can effectively reduce the modeling uncertainty and get the strong tracking ability to the state abrupt change. © 2011 Springer-Verlag.