Improved Whitening Method of Linear AR Colored Noise in Stable Distribution Environments

The whitening problem of colored noise had been resolved for Gaussian white noise through linear system, but other inputs through other systems had not been resolved which includes non-Gaussian white noise through linear or nonlinear system. This paper briefly proposed an improved iteratively reweighted least P norm (IIRLP) algorithm for whitening in alpha-stable environments. And the parameters in linear model of stable colored noise were estimated, and the whitening method was explored. Simulation and analysis show that the method is more robust and solid than the methods before.