Parameter estimation in chaotic noise and its performance simulation

Signal detection in chaotic noise has been discussed in several papers, but the estimation of signal parameter hasn´t been studied deeply. This paper focuses on parameter estimation in chaotic noise. The LS-AR algorithm is used for this purpose. First, we analyzed the proposed algorithm´s principle and its advantage for parameter estimation. Second, simulation experiment results and analysis are given out to show the performance of the proposed algorithm. Lastly, MPSV is introduced to improve the LS-AR in some specific condition and some analysis of the alternative algorithm is also given out. To sum up, the LS-AR is effective for most uses. When it is necessary to use MPSV, one should find good trade-off between computation burden and performance.