Time-Delay Estimation Based on Fractional Lower Order Statistics

Abstract

According to the algorithms based on Gaussian assumption are not accurate due to the presence of non-Gaussian signals and noise in the real environment. In this paper, a non-Gaussian noise model based on alpha stable distribution is set, and a time-delay estimation algorithm based on fractional lower order statistics is studied, where fractional lower order covariance matrix makes the time-delay estimation algorithm more accurate than traditional correlational time-delay estimation algorithm. Moreover, this paper analyzes the impact of covariance matrix parameters on the performance of the time-delay estimation algorithm based on fractional lower order statistics and compares the performance of this algorithm with correlational method. Theoretical analysis and simulation results show that this algorithm based on fractional lower order statistics has advantages of anti-jamming.