Direct-path Delay Estimation under Closely-spaced Multipath Interference

Abstract

For direct sequence spread spectrum (DSSS) signal, the minimum mean square error (MMSE) algorithms based on reiterative inverse filters have good delay resolution under closely-spaced multipath interference. However, the limited density of the correlator function dictionary leads to a model error of the filter. Under the inexact model, the performance of delay estimation decreases. To address this problem, we add a group of correction parameters into the dictionary matrix and estimate them in the reiterative filtering. Based on the sparsity of correction parameters, a threshold decision is adopted to sift the parameters that need to be estimated by maximum likelihood (ML) search. The parameters that fail to pass the threshold are set to zero. Simulation results show that, compared with least squares (LS) and MMSE algorithms, the proposed algorithm can improve the performance of delay estimation.