Coefficient estimation of multicomponent polynomial phase signal (PPS) using MAML-HAF

The problem of coefficient estimation for a multicomponent polynomial phase signal (PPS) embedded in additive white Gaussian noise, is addressed. Through analysis, we find that estimation using the multi-lag high order ambiguity function (ML-HAF) suffers from not only harmonic ambiguity problem, but also the harmonic leakage problem. Based on a new generalized transform, called multiplicative average ML-HAF (MAML-HAF), united solution of the leakage and ambiguity problem is proposed. The theoretical performance of the method is evaluated and verified by simulation results