Comparison of comb and pulse train signal designs for active sonar

Abstract

Doppler-sensitive waveforms are often used in shallow water to mitigate high reverberation levels. This paper compares three Doppler sensitive sonar waveforms including coherently-processed pulse trains, uniform frequency combs, and nonuniform co-prime frequency combs in terms of their delay-Doppler responses. It is shown that the use of comb waveforms can reduce Doppler sidelobes without trading Doppler resolution, unlike windowing. The pulse train and uniform comb are evaluated in terms of their ambiguity functions, while the co-prime comb was processed using a technique that exploits the co-prime structure to reduce the number of tones necessary to achieve a delay-Doppler response similar to that of a uniform comb. The cost of using a co-prime comb rather than a uniform comb is increased bandwidth extent. Brief discussion of the peak-to-average power ratio (PAPR) of these signals is also included.