Abstract
The performance of active sonars operating in shallow water is often limited by the reverberation level (Waite, 1998). If a target is moving relative to the reverberating scatterers, it may be possible to isolate its Doppler-scaled echo from the zero-Doppler reverberation provided that the Doppler shift is greater than the bandwidth of the transmitted pulse. In practice, this generally implies the use of narrowband long-duration continuous-wave (CW) pulses. The major disadvantages of this method are the poor range resolution of such pulses leading to poor reverberation processing with low Doppler targets. To detect slowly moving or stationary targets, linear period modulation (LPM) chirps are often used in preference. Several new classes of pulse design have been proposed which theoretically provide superior reverberation processing to CW pulses by virtue of their comb-like spectra. The paper reviews the theory behind Newhall trains, sinusoidal frequency modulated pulses and geometric comb waveforms, comparing their theoretical reverberation processing gains against CW and LPM pulses using the Q-function. Experimental results from low-frequency active sonar sea trials are also presented to verify theoretical predictions.

This publication has 2 references indexed in Scilit: