Active Sonar Deconvolution Matched Filtering Method

Abstract

The ranging accuracy and range resolution of active sonar mainly depend on the bandwidth of the signal, and the speed accuracy and speed resolution mainly depend on the signal time duration. The wide bandwidth and long duration signals can use matched filtering technique to obtain high resolution. The output signal of the matched filter has a mainlobe and a series of sidelobes on the time axis. Under multi-target environment, we expect that the target signal has good range resolution, and the main-lobe of the weak target echo is not submerged by the sidelobes of the strong interference. Low sidelobe level can be achieved by the joint processing of the matched filter and sidelobe suppressor. The traditional sidelobe suppressor based on time-domain or frequency-domain weighting method often sacrifices the main-lobe width, and degrades the range resolution. In this paper, the matched filter output of the multi-target echo is derived as the convolution operation of the reference signal (the autocorrelation function of the transmit signal) and the objective function (the multi-target unit impulse response function). The Lucy-Richardson iterative algorithm is used to realize the process of deconvolution to solve the objective function. We take the objective function as final output of the deconvolution matched filtering to achieve the high range resolution and low sidelobe level. The performance of active sonar deconvolution matched filtering method is validated by simulation and sea trial data.