Research Of Adaptive Canceller Based On Reverberation Correlation Between The Beams

Reverberation is generated with acoustic emission signals, and has close relations with characteristics of the emission signals themselves. For active sonar, the reverberation is the main restriction of the active sonar system performance. In this paper, the correlation of different directions of reverberation is studied from the statistical characteristics of reverberation, and adaptive interference cancellation technology is researched. The experimental data is taken into adaptive reverberation cancel processing. Firstly, the experimental data is analyzed in beam forming, then reverberation correlation between beams is calculated, and reverberation correlation between beams impacted by emission signal frequency, emission signal bandwidth and pulse width, seabed and so on are all researched in-depth. Secondly, from reverberation correlation between beams, the experimental data is taken adaptive reverberation cancel processing by the reference signal of different single-beam signals, and the reference signal is set through the result. Then the least-mean-square (LMS) algorithm is studied. Through the rate of convergence and the effect of convergence which is got by using of step factorμin sigmoid function, arc-tangent function, and improved arc-tangent function compared to fixed step size, then the selection of step factorμis determined. Through processing experimental data, comparison of steady state error under different filter order is to determine the filter order N. After setting the various parameters of adaptive reverberation canceller, the experimental data is taken reverberation cancel processing by applying the adaptive reverberation canceller. Experimental data is the lake trial data in two times and sea test data, and the experimental data is taken cancel processing in single-beam signal for reference signal, "parallel" cancel processing in double-beam signal for reference signal, and "series" cancel processing in double-beam signal for reference signal separately. Finally, the results of the experimental data are compared. From three experimental data processing results, it can be seen that after the adaptive reverberation cancel processing, reverberation can be suppressed to some extent, and the purpose of detecting the target signal can be achieved under reverberation background.