Line Spectrum Detection for Sonar Based on Time Reversal Convolution and Interference Suppression

摘要

This paper proposes a time reversal convolution and interference suppression (TRCIS) method, and deduces the principle of TRCIS for detecting CW signal in broadband noise. From the simulation and experiments, conclusions can be drawn such as the signal-to-noise ratio (SNR) gain of TRCIS increases linearly with the input SNR; the detection threshold is about 2 dB lower than that of the conventional line detection methods. Then, by using TRCIS as a preprocessing technique of adaptive line enhancer (ALE), the time reversal convolution and interference suppression based adaptive line enhancer (TRCIS-ALE) is proposed. Through the simulation comparison with conventional line detector, ALE and coherent accumulation based on adaptive line enhancer, the conclusion can be obtained that the SNR gain, detection threshold and ROC curve of TRCIS-ALE are all significantly better than the other three methods.