Sparsity-based frequency-domain adaptive line enhancer

The radiated lines from underwater targets are an important feature for passive sonar detection. Adaptive line enhancer (ALE) is usually applied as a preprocessing step to enhance the signal-to-noise ratio (SNR) of the lines. However, the conventional ALE based on least-mean-square (LMS) algorithm suffers from the weight noise in the adaption, which limits the SNR gain severely. Inspired by the frequency-domain sparsity of the lines, a sparsity-based ALE is developed to break through this limit. The proposed ALE is implemented in the frequency domain and a sparse penalty is incorporated into the frequency-domain adaption. By means of the sparse penalty, the weight noise is suppressed and the SNR gain is well improved. Simulation results demonstrate that the SNR gain of the proposed ALE is 9 dB higher than that of the conventional ALE. Experimental data processing also verifies the superiority of the proposed ALE.