Abstract: |
Aiming at the failure of the adaptive line spectrum enhancer under low signal to noise ratio, a weighted zero attraction frequency domain block minimum average p norm adaptive ship radiated noise line spectrum enhancement algorithm is proposed by analyzing the properties of the line spectrum in the time and frequency domains. Based on the periodic correlation of line spectra in the time domain, a block algorithm is introduced to superimpose signal gains at different time periods; Introduce the minimum mean p norm and zero attraction term based on the sparsity feature of line spectra in the frequency domain; Simulation data shows that at a low signal to noise ratio of -30 dB, the conventional adaptive filter still has a gain of about 10 dB even when it fails, and exhibits better equalization and clutter suppression effects; In actual sea trial data, good results have also been achieved, proving the reliability and accuracy of the algorithm. |