Research on Recognition Methods of underwater acoustic signal based on higher-order statistics

In acoustic signal recognition problem, different analysis methods can extract different characteristics for the same goal. Good feature fusion methods can take advantage of different traits of each feature, complement each other ,remove redundancy, get more robust new features, and improve algorithm recognition rate. This method can also complete the data compression dimensionality reduction and improve the real-time algorithm. This shows that feature fusion is extraordinary. This paper presents a signal recognition method based on high order statistics and power spectrum estimation and theoretical simulation results. This method can be used to recognize underwater acoustic signals and has a high recognition rate.