The Modeling Of Underwater Sound Field And Detecting The Vector Signal Based On EMD

With the deeper development of ocean resources and the growing of civilian and military needs,the traditional signal detection methods have been unable to meet these applications. In order to improve the processing of signals , an improved method based on empirical mode decomposition - ensemble empirical mode decomposition (EEMD) is used for signal processing, and then this method is used to simulate signal detection, the performance of detecting signals based on wavelet packet is compared with it and finally this method is used to process the actual data. The main contents are as follows:Firstly, make models of the sound field which sound waves are spreading in, to get the information of scalar signals and the vector signals. The results of the models shows that the performance of sound spreading in waveguide is better than that in free field.Secondly,the theoretical basis of the wavelet packet decomposition algorithm and its related content has been simulated, and use this algorithm to signal detection, the algorithm then provides a contrast.After that, introduce the theoretical basis of EMD method, the modal aliasing of EMD method has been analyzed, in order to solve the EMD modal aliasing problem, using the ensemble empirical mode decomposition (EEMD). Adding white Gaussian noise to signals and using the EMD characteristic binary filter banks can solve the problem of aliasing modes of EMD. Then use this method to the signal detection, we can get the EEMD energy entropy of signals ,then use a floating threshold for signal detection, simulate results and get the results. The detection method based on the EEMD has better detection performance than the method based on wavelet packet.Finally, this method is used to detect the actual data, using the floating threshold, its detection performance is better than the method based on wavelet packet ,this method of the actual signal processing is effective.