Study on reconstruction technology of underwater echo based on compressed sensing

Different from the Nyquist sampling theory, compressed sensing(CS) theory is a technique for information acquisition and processing, and it effectively combines signal sampling with compressing. It allows accurate or high probability reconstruction of signals sampled at rates many times less than the conventional Nyquist frequency the under the condition that signal is sparse or compressible, with measurement matrix projecting the high-dimensional signal into a low-dimensional space. Therefore, it can solve the problems that is a large amount of data caused by traditional sampling. This paper briefly describes the compressed sensing model, give out the underwater echo based on the model of the highlights, and we simulate the echo with the compressed sensing. Simulation results show that, compared to the traditional signal processing methods, compressed sensing can be reconstruct the original signal with data less than 20%. Also it can solve the too high sampling rate, the overload data and other evils, and have higher signal to noise ratio.