Abstract:
Traditional techniques of signal processing in both the time domain and frequency domain are often not sufficient enough to characterize the complex dynamics of underwater sources that generated, radiate, and/or reflect sonar signals. The purpose of this paper is to analyze simulated signals by means of a nonlinear technique (based on mutual information), which is a completely new method that represents a revolutionary advance. Mutual information is a measure of the amount of information that one random variable contains about another random variable. It is also a reduction of uncertainty of one random variable, which is due to the knowledge of the other. Here the results of simulated sonar signal processing are presented based on mutual information and which is then compared to the traditional correlation technique. The results indicate that the mutual information is a more powerful tool compared to the correlation technique, and the results based on mutual information provide clues leading to source identification.