Abstract: A nonlinear signal processing system and method is used to identify nonlinearity (e.g., chaos) in underwater acoustic signals, such as sonar signals. The system and method detects the underwater acoustic signal and digitizes the underwater acoustic signal to produce an acoustic time series. The acoustic time series is reconstructed using a phase space embedding algorithm to generate a phase space embedded acoustic signal. A differential radius signal is generated from the phase space embedded acoustic signal using chaotic radius computations and differential radius computations. Thresholds can be detected in the differential radius signal to reveal nonlinear or chaotic events hidden in the underwater acoustic signal.

Abstract: A method for processing acoustic signals received by first and second undater detectors, spaced from each other, for detection and determination of range and bearing of an underwater signal source, includes the steps of obtaining a time series xd of first random variables X from the first detector, obtaining a time series yd of second random variables Y from the second detector, and providing a window of a selected number of digital points of time series xd and yd.

Abstract: A method for detecting radiated and reflected acoustic signals from an unwater source, wherein the acoustic signals are present in a random noise environment and form an output therewith. The method includes the steps of operating an entropy-based acoustic receiver to receive the output, calculating the entropy of the received output, and comparing the calculated entropy with a selected entropy threshold value. A substantial difference is indicative of the presence of an acoustic signal from an underwater source. A less than substantial difference is indicative of a lack of presence of an acoustic signal from an underwater source.