Abstract

New methods of underwater acoustic-propagation research have been developed in which noiselike or pseudorandom signals, generated by shift-register encoders, are detected by correlation techniques. The principles and problems of three classes of experiments are discussed: (1) the signal received is crosscorrelated with a time-delayed and time-compressed replica of the transmitted signal; (2) two received signals are crosscorrelated and studied as a function of hydrophone separation; (3) the received signal is correlated with that received from a later repetition of the transmitted signal. It is shown theoretically that the results of the first class of experiments are different from and unpredictable from those obtained by employing the usual filtering and detection of impulsive or single-frequency signals. The severe problem arising from multipath interference in the second class of experiments, which is different for the multipath problem in the first class of experiments, is discussed and some solutions proposed. The essential identity of the second and third classes of experiments is discussed. A companion paper discusses the instrumentation for and execution of the experiments and presents some preliminary observations.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call