Maximum-likelihood estimation and detection for moving sources in waveguides

Abstract

In this paper, we examine the problem of direction finding and detection of a moving source in a waveguide. Specifically, we are concerned with detecting a wide-band low signal power source moving through the ocean. Further, we desire an algorithm that is resistant to strong interferers (e.g. surface ships) that may also be moving within or on the boundary of the waveguide. The physical model that is given in this paper was derived from moving (constant velocity) sources traveling through a waveguide. The result can be considered to be a generalization of a result derived by Hawker (1979). The amplitude of the source signals and the variance of the noise are taken as unknown deterministic quantities. To determine the parameter estimates for the purpose of localization and detection, we use the maximum-likelihood (ML) estimation algorithm. For the detection algorithm, we use the generalized likelihood ratio test (GLRT).