Abstract :
The paper analyses the problem of locating an undersea platform in time-space by passive acoustic techniques. The radiated signature of the platform is utilized to determine a collection of parameters,called a fix, from which the source/receiver geometry can be reconstructed. Two different configurations of important practical implication are considered. In one, a large array observes the spatial diversity of the source signature, with the source considered practically at rest during each observation interval. It illustrates the issues involved with passive ranging by measuring curvature with an extended array. In the second, a short array observes a moving platform for a certain period of time, processing the temporal modulations induced on the radiated signature by the relative dynamics. The study concentrates on quantifying the error performance associated with the acquisition of the global range parameter. Planar geometries, constant speed motions and linear structures are considedered. The signals are assumed narrow-band (single tone for simplicity). The disturbing observation noises are modeled white. The paper concentrates on the acquisition step, illustrating for the underwater platform location problem results develloped elsewhere. It computes the tradeoffs among the geometry, the dynamics, the statistical parameters and the attainable accuracies. The underwater acoustics problem constraints, namely the weak signal to noise ratio and the small geometry diversity available, usually place the configurations herein studied at the boundary regions where performance transitions and thresholds occur. For the nominal values assumed, it is illustrated how the predicted performance exhibits significant changes. This stresses the importance of realistically dimensioning the sensors quality and tke array baseline to attain desired levels of accuracy.
