No abstract

No abstract

Sonobuoy fields, consisting of a large network of emitter and receiver sonar sensors on buoys, are increasingly being used for detection and tracking of underwater targets in a defined maritime area. This study presents a Gaussian mixture version of a multitarget-multisensor (MS) Bayesian-type tracker developed specifically for multistatic sonobuoy fields. Its foundation is the optimal Bayesian MS filter for a single target in clutter. The multi target feature is incorporated using the linearmultitarget paradigm, which is a fast and accurate approximation assuming the density of underwater targets is low. Reliable track initiation and false track discrimination for low signal-to-noise ratio targets are achieved using the amplitude feature of reported detections. The developed tracker is capable of processing measurements with different modalities, depending on the transmitted signal waveform. It is integrated and tested within a realistic multistatic sonar emulator developed by DST Group.

Sonobuoy fields, consisting of many distributed emitter and receiver sonar sensors on buoys, are used to seek and track underwater targets in a defined search area. The authors seek a scheduling protocol, selecting both the emitter and its waveform in each time interval that optimises tracking performance. This study describes a stationary scheduling algorithm for sonobuoy fields called the continuous probability states algorithm. The algorithm replaces a full partially observed Markov decision process by a computationally feasible Markov decision process by focusing on probability of target detection. This approach is shown to result in high-quality tracks for multiple targets in a realistic simulation of a sonobuoy field.