Target tracking in multi-static active sonar systems using dynamic programming and Hough transform

Using multiple source-receiver combinations in sonar tracking systems in multistatic active sonar scenarios has been drawing a lot of attention. Significant challenges in tracking targets in this scenario are the large number of contacts acquired by the receivers, the high level of false alarm clutters and the uncertainty of the track hypothesis. Conventional tracking approaches (such as the Kalman filter) require a predetermined kinematic model of the target. This paper presents a rapid and reliable tracking algorithm for underwater targets in the multistatic active sonar scenario that combines the use of Self Organizing Maps, Dynamic Programming and the Hough transform, leaving aside the need for a target kinematic model. Results on simulated data demonstrating the improvement that can be achieved with the new method are also presented.