Tracking performance loss due to false associations of contacts from semi-coherent signal processing chains

The performance of an active sonar depends on the transmitted waveforms. Hyperbolic frequency modulated waveforms (HFM) are commonly used to obtain sonar contacts with high resolution for Time Differences of Arrival (TDoA) measurements. The signal to noise ratio of HFM matched filters is not significantly affected by the range-rate of a target. But a Doppler-shift in target echoes results in a shift of the TDoA. By transmitting two different HFM waveforms, one with a positive, the other with a negative frequency gradient, and by association of the generated contacts the different range shifts can be used to get an adequate Doppler estimation and increase the TDoA accuracy. However in noisy and fading channels, this semi-coherent data fusion suffers from false associations and can lead in multi-target scenarios or scenarios with many false contacts to a decreasing tracking performance. In this paper we analyze the impact of the false association on the tracking result in multistatic scenarios by utilizing the tracking performance metrics established in the Multistatic Tracking Working Group (MSTWG).