Multitarget Track Before Detect with MIMO Radars

Recent advances in Multiple-Input-Multiple-Output (MIMO) radar systems show that they have the potential to improve detection and localization performance of targets over bistatic and multistatic radars. Unlike beam forming, which presumes a high correlation between signals either transmitted or received by an array, the MIMO system exploits the independence between signals at the array elements due to transmit diversity. Previous works focus on waveform design, signal processing and target localization with MIMO radars while no attention has been given to tracking algorithms. In this work, the problem of tracking multiple targets using MIMO radars is considered. The scenario includes multiple targets in a widely-separated MIMO architecture in which Radar-Cross-Section (RCS) diversity can be utilized. Multi target version of Track-Before-Detect (TBD) algorithm is implemented for the collected $M \times N$ orthogonal signals at the receiver, where $M$ is the number of transmitters and $N$ is the number of receivers. Besides having the advantage of integrating information over time on unthresholded measurements to yield detection and tracking simultaneously, the TBD technique enables tracking and detecting targets in low Signal-to-Noise-Ratio (SNR) environments. Also, a modified multiple sensor TBD, which weights the target observability to the sensor as a result of target RCS diversity in the likelihood calculation to best fit the centralized MIMO tracking is proposed. Finally, Monte Carlo simulations are performed to evaluate the performance of the proposed tracking algorithm.