Predictive Tracking Simulation and Techniques for All-digital Radar

Abstract

All-digital radar architectures allow radar systems to operate with more flexibility. Multiple Input, Multiple Output (MIMO) radars are especially effective at identifying multiple targets. Combined with statistical predictive tracking techniques, All-digital radars can allocate antenna resources during future coherent processing intervals to improve target signal-to-noise ratio (SNR) over time.

This thesis describes the relationships and models that characterize an all-digital phased array. The cognitive radar system simulated in this thesis is adaptively divided into several transmitting sub-arrays that are all independently capable of aiming at multiple targets within the radar's range. This thesis also discusses predictive tracking strategies that allow the radar controller to continuously track multiple targets within the radar's range. It discusses the process of dividing an array while still maintaining a SNR above the detection threshold. This thesis also examines the limits on how many targets the simulated all-digital array can track over many coherent processing intervals. In general, the all-digital array was able to track more targets compared to a phased array that forms a single transmit and receive beam each CPI.