Narrow-Beam Scheduling for Multitarget Tracking in High-Precision Phased Array Radar Systems

Abstract

The capability of electronic beam steering enables phased array radar (PAR) technology to play a central role in modern multifunction radar (MFR) systems, which typically perform target search and tracking. In scenarios involving multiple targets, the available radar beam resources must be appropriately allocated to achieve the optimal multitarget tracking (MTT) performance. Classical strategies in radar beam resource scheduling for MTT have been extensively investigated, under the assumption that the radar beam is sufficiently broad to ensure target illumination.

Since the angular accuracy is inversely proportional to radar beamwidth, reduced beamwidths have been increasingly demanded in high-precision radar systems. However, when narrow beams are employed, targets are more likely to be missed from beam scans since their uncertainty regions may not be effectively encompassed by the beams. This effect challenges the classical beam scheduling methods, as it violates the key assumption underlying the beam-pointing control strategies and tracking performance metrics.

This dissertation focuses on developing the narrow-beam scheduling (NBS) strategies to facilitate MTT under different PAR system frameworks. The study progresses from a single multibeam PAR to PAR networks, and finally to a dual-PAR network enabling cooperative narrow-beam sensing. Noting that a missed illumination under narrow-beam sensing implies that the target is more likely located outside the beam-scanned region, filtering methods have been developed to refine the posterior knowledge of the target location. Based on the proposed filters, narrow-beam steering strategies are formulated to specify the beam-pointing angles that ensure target illumination as rapidly as possible. To quantify the expected tracking performance obtained from narrow-beam sensing, performance metrics are derived along with efficient evaluation methods. By exploiting the corresponding performance metrics as objective functions, the NBS problems under different sensing frameworks are formulated as mathematical optimizations. Owing to the complexity of obtaining optimal solutions, practical methods for determining suboptimal solutions with satisfactory performance are developed. Numerical simulation results demonstrate the superior performance of the proposed beam schedulers in different narrow-beam MTT scenarios.