Multidimensional Resource Management Optimization Strategy for High-Resolution Multitarget Tracking

In this paper, we propose a beam selection and multi-dimensional resource management (BSMRM) strategy for colocated multiple-input multiple-output (C-MIMO) radar systems to increase both target resolvability and multi-target tracking (MTT) accuracy. Specifically, the resolution performance is quantified through the normalized magnitude of the ambiguity function for both observed targets and undesired targets located within the beam. Moreover, we derive the Bayesian Cramér-Rao lower bound (BCRLB) under the probabilistic data association (PDA) fusion rule. In addition, a logarithmic barrier-based objective function is designed to quantify the global performance, while an acceptable error is introduced to ensure system robustness. Finally, the proposed strategy tackles a dual-object optimization problem to coordinate the transmit beam and power, signal bandwidth, and pulse length of each radar node, subject to the limited resource budgets. Accordingly, a three-stage-based solver is developed for the resultant optimization problem. Simulation results demonstrate that the BSMRM approach achieves target resolvability and MTT accuracy comparable to state-of-the-art algorithms, and exhibits higher efficiency in terms of spectrum occupancy and power utilization.