Improved Target Localization With Off-Grid Compressed Sensing for Multistatic MIMO-OFDM Signals

Abstract

This article addresses the challenge of accurate target localization in fifth-generation (5G) communication networks using multistatic multi-input-multi-output orthogonal frequency division multiplexing (MIMO-OFDM) waveforms. Conventional on-grid compressed sensing-based target parameter estimation methods degrade significantly when targets are located off the predefined grid points. To overcome this limitation, we propose an off-grid compressed sensing approach that uses a grid evolution technique specifically designed for the complex-valued, block sparse structure inherent in multistatic MIMO-OFDM signal. By adaptively refining the grid during the sensing process, the proposed method achieves improved target localization accuracy, particularly in off-grid scenarios. Simulation results demonstrate that this approach significantly outperforms traditional methods, enhancing localization accuracy for 5G-enabled sensor networks.