Near-Field Target Detection with Range-Angle-Coupled Matching Based on Distributed MIMO Radar.

With respect to distributed MIMO radar systems, conventional far-field detection methods fail under near-field conditions due to significant wavefront curvature, which inevitably results in target energy loss and erroneous parameter estimation. To solve this problem, we propose a near-field target detection framework based on range-angle-coupled matching in this study. Firstly, we design the linear frequency modulation by frequency division (FD-LFM) signal. In addition to offering favorable orthogonality and Doppler tolerance, the transmitter of distributed MIMO radar employs a wide beamwidth to mitigate the low scanning efficiency associated with beam positioning in distributed phased array (PA) radar systems. Secondly, we develop a three-dimensional grid-based echo model for near-field targets in range-azimuth-elevation domain. Specifically, we conceive a coherent pulse integration method via multi-dimensional matching, which enables precise delay alignment and echo accumulation across all transmit-receive pairs for accurate near-field target detection. Thirdly, we propose a parallelization scheme for distributed MIMO radar near-field processing. Our proposal not only compensates effectively for spherical wave propagation effects but also achieves real-time processing through GPU acceleration. Finally, our proposed method's feasibility of high resolution and effectiveness of near-field detection have been verified by field experimental simulation and actual measurement processing results.