VCOMP: An Efficient Probe Selection Method for 5G OTA Testing with MPAC Setup

The multi-probe anechoic chamber (MPAC) method is considered the most promising approach for over-the-air (OTA) testing of 5G massive multiple-input multiple-output (MIMO) devices. Since the cost of the MPAC system is directly influenced by the number of ports on the channel emulator, which corresponds to the number of OTA probes, the selection of probes is crucial for OTA testing. However, the classical convex-optimization-based probe selection method has become prohibitively complex due to the time-consuming computation of constrained weights and the ever-increasing number of candidate probes. Therefore, this paper proposes a variable step-size constrained orthogonal matching pursuit (VCOMP) algorithm. The proposed algorithm includes a constraint handling approach that combines closed-form solution with an iterative method, and a novel probe selection method with variable step-size initial screening and correction. Simulation results indicate that, compared with existing algorithms, the proposed algorithm achieves higher channel reconstruction accuracy in a shorter time, making it more suitable for massive MIMO testing scenarios.