A Novel Wireless Cable Method for High-Order MIMO Device Testing Based on Maximum RSRP Measurement

The wireless cable method, which can functionally realize conductive cable testing without actual radio frequency (RF) cables, is a promising solution for end-to-end over-the-air (OTA) testing of highly integrated multiple-input multiple-output (MIMO) devices. By accurately estimating and compensating for the transfer matrix between probes and device under test (DUT) antenna ports, signals from the test instrument can be independently routed to the respective DUT antenna ports. However, most channel estimation solutions for wireless cable calibration require Synchronization Signal (SS) Reference Signal Received Power (RSRP) per branch information. This cannot be directly applied to off-the-shelf commercial DUTs, as accessing SS-RSRP per branch requires specialized chipset support. Typically, only the maximum RSRP of all the DUT antenna ports can be directly obtained with commercial devices. The state-of-the-art method, based only on maximum SS-RSRP measurements, can achieve only a 2 × 2 MIMO wireless cable connection by adjusting the phase and amplitude of the probes, since the RSRP information for two DUT antenna ports can be independently discovered. Nevertheless, the applicability of this method to high-order (4×4 and above) MIMO DUTs is limited, as the RSRP information for specific DUT antenna ports may always be undiscoverable with an increasing number of DUT antennas. For this problem, this paper proposes a novel precoding-based calibration method that can cost-effectively establish wireless cable connections for high-order MIMO DUTs. Exploring the precoding matrix, the proposed method can flexibly adjust the composite transfer matrix between the testing instrument and DUT antenna ports to access the required RSRP information for different DUT antennas. The principle and framework are discussed with numerical simulations. Furthermore, experimental validation for a four-antenna DUT is provided, demonstrating the effectiveness and robustness of the proposed algorithm.