Research On Probe Optimization Algorithm For MIMO OTA Test System

Multiple-input multiple-output(MIMO)over-the-air(OTA)test has attracted widespread attention as a solution for evaluating the performance of MIMO devices in the laboratory.Unlike single antenna terminals,the performance of MIMO devices is highly dependent on the spatial correlation between spatial sub-channels,while spatial correlation can be obtained from spatial channel covariance matrices.The multi-probe anechoic chamber(MPAC)method is a solution for OTA testing.It can fully and controllably reproduce any geometry-based channel model and flexibly change channel model parameters becoming the most widely used solution.Firstly,one of the main challenges of the MPAC is the cost of the system and each probe typically needs to be connected to an expensive channel simulator.If the probes can be flexibly placed according to channel space characteristics,a larger test area can be created than the condition with the same number of fixed probe settings.On this basis,this paper studies the flexible placement of the probe in the multi-probe darkroom,optimizes the position and power weight of the probe.This paper proposes the MIMO OTA probe selection based on particle swarm optimization(PSO)and the existing genetic algorithm.The multi-shot algorithm is compared in terms of algorithm complexity and the performance.In this paper,the mean square error(MSE)of the spatial correlation of the target channel and the spatial correlation of the actual channel is used as the evaluation factor.After the probe position is determined,the convex power optimization algorithm is used to optimize the probe power weight and the superiority of the proposed algorithm is verifi ed for different test scenarios.Secondly,in the traditional OTA test,the mobile phone is usually placed in a dark room,and the channel simulator is connected to the OTA darkroom and base station simulator respectively.In the base station test,the base station needs to be placed in the OTA darkroom to connect wirelessly in a radiated manner.In high frequency scenarios,the user equipment may be a highly integrated radio frequency and antenna component configured with no accessible ports between the user equipment and the channel simulator.Aiming at this problem,this paper proposes a dual darkroom channel model,establishes an end-to-end OTA test system,and uses the MSE of spatial correlation between the target channel and the antenna at both ends of the actual channel as the evaluation factor.The probe power matching problem under the double darkroom model is studied.The simulation of different cluster delay line channel models is carried out to verify the rationality of the established model.At the same time,after verifying the rationality of the dual darkroom channel model,the model was optimized by PSO.