Machine Learning-Based Compensation Methods for Weight Matrices of SVD-MIMO

Abstract

This paper proposes and evaluates machine learning (ML)based compensation methods for the transmit (Tx) weight matrices of actual singular value decomposition (SVD)-multiple-input and multiple-output (MIMO) transmissions. These methods train ML models and compensate the Tx weight matrices by using a large amount of training data created from statistical distributions. Moreover, this paper proposes simplified channel metrics based on the channel quality of actual SVD-MIMO transmissions to evaluate compensation performance. The optimal parameters are determined from many ML parameters by using the metrics, and the metrics for this determination are evaluated. Finally, a comprehensive computer simulation shows that the optimal parameters improve performance by up to 7.0 dB compared with the conventional method.

[Paper] Field Experiment with FPUs for Mobile Relay Programs with Auto-switching of Total Modulation Orders

“…Therefore, the MER for the received signal r s and the transmitted signal t s for each sample s is measured by s (|t s | 2 )/ s (|r s − t s | 2 ). In addition, as shown in references 13), 20) , the MER for each independent channel i in an ideal environment is expressed by MER i = CNRξ i 2 p i using the received CNR CNR, the singular value ξ i of the channel matrix H, and the power allocation p i . Moreover, MER considering interference components has also been proposed 13), 20) .…”