Learnable Model-Driven Performance Prediction and Optimization for Imperfect MIMO System: Framework and Application

TL;DR

This paper introduces a learnable, model-driven framework for performance prediction and optimization in imperfect MIMO systems, effectively handling unknown interference and CSI uncertainty through neural networks and iterative estimation.

Contribution

It presents a novel framework combining model-driven approximations with neural networks for adaptive performance prediction and optimization in complex MIMO scenarios.

Findings

Neural network improves sum rate and detection error prediction.

Iterative CSI uncertainty estimation enhances optimization accuracy.

Deep unfolded algorithm balances convergence speed and robustness.

Abstract

State-of-the-art schemes for performance analysis and optimization of multiple-input multiple-output systems generally experience degradation or even become invalid in dynamic complex scenarios with unknown interference and channel state information (CSI) uncertainty. To adapt to the challenging settings and better accomplish these network auto-tuning tasks, we propose a generic learnable model-driven framework in this paper. To explain how the proposed framework works, we consider regularized zero-forcing precoding as a usage instance and design a light-weight neural network for refined prediction of sum rate and detection error based on coarse model-driven approximations. Then, we estimate the CSI uncertainty on the learned predictor in an iterative manner and, on this basis, optimize the transmit regularization term and subsequent receive power scaling factors. A deep unfolded projected gradient descent based algorithm is proposed for power scaling, which achieves favorable trade-off between convergence rate and robustness.