Federated Dynamic Neural Network for Deep MIMO Detection

TL;DR

This paper introduces a federated dynamic neural network for MIMO detection that adaptively chooses between two detectors for each sample and reduces communication overhead through federated learning techniques.

Contribution

It proposes a novel federated dynamic detection network that combines sample-wise adaptive routing with federated learning and gradient sparsification to improve MIMO detection efficiency.

Findings

The DDNet detector outperforms existing detectors across various system conditions.

Federated DDNet, especially FedGS-DDNet, reduces transmission overhead by over 25%.

The proposed methods maintain high detection accuracy with lower communication costs.

Abstract

In this paper, we develop a dynamic detection network (DDNet) based detector for multiple-input multiple-output (MIMO) systems. By constructing an improved DetNet (IDetNet) detector and the OAMPNet detector as two independent network branches, the DDNet detector performs sample-wise dynamic routing to adaptively select a better one between the IDetNet and the OAMPNet detectors for every samples under different system conditions. To avoid the prohibitive transmission overhead of dataset collection in centralized learning (CL), we propose the federated averaging (FedAve)-DDNet detector, where all raw data are kept at local clients and only locally trained model parameters are transmitted to the central server for aggregation. To further reduce the transmission overhead, we develop the federated gradient sparsification (FedGS)-DDNet detector by randomly sampling gradients with elaborately calculated probability when uploading gradients to the central server. Based on simulation results, the proposed DDNet detector consistently outperforms other detectors under all system conditions thanks to the sample-wise dynamic routing. Moreover, the federated DDNet detectors, especially the FedGS-DDNet detector, can reduce the transmission overhead by at least 25.7\% while maintaining satisfactory detection accuracy.