Decentralized Beamforming for Cell-Free Massive MIMO with Unsupervised Learning
This addresses the challenge of communication overhead in wireless networks for improved spectral efficiency, though it appears incremental as it builds on existing beamforming approaches with new learning-based optimizations.
The paper tackled the problem of high signaling overhead in cell-free massive MIMO systems by proposing unsupervised deep neural networks for decentralized beamforming, achieving near-optimal sum-rate and reducing computational complexity by 10-24x compared to conventional methods.
Cell-free massive MIMO (CF-mMIMO) systems represent a promising approach to increase the spectral efficiency of wireless communication systems. However, near-optimal beamforming solutions require a large amount of signaling exchange between access points (APs) and the network controller (NC). In this letter, we propose two unsupervised deep neural networks (DNN) architectures, fully and partially distributed, that can perform decentralized coordinated beamforming with zero or limited communication overhead between APs and NC, for both fully digital and hybrid precoding. The proposed DNNs achieve near-optimal sum-rate while also reducing computational complexity by 10-24x compared to conventional near-optimal solutions.