A Deep Learning Clustering Beamforming Approach for Future 6G Mobile Ad Hoc Networks

Abstract

The rapid advancement of wireless technologies and the increasing demand for seamless, high-speed connectivity has spurred the need for efficient routing protocols in 6G Mobile Ad Hoc Networks (MANETs). This paper proposes a modern approach, the Deep Learning Clustering Beamforming Massive MIMO Routing Protocol (DLCB), that is designed to achieve high level throughput by exploiting beamforming advantages and for enhancing communication efficiency for future next generation 6G MANETs. The DLCB algorithm establishes robust and efficient routes for data transmission by adopting a clustering strategy based on deep learning allowing to select cluster heads by taking advantage of Graph Neural Networks (GNN) features and creating effective cluster paths. Furthermore, this approach integrates IEEE 802.11ay and Massive MIMO technologies to exploit their benefits in terms of high-speed data transmission, improved network capacity, and better signal quality. DLCB aims to facilitate seamless data transmission in dynamic network environments, paving the way for enhanced network performance in 6G MANETs. Experimental evaluations and simulations are conducted to validate the performance of the proposed approach through the use of Omnet++ Network Simulator.