Deep Learning Based Energy, Spectrum, and SINR-Margin Tradeoff Enabled Resource Allocation Strategies for 6G

Abstract

In the rapidly evolving landscape of wireless communication systems, the forthcoming sixth-generation technology aims to achieve remarkable milestones, including ultra-high data rates and improved Spectrum Efficiency (SE), Energy Efficiency (EE), and quality of service. However, a key challenge lies in the transmission at Terahertz frequencies, which entails significant signal loss, resulting in reduced signal-to-interference and noise ratio margins (Γ). Increased transmit power can ameliorate Γ and SE, thereby sacrificing EE. Consequently, it necessitates strategic Resource Allocation (RA) to uphold an optimal trade-off amid SE, EE and Γ. In this paper, we propose a series of RA strategic algorithms harnessing the Transfer Learning, Growth-Share (GS) matrix, Game Theory (GT), and service priorities to tailor the aforementioned trade-off. This endeavour renders the network more intelligent, self-sufficient, and resilient. Furthermore, we have seamlessly integrated Device-to-Device communication scenarios into our proposed algorithms, enhancing SE and network capacity. The proposed integration aims to strengthen overall system performance and accommodate the evolving demands of future wireless networks. Our primary contribution lies in the development of the GS-GT-based Optimal PathFinder (GS-GTOPF) algorithm to identify optimal paths based on SE using Deep Neural Networks. Thereafter, we formulate an enhanced version of it by integrating service priorities (GS-GTOPF-SP). This refinement has been further advanced by reducing the Computational Time (CT), resulting in GS-GTOPF-SP-rCT. Further improvement is achieved by introducing the angle criterion (GS-GTOPF-SP-rCT-θ). Extensive simulations demonstrate that angle criterion integrated algorithm, showcases a remarkable 76.12% reduction in CT while maintaining an accuracy surpassing 95% compared to GS-GTOPF. Moreover, prioritizing high-priority services leads to a significant enhancement of 12.97% and 62.95% in SE, 16.14% and 81.97% in EE, and 12.27% and 25.95% in Γ when compared to medium and low-priority services. © 2023 IEEE.