Loading...

Resource Allocation in 6G Wireless Networks with Machine Learning

Zarini, Hossein | 2025

1 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 58904 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Movaghar Rahimabadi, Ali; Hessabi, Shaahin
  7. Abstract:
  8. In recent years, wireless network services have been classified into three main categories: enhanced Mobile Broadband (eMBB), ultra-reliable and low-latency communications (uRLLC) and massive Machine-Type Communications (mMTC). Most existing research has focused on the coexistence of two of these services, and to date, no study has investigated the coexistence of all three services from the perspective of radio resource allocation. In this research, downlink transmission of a non-orthogonal visible light communication (VLC) system, supported by unmanned aerial vehicles (UAVs), is investigated to enable the coexistence of all three services. A resource allocation problem is formulated to optimize association of users with UAVs, as well as transmit power and flight trajectory of UAVs to establish a multi-objective trade-off between total weighted power consumption per UAV and quality of experience (QoE) of its associated eMBB users, while satisfying the requirements of all users. Considering uncertainty in channel state information, a generalized Benders decomposition approach is employed, which combines convex optimization tools with deep multi-agent deep reinforcement learning algorithms to solve the problem. Moreover, the upper and lower bounds of the reward function of UAVs as learning agents are analytically derived. Simulation results demonstrate that the proposed method, compared to conventional single-agent approaches, achieves lower power consumption of UAVs and improved QoE of eMBB users. Furthermore, the results show that in a small-scale system, the performance degradation of the proposed method—compared to adoption of globally optimal association of users to UAVs—is negligible
  9. Keywords:
  10. Enhanced Mobile BroadBand (eMBB) ; Radio Sources Localization ; Ultra-Reliable Low-Latency Communication (URLLC) ; Unmanned Aerial Vehicles (UAV) ; Visible Light Communication ; Massive Machine Type Communication (mMTC)

 Digital Object List

 Bookmark

...see more