We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Energy-Efficient Access Point Selection Scheme for Reconfigurable-Intelligent-Surface-Assisted Cell-Free Massive MIMO Systems.
- Authors
Wang, Weiran; Song, Jiaqi; Zhou, Jianguo
- Abstract
Reconfigurable intelligent surface (RIS)-assisted cell-free (CF) massive Multiple-Input Multiple-Output (MIMO) technology exhibits significant potential in enhancing the energy efficiency of 6G mobile communications. Nevertheless, recent studies suggest that both access points (APs) and RISs encounter challenges related to a high energy consumption during operation. To address this issue, strategies involving AP hibernation and RIS shut-off are proposed. Subsequently, an optimization problem is formulated to jointly optimize RISs, beamforming vectors, and AP selection with the aim of maximizing the energy efficiency (EE). Initially, the non-convex optimization problem for maximizing energy efficiency is decomposed into three sub-problems. These sub-problems are subsequently reformulated using fractional programming and variational programming techniques and then solved using the successive convex approximation (SCA) algorithm, Dinkelbach algorithm, and greedy algorithm, respectively. Subsequently, an alternate optimization algorithm based on block gradient descent is introduced to iteratively solve the four-variable optimization problem, thereby obtaining an approximate solution to the original problem. The simulation results demonstrate that the algorithm significantly reduces energy consumption. Specifically, compared to the scheme without the hibernation strategy, the energy efficiency (EE) is enhanced by 35%.
- Subjects
MIMO systems; OPTIMIZATION algorithms; GREEDY algorithms; ENERGY consumption; HIBERNATION; POTENTIAL energy; FRACTIONAL programming; MACHINE-to-machine communications
- Publication
Electronics (2079-9292), 2024, Vol 13, Issue 7, p1397
- ISSN
2079-9292
- Publication type
Article
- DOI
10.3390/electronics13071397