Optimal Base Station Placement for Beyond 5G Networks with Non-Convex Topology
2026-07-06 • Emerging Technologies
Emerging Technologies
AI summaryⓘ
The authors studied how to best place a millimeter-wave base station inside a complicated U-shaped area, which is hard because of its shape and tricky goals like maximizing data rates fairly. They turned the placement problem into a Markov Decision Process and used two types of deep reinforcement learning methods: one that picks spots on a grid (DQN) and another that breaks the area into smaller parts with separate agents (DDPG). Their results show that placing the base station optimally covers the area well and is very fair in service distribution. The multi-part DDPG method gave better results than the grid-based method while being simpler.
millimeter-wave (mmWave)base station (BS)non-convex topologysum-rate maximizationmax-min fairnessMarkov Decision Process (MDP)Deep Q-Network (DQN)Deep Deterministic Policy Gradient (DDPG)Jain indexdeep reinforcement learning (DRL)
Authors
Mohamed Shalma, Amr Mansour, Ahmed El-Mahdy
Abstract
This paper investigates the optimal placement of a millimeter-wave (mmWave) base station (BS) within a realistic U-shaped environment with non-convex topology. The problem is challenging and NP-hard due to the non-convex topology and the non-convex objective functions which are the sum-rate maximization and max-min fairness, the latter being additionally non-smooth. To address this challenge, the BS placement is formulated as a Markov Decision Process (MDP). Then, we propose two deep reinforcement learning (DRL) techniques: First, the deployment area is discretized into a grid and optimized using a Deep Q-Network (DQN). Second, the U-shaped region is partitioned into continuous subspaces, where a Deep Deterministic Policy Gradient (DDPG) agent is dedicated to each subspace then the best BS placement is selected among partitions. Results demonstrate that optimal placement achieves full coverage and yields a Jain index of 0.99. Furthermore, the proposed partitioned multi-space DDPG achieves better solution than DQN with lower complexity.