Abstract
The rapid advancement of wireless communication technologies has significantly increased energy consumption in cellular networks. To address this issue, we propose a multilevel sleep mode strategy for multi-base station (BS) collaboration, assisted by cell zooming and joint user association. Sleep mode enables BSs with low traffic to enter different sleep states for energy savings, while cell zooming dynamically adjusts BS coverage. To mitigate the delay introduced by sleep mode, an energy-saving scheme under delay constraints is developed. The optimization is modeled as a Markov decision process (MDP) and solved using the proximal policy optimization (PPO) algorithm. Moreover, a bidirectional long short-term memory (Bi-LSTM) network is employed for traffic prediction to enhance decisionmaking. Simulation results validate the superiority of the proposed scheme in energy savings compared to benchmark methods.