Autonomic Joint Session Admission Control Using Reinforcement Learning
Jie Chen · Beijing Youdian Xueyuan xuebao · 2007
A reinforcement learning based joint session admission control(JOSAC) algorithm is proposed to realize the autonomic and distributed joint resource optimization between the heterogeneous radio access technologies(RAT) in a reconfigurable system.By introducing Q-learning into the admission control algorithm and adjusting the strength of the reinforcement signals for different types of sessions considering the inherent characteristics of different RATs,the characteristics of RATs is driven to absorb the suitable traffic for a proper service distribution.That will improve the efficiency of resource utilization.The simulation results show that,through the trial-and-error on-line learning process,overlapping RATs can converge to the optimized admission control policies that reduce the overall blocking probability and achieve lower handover dropping probability as well as higher revenue.