A Reinforcement Learning Based Joint Call Admission Control for Heterogeneous Wireless Networks

Currently, there are many wireless networks based on different radio access technologies (RATs). Despite this, new kind of networks will be developed to complement those already existing today. As there will be no RAT able to give users full service requirements with universal coverage, the next generation wireless networks will integrate multiple technologies, working jointly on a heterogeneous way. Heterogeneous networks necessitate joint radio resource management (JRRM) mechanism to enhance better resource utilization and give users better quality of service. Joint call admission controls (JCAC) are a kind of JRRM mechanisms. In this paper, we present a JCAC approach to heterogeneous wireless network management based on reinforcement learning to treat call admission and technology selection, enhancing the network’s performance. The effectiveness of this approach is assessed in terms of blocking rate results obtained by two simulation scenarios. Joint call admission control, JCAC, resource allocation, reinforcement learning, heterogeneous networks.