A semi-Markov decision process-based joint call admission control for inter-RAT cell re-selection in next generation wireless networks

In the next generation wireless networks (NGWNs), where different radio access technologies (RAT) will coexist and work in collaboration to provide ubiquitous access, a mechanism called Joint Call Admission Control (JCAC) will play an important role by deciding whether or not an incoming service request will be accepted according to an admission constraint as well as determining in which RAT (among the available) it will be connected. In this paper, we propose an optimal JCAC for inter-RAT cell re-selection problem also referred to as initial RAT selection in co-located wireless networks, which supports both real-time services and non-real-time services. To properly meet the JCAC goals, we propose a cost function that weigh two criteria: the blocking cost function, which takes into account the priority of each service class in each RAT, and the alternative acceptance cost, which reflects the multiplicity of RATs working in a collaborative fashion, mandatory in NGWN. We use the framework of Semi-Markov Decision Process (SMDP) to formulate the optimization problem and the value iteration algorithm to compute the optimal policy. Our model still takes into consideration the ratio between the radius of the co-located RATs and shows how it may impact on optimal initial RAT selection. Numerical results, supported by an analysis of the structure of the optimal policy, show that the proposed optimal JCAC selects for real-time service class the biggest RAT and for non-real-time service class the smallest one. This optimal JCAC policy is ratified by the current trend in the design of NGWN and also follows the 3rd Generation Partnership Project (3GPP) expectations.