Stochastic model of a user-centric joint call admission control scheme for heterogeneous wireless networks

Next generation wireless networks (NGWN) will be heterogeneous, comprising a number of radio access technologies (RATs) coexisting in the same geographical area. In NGWN, a joint call admission control (JCAC) algorithm is needed to select the most appropriate RAT for each incoming call. The JCAC algorithm must be user-centric (i.e., consider user preferences in making RAT selection decisions) in order to enhance userspsila satisfaction. This paper investigates a user-centric JCAC (UJCAC) algorithm for heterogeneous wireless networks. The UJCAC algorithm uses Fuzzy multiple-attribute decision-making technique to select the most appropriate RAT for each incoming call. We develop a Markov model for the UJCAC algorithm to evaluate the overall call blocking probability and handoff call dropping probability in heterogeneous wireless networks. Performance of the UJCAC algorithm is evaluated using a three-RAT heterogeneous network. Simulation results show that a UJCAC algorithm can lead to highly unbalanced network load, and high call blocking/dropping probability. Therefore, a flexible UJCAC algorithm is more suitable for NGWN. In flexible UJCAC, if the mostly preferred RAT for incoming call is fully loaded, the next preferred RAT will be selected for the call. The call will be blocked only if none of the available RATs can accommodate the call. Therefore, call blocking/dropping probability will be reduced.