As the demand of the mobile users are increasing day by day, wireless/mobile multimedia networks still need advancement in terms of, reliable traffic performance, link availability, efficient bandwidth utilization, and user mobility, that can attain extremely consistent wireless communication and data transmission over the networks. Due to the emerging demand of multimedia services a high-speed network and call admission control (CAC) scheme is required, which not only guarantees the quality of services (QoS) for new and handoff calls but also results in optimum resource utilization in bursty traffic network environments. The main objective of this integrated neural fuzzy based CAC scheme is to improve QoS with decent resource allocation, such that it minimizes the probability of call dropping and call blocking in mobile multimedia networks. The proposed neural fuzzy CAC scheme is a hybrid approach that integrates the semantic rule ability of fuzzy logic (FL) controller and self-training capability of a neural network (NN) which is further enhanced to construct an efficient computational model for traffic control and fair radio resources allocation for new calls and handoff calls. The simulation results conclude that a neural fuzzy based CAC can achieve minimal call dropping probabilities and maximum resource utilization in high-speed networks as compared to fuzzy logic based CAC and conventional CAC or existing CAC schemes
Reinforcement-learning-based call admission control and bandwidth adaptation in mobile multimedia networks
