Mobility based call admission control and resource estimation in mobile multimedia networks using artificial neural networks

Mobility management and resource utilization are the two most important research issues in wireless mobile multimedia networks. In order to provide the guarantee of handoff probability; we propose a mobility based call admission control using artificial neural network to predict the accurate future position of mobile user based on the mobility history of the user. For mobility prediction we have used very common and popular error back propagation neural network model. The concept of a error back propagation neural networks is completely distributed that shows current position and predict next movement of a mobile user, with this mobility information resources are estimated and reserved to improve the overall quality of service (QoS) in terms of continuous service availability and handoff dropping probability. For better resource utilization, the proposed scheme will predict when and where the next handoff takes place and in which cell the handoffs will occur. In addition, a call admission control scheme is used to decide whether a call is accepted or rejected in the current traffic which is based on the amount of resources after a handoff. The allocation of radio resources are made by the queuing scheme in which the handoff calls and new calls are queued to get the services. The proposed scheme increases the ratio of mobility prediction and decreases the call dropping probability and handoff dropping probability up to a maximum extent.