Call Admission Control Using Neural Network and SVM

Abstract: This research work introduces and conditionally look over a software based method for Call Admission Control using Neural Network and SVM. In this paper, Call Admission Control scheme using Neural Network and SVM is proposed for better QoS. THE rising demand for mobile communication services is increasing the importance of efficient use of the limited bandwidth and frequency spectrum. In recent years, considerable efforts have focused on the Channel Allocation and Call Admission Control (CAC) problems and many schemes that range from static to dynamic strategies have been proposed in the literature. Call Admission Control is a provisioning strategy used to limit the number of call connections into the networks in order to reduce the network congestion and provide the desired Quality of Service (QoS) to users in service. Traditional CAC schemes that mainly focus on the tradeoff between new call blocking probability and handoff call blocking probability cannot solve the problem of congestion in wireless networks. To overcome the problems arises due to traditional CAC schemes we propose a new CAC using hybrid technique i.e. SVM and Neural network.

Keywords: Call admission control, WCDMA, Offered traffic, Total carried traffic.