Revenue optimisation and user prioritisation using call admission control strategies in multi-service 4G cellular wireless networks

The next generation of mobile networks will utilise multiple radio access technologies, seamlessly integrated to enable total ubiquitous network connectivity and high capacity for a multitude of future bandwidth intensive services. An efficient and proactive radio resource management system will be critical in such networks to enable full utilisation of network capacity, whilst providing sufficient capacity for services that generate maximum revenue for the network operator and for handover traffic. In this paper we investigate the use of capacity surfaces, policy based management and evolutionary computing in the development of a call admission control strategy. The call admission control algorithm is assessed in terms of resource utilisation, prioritisation of user/service requests and revenue generation. It is shown that one of the proposed strategies which uses a combination of neural network service prediction and genetic algorithm optimisation of service selection fulfils all three criteria and is therefore an ideal candidate for call admission control in next generation mobile networks.