Mobile Data Offloading addressing the Service Quality vs. Resource Utilisation Dilemma

Mobile operators are integrating WiFi into their cellular networks as a way to address the issue of mobile data deluge. Although prominently used for its high data rates and unlicensed frequency band of operation, WiFi's contention based medium access does not guarantee that a user's quality of service will necessarily be met, especially if mobile devices simply 'offload' to WiFi in preference to cellular. A network selection algorithm offers a way to enhance user experience, by always selecting the best radio access network for the user. In the literature, several techniques have been applied to the network selection problem. While most proposals aim to optimise quality of service for the user, the impact on resource utilisation for the networks is often neglected. This work proposes a network selection algorithm which attempts to find a good trade-off between user quality of service and efficient resource utilisation and elaborates on its performance evaluation through a simulation based study. Findings indicate that the proposed algorithm is able to provide enhanced QoS to the user as compared to conventional algorithms in terms of achieving higher throughput, lower delay as well as a higher overall system performance.