Multi-access interface selection based on data mining algorithm

Multi-access networks are envisioned to guarantee a continuous service while addressing different application requirements in terms of data rate, coverage and power consumption. Different access technologies can be effectively combined in a multi-access scenario where the overall system is able to achieve a better performance leveraging on the different characteristics of the access technologies. One of the main challenges in realizing a multi-access system is the process of properly selecting a network interface. This paper proposes a network selection algorithm based on the k-nearest neighbor data mining classification technique. This work considers Ethernet, Wi-Fi, and 3GPP technologies; however, the proposed approach could be easily adapted to take into account other access technologies.