Abstract
Fast session transfer (FST) is introduced by wireless gigabit (WiGig) standards to transfer the data communication session among wireless fidelity (Wi-Fi) and WiGig channels based on the availability. In the conventional Wi-Fi/WiGig FST decision-making, the session of data transmission is transferred from Wi-Fi to WiGig whenever it is possible regardless of the traffic loads on both bands and the expected WiGig blocking probability. Also, WiGig to Wi-Fi FST occurs whenever the WiGig signal is lost even if the WiGig link is undergoing an instant path blocking. This results in Wi-Fi/WiGig load imbalance as well as large average delay in user data delivery. In this study, a novel Wi-Fi/WiGig FST decision-making algorithm is proposed utilising fuzzy logic while considering the coverage probability of the WiGig link, the sizes of the users' remaining messages relative to the available data rates of both bands, and the probability of WiGig path blocking. Furthermore, a WiGig interruption-classification methodology is proposed to detect the occurrence of the WiGig instant path blocking and then decide if it is better to wait for WiGig signal recovery or immediately transfer to Wi-Fi. Simulation analysis demonstrates the effectiveness of the proposed Wi-Fi/WiGig FST decision-making algorithm, over the conventional one.