Gossipmule: Improving association decisions via opportunistic recommendations

The IEEE 802.11 association procedure lacks Quality of Experience (QoE) mechanisms. High delays and an association decision based on the Received Signal Strength Indicator (RSSI) do not lead to a QoE-based association among stations (STAs) and access points (APs). This paper presents Gossipmule, a novel approach that takes advantage of opportunistic communication among STAs to improve the association procedure in 802.11 wireless networks. STAs in Gossipmule exchange information prior to association by gossiping on the capabilities and performance of the APs. The knowledge gathered by this participatory sensing is used to rank the APs and update the recommendation that will be given to other peers. Results from simulations show a measurable improvement regarding the current delay and the achieved throughput.