Enabling fast handoff in dense 802.11 WMNs via temporal-spatial knowledge and opportunistic probing

Recently the deployment of IEEE 802.11 Wireless Mesh Networks (WMNs) has been growing exponentially attributing to sharing economy of WiFi. As a result, the handoff delay caused by mobile client (MC) probing for the next preferred AP imposes more challenges for time-sensitive mobile applications such as VoIP, video chat, and IoT in a dense WMN with more and more MCs and APs. In this work, we exploit the characteristic of dense network and propose two fast handoff schemes, namely T-NRS (Temporal-Network Radio Signature) and OppoScan. T-NRS scheme leverages historical knowledge of associated APs in time series together with radio signature technique based on spatial knowledge to assist in fast handoff, while OppoScan scheme opportunistically leverages nearby MCs and APs to produce the required information of neighboring AP for handoff and thus significantly decrease the number of switching channel of APs. We implement and evaluate the mechanisms in testbed of about 200 APs in shopping malls. The results prove that the performance of the proposed schemes is satisfactory while cost less than previous approaches.