{"title":"SDWDS: Fault Recovery Automation in IoTs","authors":"Chin-Ya Huang, Hong-Yi Wang, Yu-Pei Wu","doi":"10.1109/SAHCN.2018.8397152","DOIUrl":null,"url":null,"abstract":"To assist various intelligent applications, IoT devices send their monitored data to Internet for further process. WiFi access points (APs) would be a candidate to support data transmission in IoTs due to its large wireless service coverage and low cost. An AP connects to the Internet through Ethernet to provide Internet connectivity to its connected IoT devices. However, data transmission may be failed when an AP loses its Ethernet connectivity. To resolve this problem, a software define networking (SDN) and wireless distribution system (WDS) assisted fault recovery automation scheme, SDWDS, is proposed. In SDWDS, packets can be directly and wirelessly forwarded between two APs when the Ethernet link fails at one of the APs. Specifically, when the Ethernet link failure of an AP is detected by the SDN controller, the AP will switch to WDS mode to wirelessly connect to its neighboring APs. The SDN controller in further reroutes corresponding packets through the newly associated AP. Consequently, the IoT devices can sustain Internet access even though their associated AP cannot directly communicate with the Internet via Ethernet. Additionally, the proposed SDWDS is implemented in commercial APs for performance evaluation. Preliminary results show the SDWDS can automatically recover more than one Ethernet link failures in the network. However, due to the limitation of the commercial AP, the creation of WDS requires rebooting the AP which costs about 20 seconds recovery latency. In the future, we will introduce virtualization or multi- connectivity techniques to more efficiently support the fault recovery automation in IoTs.","PeriodicalId":139623,"journal":{"name":"2018 15th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 15th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SAHCN.2018.8397152","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
To assist various intelligent applications, IoT devices send their monitored data to Internet for further process. WiFi access points (APs) would be a candidate to support data transmission in IoTs due to its large wireless service coverage and low cost. An AP connects to the Internet through Ethernet to provide Internet connectivity to its connected IoT devices. However, data transmission may be failed when an AP loses its Ethernet connectivity. To resolve this problem, a software define networking (SDN) and wireless distribution system (WDS) assisted fault recovery automation scheme, SDWDS, is proposed. In SDWDS, packets can be directly and wirelessly forwarded between two APs when the Ethernet link fails at one of the APs. Specifically, when the Ethernet link failure of an AP is detected by the SDN controller, the AP will switch to WDS mode to wirelessly connect to its neighboring APs. The SDN controller in further reroutes corresponding packets through the newly associated AP. Consequently, the IoT devices can sustain Internet access even though their associated AP cannot directly communicate with the Internet via Ethernet. Additionally, the proposed SDWDS is implemented in commercial APs for performance evaluation. Preliminary results show the SDWDS can automatically recover more than one Ethernet link failures in the network. However, due to the limitation of the commercial AP, the creation of WDS requires rebooting the AP which costs about 20 seconds recovery latency. In the future, we will introduce virtualization or multi- connectivity techniques to more efficiently support the fault recovery automation in IoTs.