Protocol-Adaptive Strategies for Wireless Mesh Smart City Networks
Lefteris Mamatas, Vassilis Demiroglou, Sarantis Kalafatidis, Sotiris Skaperas, Vassilis Tsaoussidis
Abstract
Wireless mesh networks, especially those typically utilized in smart city deployments for their low-cost and adaptable topologies, are characterized by challenging requirements for communication performance, reliability, as well as adaptability to dynamic network conditions. In this context, Named Data Networking (NDN) introduces a novel packet naming scheme and in-network caching for efficient data retrieval. Although NDN reliability can be damaged by prolonged delays and intermittent connectivity, this impact can be largely canceled by incorporating the Delay-Tolerant-Networking (DTN) paradigm. Hence, we argue that the challenging, dynamic network conditions of smart cities can be handled accordingly by a protocol-adaptive solution that deploys and configures on-demand the most appropriate protocol strategy per node. Software-Defined Networking (SDN) provides the missing features of intelligent centralized control and programmability. Here, we propose REWIRE: an SDN-based protocol-adaptive solution for smart city networking with low-delay communication and reliable interactions. We employ SDN control features, containerized non-IP protocol stacks, clustering and change point (CCP) mechanisms. We also conduct a preliminary investigation of our solution based on real experimentation over two novel smart city testbeds.