A Hierarchical Security Framework for Defending Against Sophisticated Attacks on Wireless Sensor Networks in Smart Cities

In smart cities, wireless sensor networks (WSNs) act as a type of core infrastructure that collects data from the city to implement smart services. The security of WSNs is one of the key issues of smart cities. In resource-restrained WSNs, dynamic ongoing or unknown attacks usually steer clear of isolated defense components. Therefore, to resolve this problem, we propose a hierarchical framework based on chance discovery and usage control (UCON) technologies to improve the security of WSNs while still taking the low-complexity and high security requirements of WSNs into account. The features of continuous decision and dynamic attributes in UCON can address ongoing attacks using advanced persistent threat detection. In addition, we use a dynamic adaptive chance discovery mechanism to detect unknown attacks. To design and implement a system using the mechanism described above, a unified framework is proposed in which low-level attack detection with simple rules is performed in sensors, and high-level attack detection with complex rules is performed in sinks and at the base station. Moreover, software-defined networking and network function virtualization technologies are used to perform attack mitigation when either low-level or high-level attacks are detected. An experiment was performed to acquire an attack data set for evaluation. Then, a simulation was created to evaluate the resource consumption and attack detection rate. The results demonstrate the feasibility and efficiency of the proposed scheme.