FogIoT: Fog Computing based Security Frameworks for Software-defined IoT

Software-defined Networking (SDN) has been implemented in Internet of things (IoT) networks to attain centralized control and improve network management. However, the SDN centralized approach can be compromised by malicious sensor nodes engaging in activities like message dropping or flooding, which can negatively impact the network’s functionality. The security measures in the IoT networks are often based on an idealized trusted environment, which is not practical due to resource constraints and high computational requirements. In this paper, we proposed a lightweight cryptographic approach for detecting potential security threats in wireless fog-enabled SDN-IoT network named as FogIoT. To improve detection of attacks on control plane traffic, the proposed approach examines the credibility of network nodes using reliability metrics and a reliability index. We also propose an acknowledgment-based mechanism to capture reliability and used a weighted averaging approach to determine the confidence index for nodes to ensure credibility. We demonstrated a simulation study to analyze the performance of the lightweight cryptography algorithm and results from simulations show that FogIoT is effective in detecting various types of attacks such as sink-hole, denial of service, and reputation-deceptive.