A Multi-Tier MQTT Architecture with Multiple Brokers Based on Fog Computing for Securing Industrial IoT

Abstract

With the rapid growth of internet-connected devices and their resource-constrained capabilities, the current authentication mechanisms are unable to meet the complex IoT application requirements, such as in the Industrial Internet of Things (IIoT), due to the increased computation, communication, and storage overhead arising from these mechanisms. In the IIoT, machine-to-machine (M2M) communication is an underlying technology where devices (e.g., sensors, actuators, and controllers) can be enabled to exchange information autonomously; thus, the massive data generated by these devices can increase latency, network congestion, and the complexity of security management. Message queue telemetry transport (MQTT) is one of the promising M2M protocols used in the IoT that could encounter such issues because it relies on a central broker in the cloud and implements a heavyweight authentication mechanism based on TLS. Therefore, this paper proposes an MQTT architecture with multi-tier brokers based on fog computing, where each broker is deployed with an authentication manager. In addition, the paper presents a lightweight mutual authentication scheme based on hash function and XOR operation. Comparing the results given in the benchmark, the overall performance of our scheme shows that storage and communication overheads are reduced to 89% and 23%, respectively. Furthermore, our system can resist against several cyberattacks and provide scalability.

Security analysis of fog computing environment for ensuring the security and privacy of information

“…Multitier broker‐based message queue telemetry transport (MQTT) (E6) was designed by, 47 in which lightweight authentication was introduced. The broker was designed at the authentication manager using simple XOR and hash functions.…”

“…Authentication and key agreement : The role of fog nodes is data gathering and aggregation, in which acquiring authentication with various levels of the gateways creates security issues. Hence, an end‐to‐end authentication approach and lightweight authentication techniques are considered significant criteria for acquiring secure information sharing in fog computing 40–49 …”

“…Distributed architecture : The decentralized nature of fog devices makes security issues more challenging because many edge devices utilize the same code for information sharing. Thus, the redundancy of a system model must be minimized to enhance the security of a fog environment 42,45,47–49,51 …”

“…Thus, the authentication and access control mechanism is essential for fog computing to accomplish secure information sharing and processing. Hence, the security of a model is assured by providing the appropriate authentication 45,47,58,65 …”

“…The analysis based on real or random was employed for the formal security, and informal analysis based on the attacks was performed to depict the resilience of a model. Multitier broker-based message queue telemetry transport (MQTT) (E6) was designed by, 47 in which lightweight authentication was introduced. The broker was designed at the authentication manager using simple XOR and hash functions.…”