Model improvement and performance analysis of a fuzzy logic based secured MQTT protocol for IOT devices

Ensuring secure communication among smart IoT devices remains a critical challenge due to the inherent vulnerabilities of lightweight messaging protocols. This study focuses on enhancing the Message Queuing Telemetry Transport (MQTT) protocol by embedding a fuzzy logic–based security mechanism within its publish–subscribe architecture. The proposed Fuzzy-Based Secured MQTT (FBSM) model introduces two lightweight traffic variables the Connection Message Ratio (CMR) and the Acknowledgment Message Ratio (AMR) as inputs for real-time anomaly detection. Unlike earlier approaches that rely on complex machine learning models or static thresholds, the novelty of this work lies in its minimalist fuzzy variable design, real-time deployment on Node-RED, and systematic benchmarking against both MQTT and HTTP protocols. Experimental evaluation using Wireshark and Node-RED demonstrates that the throughput values are 72.68, 73.67, 159.74, and 607.5 bytes per millisecond for FBSM compared to 65.6, 41.52, 21.098, and 17.24 bytes per millisecond for MQTT at message counts of 1, 10, 100, and 1000, respectively. However, from test, the latency of the FBSM system is seen to be 0.07 sec higher than the conventional MQTT as message is been transferred. These results confirm that the proposed model significantly enhances resistance to denial-of-service and congestion-based attacks while maintaining the lightweight advantages of MQTT. The study therefore provides a scalable and practically implementable security enhancement for IoT communication, with strong potential for application in smart homes, healthcare, and industrial systems