A novel intelligent approach for <scp>man‐in‐the‐middle</scp> attacks detection over internet of things environments based on message queuing telemetry transport

One of the most common attacks is man-in-the-middle (MitM) which, due to its complex behaviour, difficult detect by traditional cyber-attack detection systems. MitM on internet things systems take advantage special features protocols and cause system disruptions, making them invisible legitimate elements. In this work, an intrusion (IDS), where intelligent models can be deployed, approach type attack considering network alterations. Therefore, paper presents a novel method develop model used IDS, being based hybrid process. The first stage process implements feature extraction method, while second one applies different supervised classification techniques, both over message queuing telemetry transport (MQTT) dataset compiled authors in previous works. contribution shows excellent performance for any compared methods. Likewise, best results are obtained using with highest computational cost. Thanks this, functional IDS will able prevent MQTT attacks.