The internet of things (IoT) is unquestionably one of the most adaptable technologies available today. The IoT is scalable and varied due to the presence of the internet, the expanding capability of network association, and the diversity of connected objects. It has also resulted in the completion of good homes, structures, and even cities over time. The IoT's expanding reality, on the other hand, argues that addressing its potential implications is also necessary. Due to the resource-constrained nature of IoT, an IoT network is vulnerable to security breaches. The Distributed Denial-of-Service (DDoS) attack can result in the removal of network services to users in various ways as a result of leaks, which can result in a crash in important IoT use cases. Our proposed subject encourages the use of SDN and cloud assistance to mitigate DDoS attacks on IoT systems. We've devised a one-of-a-kind mechanism called learning-driven detection mitigation (LEDEM) that identifies DDoS and mitigates it using a semi-supervised machine-learning algorithmic program. We ran LEDEM through its paces in the testbed, simulating topologies, and comparing the results to the progressed solutions. We tend to obtain an increased accuracy rate of 96.28 percent in DDoS attack detection.

Keywords : IoT, DDoS, LEDEM, SDN