A Review Of The Issues And Challenges In IoT Security Using Machine Learning Techniques

In a typical IoTnetwork, a sensor connects to a controller using a wireless connection. Controllers collect data from sensors and sends the data for storage and analysis[1]. These controllers work with actuators that translate an electrical input to a physical action. The internet of things (IoT), have found application in different areas of human endeavor including healthcare, government, supply chain, cities, manufacturing, etc. and it is estimated that the number of connected devices will reach 50 billion by 2020[2]With the increasing number of devices comes an increase in the the varying number of security threats to the IoT network [3]. To contain these threats, a secure-by-designapproach should be adopted as this will help the IoT devices to anticipate and neutralize the ever changing nature of the threats as against older systems where security was handled as it presents itself [2]This paper x-rays the security challenges in IoT networks and the application of machine learning (Supervised learning, Unsupervised learning and Reinforcement learning) in tackling the security challenges.