Trust based security model for detecting misbehaving nodes and enhancing routing in an internet of things environment

Authentication in Internet of Things (IoT) is essential due to its unique features which includes sensing, connectivity, intelligence, scalability, self-configuration and an on open dynamic environment. The Internet of Things (IoT) devices must establish a mutual trust without a prior familiarity which might expose them to potential attacks. Therefore, a trust-based solution is crucial for Internet of Things security. This discussion is mainly focused on four different important issues namely Support Vector Machines (SVM), Dynamic Source Routing, Cryptographic Techniques and Bayesian Network Model. This survey first outlines the challenges in Support Vector Machines to detect the malicious nodes. The merits and demerits of support vector machines under this review are also considered. This paper brings a close outlook to detect malicious nodes using support vector machines. Dynamic Source Routing (DSR) is used for an efficient routing from source to destination to enhance trust among IoT nodes. The Cryptographic techniques are used to enhance the security in communication among IoT nodes. The Bayesian networks and random forest to detect the trustworthiness of IoT nodes with a hybrid approach.