SECURED INTERACTION MODEL FOR IOT DEVICES THROUGH MACHINE LEARNING TECHNIQUE

Abstract

The IoT has received a lot of attention ever since its inception. It is now ranked in the top 10 innovations to transform the world. Due to the lack of serious work by all generations, more people are currently engaged in the research of IoT. It has now become applied in our everyday life. Nonetheless, it is possible to maintain the privacy of individuals as one of the factors of IoT applicants. Unless the above issue is addressed effectively, it will greatly affect the advancement of the IoT. Currently, the encryption business is concerned with the advancement of entirely homomorphic technologies. The users can directly influence the symmetric encryption system in order to restore the outputs and decode the result. The outcome is the same as simple text results that are not coded. This article will firstly critically analyze the response to an online privacy problem within the existing IoT, and thereafter is an input to the suggestion that the fully homomorphic methodology be adopted in the internet, to enhance the security of the activities within a network. The interaction model of a privacy and security of an IoT device is developed based on a fully homomorphic method emphasizing a basic architecture and the development of the existing IoT architecture. A theoretical framework of homomorphic encryption is introduced to some security concerns and the method having simple encryption algorithms is enhanced. The outage was not successful by 24 percent with this applied cryptographic privacy technique.