Profiling IoT devices for unusual responses of operations using machine learning techniques

4^th International Conference and Business Expo on Wireless, Telecommunication & IoT

July 19-20, 2018 | London, UK

Khusanbek Gafurov and Chung Tai Myoung

Sungkyunkwan University, South Korea

Posters & Accepted Abstracts: Int J Sens Netw Data Commun

Abstract :

There will be more IoT devices close to 50 billion in 2020. One way it rises various opportunities for society, businesses and alike on the other hand more research work for engineers to optimize IoT devices and provide ensured security. IoT devices use new protocols and services that have not tested or used in the practice widely and because of separation of IoT objects, various issues may arise. For this reason, new solutions for preventing device out of malfunctioning, making it reliable and efficient are required. In this paper, we want to add a contribution to the security of IoT devices. While sensors generate large amount of data and send it to processing unit data can be modified signal injection which may result improper functioning of the peripherals. In order to prevent this we used existing dataset and profile the features of unusual responses of operations using machine learning techniques. In further work, we will explore new methods of analyzing data modification with other techniques of machine learning or beyond.

Biography :

E-mail: x.gafurov@gmail.com