Sensor search using clustering technique in a massive IoT environment

Abstract

The Internet of things contains billions of interconnected things such as physical objects, animals, or human beings that give the power to transfer information over a network without human interaction. The huge amount of data continuously generated by these sensing devices raises the challenge of searching for the most relevant sensor data according to user’s query. Also, the queries specified by users are in natural human language which cannot be processed by sensors. To handle these challenges, devices in an IoT environment are grouped together to form clusters which reduces the search space. Every device is given a unique ipv6 address to identify it within the network. Then, the user’s abstract query is transformed into low level form which can be recognized by sensors. Experimental results show that the search time and response time is improved using this approach in large scale IoT environments.