Validating sensor nodes in Wireless sensor networks using scoring algorithm

Abstract

Sensor networks are frequently used to collect data in the environment such as agriculture, forest monitoring, healthcare, and military battlefield. In Wireless Sensor Networks (WSN), nodes are used to monitor the environment and gather data where sinks can be used to collect the data from the sensor nodes and transfer them to the back-end server for processing. These sensible data are moved from one node to another node in the network. Such data should not be considered for public accessibility by the nodes in the network where the visibility and ease of access can only be achieved through either authenticated nodes or right authenticated persons. As sensor node can collect an important data (such as medical or military data), security is a critical issue. Hence, the sensor network needs a secure authentication mechanism to solve this problem and protects the unauthorized access. Therefore, the authentication mechanism used by the node and the sink node must be very efficient in terms of both computational time and energy consumptions. This is especially important for nodes with computing capabilities and battery lifetime is very low. Moreover, for extremely lightweight devices, efficient security solutions with simple mathematics operation and low energy consumptions are still required. To make an authentication decision in real-time, a scoring algorithm examines the user model and the user’s recent behavior, and outputs a score indicating the likelihood that the correct user is using the device. The score is used to make an authentication decision.