Indian Journal of Science and Technology

Abstract

Background/Objectives: A wireless sensor network, from the large number of nodes (which may reach to thousands of nodes) has been formed. These nodes, each are called a sensor that can sense a special feature of the environment (temperature, humidity, pressure, etc.) and to send to its neighbours.Trust level algorithms are often based on conditions and criteria of current environment in a particular application of wireless sensor networks are defined. The ultimate goal of these algorithms is reach to a network, respondingtofunctional and economic needs of operating environments in wireless sensor networks. Methodology: Each node from two sources is used to obtain trust values of its neighbouring nodes.At firstobtained the known value of direct trust of intended node;next,of indirect trust that by the rest of the common node in the radio range of the two nodes is obtained.Then by combination of direct and indirect trust, a total trust value for each node is calculated. Findings: For simulation, the MATLAB software is used; the number of 30 nodes are at interval of 500*500 meters randomly. A subnet of 5 nodes with node numbers of 10, 12, 15, 16 and 19 are examined to broadcast reports in specified time intervals.Neighbouring nodes view these reports and perform calculations and simulation continues for100 seconds. This simulation both for different scenarios such as: the presence of 5% misbehaving node in the neighbourhood, 10% misbehaving node in the neighbourhood, 20% of misbehaving node in the neighbourhood, being carried out. Applications/Improvements: Since the energy consumption is still low and scalability is also intermediate, according to this proposed method,it did not make changes in other parameters and by keeping them fixed increased the reliability. It is clear that the proposed method is more successful than other methods.

Keywords: Gaussian, Trust and Reputation, Wireless Sensor Networks