Scalability Evaluation of Trust and Reputation Models for Wireless Sensor Networks

Abstract

A wireless sensor network consists of a group of distributed sensor nodes to monitor physical and / or environmentalconditions, such as sound, motion, temperature, pressure or pollutants and to cooperatively deliver their data through the networkinfrastructure to a main location, generally called sink. Inspite of its several advantages, due to its inherit features, wireless sensornetworks are open to many security risks. Although conventional data security solutions are effective for other types of networks,they are not effectively applicable to wireless sensor networks. Therefore, trust and reputation management approaches havebeen proposed as an alternative. In a wireless sensor network, trust and reputation management enables a node to make theirown opinion about how trustworthy or reputable another node is. This way, it reduces the opportunities of being defrauded andaugments the probability of a successful transaction. It is known that although in the literature there are many trust and reputationmodels, stil there is a lack of state-of-the-art models, standart data structures and application programming interfaces, generictesting tools, and security threats analysis. Accordingly, in this study a contribution to the last one is made and the effectiveness oftwo well-known trust and reputation models proposed for wireless sensor networks, namely PeerTrust and PowerTrust, is evaluatedin terms of accuracy, path length, and power consumption.