Collusion-resilient quality of information evaluation based on information provenance

The quality of information is crucial for decision making in many dynamic information sharing environments such as sensor and tactical networks. Information trustworthiness is an essential parameter in assessing the information quality. In this paper, we present a trust model to evaluate the trustworthiness of information as well as information publishing entities based on information provenance. In our trust model, decision makers can give an evaluation on the information they receive and further adjust the evaluation result to a more accurate value by considering two factors: information similarity and path difference. We introduce Collusion Attacks that may bias the computation and present a mechanism to detect and reduce the effect of Collusion Attacks. Based on the final adjusted information trust, feedback is given to the information publishing nodes to adaptively update their trust scores. Therefore, our collusion-resistant scheme can dynamically assess the trustworthiness of information as well as participating entities in a network and thus effectively enhance the network security. Detailed analysis of the proposed approach is presented along with simulation results.