In Reputation We Believe: Query Processing in Mobile Ad-Hoc Networks

Abstract

Current research on data management in mobile ad-hoc networks focuses on discovering sources and acquiring information. In this mode of operation, mobile devices assume answers to be correct and do not verify the veracity of the information or their providers. This assumption is suitable for most client-server environments; however, peer-to-peer environments lack the intrinsic stability of ``anchored'' sources. In mobile peer-to-peer environments, some sources may provide faulty information, which can lead to incorrect conclusions. Consequently, devices need a mechanism to evaluate the integrity of their peers and the accuracy of peer provided information. To address this problem we propose a query processing model that relies on distributed trust and belief. In our model, each device maintains and shares beliefs regarding the degree of trust it has for its peers -- where trust is determined by experience and reputation. Additionally, each device associates a value indicating its belief in the accuracy of the information the device holds. Each device, when querying its peers, uses the trust it has placed in the peers, in conjunction with the peers' accuracy belief of their information, to determine the reliability of the responses to its query. We have implemented our model in GloMoSim and provide experimental results for different combinations of trust and accuracy algorithms.