Subjective Logic-Based In-Network Data Processing for Trust Management in Collocated and Distributed Wireless Sensor Networks

While analyzing an explosive amount of data collected in today's wireless sensor networks (WSNs), the redundant information in the sensed data needs to be handled. In-network data processing is a technique which can eliminate or reduce such redundancy, leading to minimized resource consumption. On the other hand, trust management techniques establish trust relationships among nodes and detect unreliable nodes. In this paper, we propose two novel in-network data processing schemes for trust management in static WSNs. The first scheme targets at networks, where sensor nodes are closely collocated to report the same event. Considering both spatial and temporal correlations, this scheme generates trustworthiness for each sensor node-based solely on the data reported by the sensor. The second scheme is designed for networks, where sensor nodes are randomly distributed to sense ongoing events. This scheme produces node trustworthiness considering the variations in node observations. In addition, we propose an energy saving mode to reduce energy consumption and/or increase the reliability of a network. For trust generation, we employ a subjective logic-based scheme to mitigate trust fluctuations caused by various factors. Extensive simulations based on four scenarios are performed to evaluate the effectiveness of our proposed schemes.