A trustworthiness-enhanced reliable forwarding scheme in mobile Internet of Things

Abstract

In intelligent Internet of Things (IoT), finding a trustworthy route that has higher delivery ratio and lower transmission latency is a key challenge of data forwarding for mobile applications. This paper proposes a trustworthiness-enhanced reliable forwarding (TERF) scheme for mobile IoT to minimize the disadvantages of selfish or malicious nodes to data transmission. TERF is built on a dual trustworthiness framework that consists of the local and global trustworthiness among nodes. First, mobile IoT network is modeled as a weighted directed graph that changes over time. Second, based on the dynamic graph, TERF redefines the contact probability and the service degree, and employs them to measure the local trustworthiness between nodes. The former reflects one node's familiarity with another node, and the latter helps reduce the interference from malicious nodes. They can improve the local trustworthiness prominently. Third, to refrain from one-sidedness of the local trustworthiness, TERF considers the role of nodes in the entire networks, and constructs the global trustworthiness between nodes based on the personal centrality and the social similarity. The social similarity reflects the associations between mobile device nodes, while the personal centrality reflects the relative importance of nodes and can avoid the inefficiency from selfish nodes. More importantly, TERF uses the dot product of the local trustworthiness vectors between two nodes to compute the social similarity, and employs the nodes' degree and the local trustworthiness to calculate the personal centrality. Finally, the TERF algorithm is developed. The experimental results prove that TERF has high stability, and outperforms SimBet, PROPHET and Bubble Rap with respect to message delivery ratio, average latency, average hop-count distribution and the network cost.