A Trust With Abstract Information Verified Routing Scheme for Cyber-Physical Network.

Cyber-physical network (CPN) as well as cyber-physical social computing and networking (CPSCN), as the next generation of computing and networking, utilize a wide variety of sensing devices with significant sensing capabilities. However, it face huge challenge that data packet is transmitted to destination security. Data packets can be dropped by malicious nodes in CPN/CPSCN, causing serious damage to the network. Thus, ensuring the security of data transmission is a crucial issue for wireless sensor networks which is one of the important component of CPN/CPSCN. In this paper, a trust with abstract information verified (TAIV) routing scheme is proposed to improve the data packet arrival ratio. In the TAIV scheme, backbone routing paths are created by constructing an optimal connected dominating set (CDS). CDS routing ensures communication among all nodes in the network. In addition, as the source node sends the data packet to the destination through the backbone routing path, the abstract information is also transmitted to the destination through the auxiliary routing path to allow the destination node to know whether data has been sent to it. Depending on whether the data packet is transmitted to the destination or dropped by the malicious node and whether the abstract message is transmitted to the destination successfully, the destination will reduce or increase the trust in nodes in this routing path accordingly. After a period of time, the trust information of nodes will be diffused by information exchange. If the trust information is sufficient, the malicious node, whose trust is lower than that of the normal node, will be identified and eliminated from the backbone routing path, reducing the number of malicious nodes in the backbone. Thus, the data packet arrival ratio will be improved. The theoretical analysis shows that compared with the previous scheme, the data packet arrival ratio increases by 4.251%-80.208%, and nearly all malicious nodes are identified while retaining long network lifetime.