Research Of Trust Evaluation Mechanism For Node In Sensor-Cloud Based On Fog Computing

The sensor cloud serves people better by integrating wireless sensor network technology and cloud computing technology to provide users with ideal sensor data conveniently.However,with the increase of the scale,the security problems in the sensor cloud are gradually emerging.On the one hand,the increasing communication capability enables more and more underlying nodes to communicate with users,cloud and other Internet of things systems more closely.On the other hand,increased connectivity also makes the system more vulnerable to network failures and malicious attacks,where a single failure or malicious node can cause the entire system to crash.In order to solve the security problem in the sensor cloud system,this paper proposes to introduce the fog node to conduct trust evaluation for the underlying sensor node.Fog computing was first proposed by cisco in 2011.Compared with the remote cloud center,fog nodes are closer to the underlying nodes,with certain local computing capacity,wider geographical distribution and support mobility.Fog nodes can be used as a bridge between the cloud and the sensor network in the sensing cloud system to help the cloud better manage the underlying sensing network.The introduction of mobile fog nodes needs to address the following three key issues:1.Quality of service of mobile fog nodes.In the process of selecting mobile fog nodes in the cloud,it is difficult for the cloud to obtain the true trust evaluation service quality of the mobile fog node,and thus the cost cannot be determined.At the same time,in the process of trust evaluation of sensing nodes by mobile fog nodes,selfish mobile fog nodes may reduce the service quality of trust evaluation.2.The issue of trust evaluation mechanism.In the existing trust evaluation methods,the cloud is too far away from the sensor network,resulting in the cloud being unable to obtain a fine-grained trust evaluation of each sensor node.At the same time,the sensor node to obtain the trust evaluation of other remote nodes needs to use multihop indirect trust evaluation,which reduces the accuracy of trust evaluation and increases the energy consumption.3.Trust evaluation delivery efficiency.The introduction of mobile fog nodes can effectively improve the accuracy of trust evaluation and reduce the energy consumption of trust evaluation.However,the delay of trust delivery through mobile fog nodes is high and the timeliness is weak.In view of the above three problems,this article proposes the following four solutions.1.For the service quality of mobile fog nodes.The scientific problem is how to make sensor cloud service providers to obtain higher service quality of mobile fog nodes at a lower cost,thereby improving the accuracy of trust evaluation.An auction-based incentive mechanism and a service quality-based incentive mechanism are proposed.The former ensures that the cloud can recruit mobile fog nodes with higher service quality at a lower cost.On the basis of the former,the latter ensures that mobile fog nodes do not reduce the quality of trust evaluation when performing trust evaluation tasks.2.For the issue of trust evaluation mechanism.The scientific problem is how to identify malicious nodes through their various behaviors.A trust evaluation mechanism based on mobile fog nodes is proposed.By comparing current communication behaviors,data collection behaviors of the sensor node with its historical communication behaviors,data collection behaviors,and communication behaviors,data collection behaviors of nodes around the sensor node,the mobile fog nodes are capable to evaluate whether the sensor node is malicious or not.3.On the premise that the size of the sensor network is small and the moving distance of a single mobile fog node is sufficient to cover the entire network.Aiming at the problem of trust evaluation delivery efficiency.The scientific problem is how to effectively reduce the moving distance of the mobile fog node and thereby reduce the trust delivery delay when the mobile fog node establishes direct trust evaluation with all sensor nodes.The paper improves existing algorithm and proposes a optimized mobile fog node algorithm,which considers the minimum dominating set problem and the traveling salesman problem together,and reduces the moving distance of the mobile fog node by increasing the number of relay nodes4.On the premise that the sensor network is large and single or multiple mobile fog nodes can only perform direct trust evaluation on some sensor nodes.Aiming at the problem of trust evaluation delivery efficiency.The scientific problem is how to how to effectively improve the probability of direct trust evaluation and shorten the transmission chain of indirect trust evaluation.A path planning algorithm for single fog nodes and an optimized path planning algorithm for multiple fog nodes are designed.The moving path of fog nodes is constructed based on the number of neighbors-distance-based spanning tree.Compared with existing algorithms,the proposed method can effectively reduce the number of transfer hops based on the trust evaluation of mobile fog nodes.Aiming at the problem that the underlying sensor nodes in the sensor cloud system are unreliable,this paper introduces mobile nodes to form fog computing layer and perform fine-grained trust evaluation.Based on this,some specific issues are studied.The theoretical proof and experiments verify that the proposed scheme can ensure the trust of the underlying sensor nodes,thereby help sensor cloud to obtain correct sensing data,and mine effective information.