An Forwarding Strategy Trust Model Based On Gray Correlation And Collaborative Computing In Opportunistic Network

The environment of Opportunistic network is complex,the existence of selfish nodes and malicious nodes pose a significant threat to network.In order to improve the transmission success rate of nodes in the current harsh environment,reducing the transmission delay.This paper proposes a data forwarding model GCTM(Gray correlation and Collaborative computing Trust Model)based on grey relational and collaborative computation.Data transmission relies on mobile nodes to complete,but in reality,the nature of the node cannot directly identify the nodes whether provide services to the network or attacks the network.Only considering routes to improve the performance of the network has not met the opportunity network demand today,many domestic and foreign scholars put forward a trust model for the complex environment of opportunity network.Due to the current opportunity network is rampant with selfishness and malicious nodes,we need to establish the corresponding trust model in particular to improve the network performance of the node.How to solve the unsafe environment of the opportunity network? we put forward a feasible and effective trust model,which is the focus of the research work in this paper.The relevant work and major contributions of the dissertation are as follows:Firstly,this paper proposes a direct trust model based on grey relational analysis.This trust model aiming at the complex environment in opportunity network.This model analyzes the mobility of nodes and the differences between common nodes and selfish and malicious nodes,and the node characteristics and communication characteristics between the first-level nodes and two-level nodes.The contact density,the delivery credibility and the position closeness are chosen as the multi-dimensional trust parameters of the model,to calculate the direct trust value of the node,and the weighting coefficient of the direct trust value is adjusted by Grey relational algorithm.The static weighting coefficient is improved to automatically adjust the dynamic weighting coefficients of each trust parameter based on the intrinsic relation between the trust parameters,and the accuracy of the direct trust computation of the Opportunity Network trust model is improved.Secondly,we aiming at the attacking characteristics of selfish and malicious nodes in the opportunity network,and the sparse environment,a new filtering algorithm based on collaborative computation is proposed to calculate the recommend trust.Using the communication function of neighbor node,the evaluation node cooperated with the Neighbor node in the collaborative computation.The parameters of the recommended trust are calculated by contact Intimacy and delivery credibility.When the recommended trust module is running,the evaluation node sends a collaborative computation request to its’ neighboring neighbor nodes,and then the neighbor nodes send the recommended information to the evaluation node after the request is received,but because some selfish and malicious nodes exist in the network,it is necessary to filter the recommended information.In this paper,the KNN algorithm is used to filter and the information is merged.The KNN algorithm can effectively identify the false recommendation trust sent by the selfish node and the malicious node,and reduce the interference of the selfish and malicious nodes on the opportunity network.Using collaborative computing technology can makes the calculation result more authoritative.In order to verify the validity of The GCTM,The one simulator is used as the experiment.Comparing analysis of MaxProp,direct Delivery and the MDT trust model based on multidimensional Direct trust,using transmission success rate,transmission delay,routing overhead as the simulation contrast parameters of the experiment,the comparison experiment of different selfishness and the number of malicious nodes is set up in simulation environment.The experimental results show that the GCTM in the relationship with Epidemic,MaxProp,direct Delivery and MDT Trust model has the optimum comprehensive performance.