Research On Trust-oriented Social-aware D2D Caching Algorithm

With the wide use of mobile applications,the data volume and content diversity in wireless networks are increasing explosively,which brings heavy traffic burden to traditional cellular facilities.How to cope with the massive data demand and guarantee the safe and reliable quality of service has become a severe challenge for the mobile network.Content caching is a popular content distribution technology,which is widely used in wireless networks to alleviate the traffic load.Using content caching in D2 D network can combine the advantages of D2 D communication and content caching,which is significant to improve the performance of cellular network.In cache-enabled D2 D networks,content confidentiality is becoming one of the key issues in the caching process,as any user in the network can be a potential eavesdropper.Indiscriminate distribution of content not only results in content leakage,but also affects the overall performance of the network.Because mobile users have a stable social structure,and social trust is a key factor in building social groups.When only the users who trust each other can share the content,the eavesdropper can be reduced effectively,thus the security and reliability of the content distribution can be guaranteed.At the same time,this kind of social feature between users has a great effect on optimizing cache strategy,not only to overcome the influence of user selfishness,but also to optimize the selection of cache node and transmission path.Therefore,this paper innovatively introduces the concept of social trust into D2 D cache to ensure the security of content distribution by building trust bridges between users,and designs efficient and reliable caching algorithms combined with relevant social features.Firstly,a multi-dimensional trust evaluation mechanism is established by analyzing the history of user interaction and request information.The cooperative capacity,preference similarity and social reciprocity between users were evaluated to obtain multi-dimensional trust relationship.Then,a rough set algorithm based on decision theory is used to extract trusted neighboring users as cache partners.Then,considering the influence of user selfishness on the cache performance,this paper models the cache problem as a trust-oriented cooperative cache game to encourage the user to participate in the cache cooperation.The goal of the optimization is to minimize the total cost of all nodes in the system,and a heuristic cache incentive algorithm is designed to obtain the optimal solution.Experimental results show that the proposed algorithm has better performance than other algorithms in terms of total system cost,cache hit ratio and average request delay.Finally,a D2 D proactive cache algorithm based on deep learning and social trust relation is designed.In this algorithm,the encounter probability between users is evaluated by the trust relationship and similarity between users,and the accuracy of the request prediction is improved by combining the deep learning.A content placement algorithm based on greedy strategy is proposed to maximize the system cache hit ratio.Simulation results show that the proposed proactive cache algorithm has significant advantages over other algorithms in cache hit ratio.