Mobile distributed caching(MDC)as an emerging technology has drawn attentions for its ability to shorten the distance between users and data in the wireless network.However,the DC network state in the existing work is...Mobile distributed caching(MDC)as an emerging technology has drawn attentions for its ability to shorten the distance between users and data in the wireless network.However,the DC network state in the existing work is always assumed to be either static or real-time updated.To be more realistic,a periodically updated wireless network using maximum distance separable(MDS)-coded DC is studied,in each period of which the devices may arrive and leave.For the efficient optimization of the system with large scale,this work proposes a blockchain-based cooperative deep reinforcement learning(DRL)approach,which enhances the efficiency of learning by cooperating and guarantees the security in cooperation by the practical Byzantine fault tolerance(PBFT)-based blockchain mechanism.Numerical results are presented,and it illustrates that the proposed scheme can dramatically reduce the total file download delay in DC network under the guarantee of security and efficiency.展开更多
Information-Centric Networking(ICN), an alternative architecture to the current Internet infrastructure, focuses on the distribution and retrieval of content by employing caches in a network to reduce network traffic....Information-Centric Networking(ICN), an alternative architecture to the current Internet infrastructure, focuses on the distribution and retrieval of content by employing caches in a network to reduce network traffic. The employment of caches may be accomplished using graph-based and content-based criteria such as the position of a node in a network and content popularity. The contribution of this paper lies on the characterization of content popularity for on-path in-network caching. To this end, four dynamic approaches for identifying content popularity are evaluated via simulations. Content popularity may be determined per chunk or per object, calculated by the number of requests for a content against the sum of requests or the maximum number of requests. Based on the results, chunk-based approaches provide 23% more accurate content popularity calculations than object-based approaches. In addition, approaches that are based on the comparison of a content against the maximum number of requests have been shown to be more accurate than the alternatives.展开更多
基金Supported by the National Natural Science Foundation of China(No.61571021,61901011)the Program of China Scholarship Council(No.201806540039).
文摘Mobile distributed caching(MDC)as an emerging technology has drawn attentions for its ability to shorten the distance between users and data in the wireless network.However,the DC network state in the existing work is always assumed to be either static or real-time updated.To be more realistic,a periodically updated wireless network using maximum distance separable(MDS)-coded DC is studied,in each period of which the devices may arrive and leave.For the efficient optimization of the system with large scale,this work proposes a blockchain-based cooperative deep reinforcement learning(DRL)approach,which enhances the efficiency of learning by cooperating and guarantees the security in cooperation by the practical Byzantine fault tolerance(PBFT)-based blockchain mechanism.Numerical results are presented,and it illustrates that the proposed scheme can dramatically reduce the total file download delay in DC network under the guarantee of security and efficiency.
基金funded by the Higher Education Authority (HEA)co-funded under the European Regional Development Fund (ERDF)
文摘Information-Centric Networking(ICN), an alternative architecture to the current Internet infrastructure, focuses on the distribution and retrieval of content by employing caches in a network to reduce network traffic. The employment of caches may be accomplished using graph-based and content-based criteria such as the position of a node in a network and content popularity. The contribution of this paper lies on the characterization of content popularity for on-path in-network caching. To this end, four dynamic approaches for identifying content popularity are evaluated via simulations. Content popularity may be determined per chunk or per object, calculated by the number of requests for a content against the sum of requests or the maximum number of requests. Based on the results, chunk-based approaches provide 23% more accurate content popularity calculations than object-based approaches. In addition, approaches that are based on the comparison of a content against the maximum number of requests have been shown to be more accurate than the alternatives.
