Research On Content Caching Strategy In Fog Radio Access Network

With the rapid development of mobile communication and the sharp increase of smart mobile devices,wireless data traffic has experienced explosive growth in recent years,thus injecting tremendous traffic into the network.Fog Radio Access Network(FRAN)is proposed as a promising wireless network architecture to accommodate the fastgrowing data traffic and improve the performance of network service.By deploying content caching in F-RAN,fast and repeatable data access can be achieved,which reduces network traffic and transmission latency.Due to the capacity limit of caches,it is essential to predict the popularity of the content and pre-cache them in edge nodes.The thesis focuses on the problem of edge caching based on content popularity prediction in F-RAN.In the thesis,the problem of joint storage resource allocation and content placement in F-RAN is studied.Firstly,we propose an intelligent F-RANs framework based on federated learning(FL)for content popularity prediction,which does not require gathering user data centrally on the server for training and it can effectively ensure the privacy of users.FL is applied to content popularity prediction,which can accurately predict the content popularity distribution in the network.Secondly,in order to maximize resource utilization,a traffic-based allocation algorithm is proposed to maximize the utilization of the storage resources of the fog server.Finally,to minimize the cost of network traffic,two heuristic algorithms are proposed to address the problem of content placement,so as to pre-cache the content required by users in the associated fog server.Theoretical analysis and simulation results show that the proposed algorithm has lower computational complexity and can effectively reduce the cost of network traffic.Considering the heterogeneity of terminals resources and the security of data transmission,a system that combines IoT equipment,fog computing,remote cloud and blockchain is investigated in the thesis.In the system,a blockchain-assisted active content caching scheme based on federated learning is proposed.Firstly,considering the resource constraints of user equipment in the real network,a federated learning scheme based on user selection is presented to effectively improve the training process of federated learning and reduce the time for model training.Secondly,a context-aware adversarial autoencoder is used to predict the highly dynamic content popularity.Finally,in order to ensure the security of transmitted data,we have incorporated blockchain technology in the algorithm and design four smart contracts for decentralized entities to record and verify the transactions to ensure the security of data.Simulation results show that the proposed scheme can effectively improve the cache hit ratio.