Coding-based Cooperative Caching Scheduling Algorithms

With the rapid development of wireless communication technology, widely usage of mobile devices and explosive growthof mobile data, data dissemination strategy becomes a key factorin the followint future of the emerging mobile application development. Data broadcasting can provide mobile support station disseminate data to a large scale of clients, and is widely used in the new mobile application system based on location information services, such as intelligent transportation system. In order to meet the demand of real-time data transportation, the client will store data in the its local cache, and cooperative caching via Peer-to-Peer(P2P) network can further improve the efficiency of the system.Network Coding technology was put forward at the beginning of this century, which principle is to add the coding function on node in the traditional routing protocol, which would increase the data disseminated by a node in a broadcast tip. Network coding can effectively improve the bandwidth utilization and increase the throughput of the system, which raise a heat attention from the Research.Based on C/S and P2 P hybrid cooperative data dissemination architecture framework for server side and client side, we proposed a scheduling algorithm and related cooperative caching strategy research, target for improving the system overall performance(such as bandwidth utilization, service response time and cache hit rate), called MCRCA protocol. In order to further improve the data distribution efficiency and enhance system in large-scale data application environment scalability. We combined network coding and on-demand broadcast, design client grouping strategy in certain coding scheme, and client cache replacement scheme as well. Finally, through the simulation experiments, the effectiveness of the proposed scheme is provided.In this paper, we first defined the relevant variables in the communication process and data transmission constraints, made problem formulation then proved it as NP hard problem, we proposed a scheduling algorithm called Coding-based Maximum Response with Collision Avoidance(CMRCA) and made an example of the running process. In the discussion of cache replacement strategy, we began with analysis in characteristics of traditional caching management algorithms, and illustrated their limitations in the coding-based data broadcast system, put forward "decoding contribution" as evaluation index, and thus the design of matched with the strategy of MCRCA cache replacement strategy.Finally, the CMRCA protocol are carried out to validate therir effiency and capability by means of Visual Studio.Through the analysis of simulation results to verify the method proposed in this paper not only improve the system in the downlink channel bandwidth utilization(client-server) and the cache hit rate, and also reduce the latency.