Research On Key Technology In Wireless Caching Network

Mobile Internet and virtual reality technology developed rapidly, massive mobile devices accessed the network, and the demand of the user terminal for the data type began diversification, especially the proportion of video content for data types showed increasing trend, which makes mobile data traffic rapid growth, and further increases the demand for communication network bandwidth. By way of continuous cell division and resource reuse traditional cellular communication system enhanced network spectrum utilization and network capacity. With mobile data traffic into the mobile era of big data, these methods can not meet the current needs of the business. With increasing of the mobile terminal computing power and storage space, wireless caching networks based on D2D (Device-to-Device, D2D) communication and cache technology greatly reflects that the cache technology provides a new direction for the development of mobile communication.The wireless caching networks mainly combined D2D communication technology with cache technology, the traditional cellular network in neighboring users can establish a D2D communication link which doesn't go through the base station (Base Station, BS) relay to directly transfer data to each other, thus this way greatly improves the data transmission rate and greatly reduces the transmission time delay. In the auxiliary of terminal cache technology, the user terminal can cache hot video beforehand during the off-peak hours, so the user finding the requested file in the local cache, and the close user's local cache is directly determined by the cache strategy. So it is meaningful for the wireless cache network to major in cache strategy, caching rational content and the effective transmission of cache contents is a key technology for the wireless cache network. In view of the above questions, we studied on caching strategy for different communication scenario in this paper. Firstly, from the cache effective transmission, we researched on the communication strategy based on recommendation algorithm, then, from caching rational content, we studied the cache strategy based on the outage probability; finally optimal cache strategy based on the offloading probability for D2D assisted wireless caching networks was drived by us. This thesis studies the key technology of wireless caching networks as follows:First, under the video network scenario, we propose a communication strategy based on recommendation algorithm. By analyzing the user's historical video playback records to get the user's hobbies, and then calculate the similarity between users where distributed in the range of D2D communication, and finally we introduce collaborative filtering recommendation algorithm to change the user's request behavior. Under this communication strategy, it is observed that the introduction of recommendation and caching techniques can reduce the cost of content providers by simulation results.Second, we study on the caching strategy based on the outage probability. Under the condition of satisfying the minimum target SINR constraint, the outage probability is minimized by activating the D2D link, and then the mathematical modeling of the problem is carried out. When we reduce the complexity of solving the optimization problem, we propose a search algorithm with low complexity compared to traditional sub - gradient algorithm. Simulation results compared with the existing cache strategy,the performance advantage is very obvious. At the same time, it can be proved that our search algorithm is a relatively less complex solution.Third, we study on the caching strategy based on the offloading probability. Firstly,the problem of maximizing network offloading probability is studied under the two-tier heterogeneous network model of wireless caching networks. The solution method based on KKT optimization condition is given, and the simulation results show that how the proposed caching strategy affects the probability, and this way realizes data flow diversion of the cellular network and reduces the network'load by the self-service and D2D offloading.