Research on Satellite Network Caching Strategy Based on Content Popularity
Aiming at the problem that the traditional satellite network caching strategy only considers the global content popularity,and does not take into account the large differences in popular content in different regions,so that the high popularity content cannot be accurately found and placed,resulting in a large user acquisition delay.A satel-lite network caching strategy based on a mixture of global popularity prediction and local popularity calculation is pro-posed.This strategy uses the LSTM model to predict the popularity of global content,and calculate the popularity of local content,partition satellite cache nodes,define the partition scale factor of cache nodes,cache global and local popular content respectively,minimize the average acquisition delay of user content as the optimization goal,and solve the cache placement problem by using BPSO.Simulation results show that compared with the traditional random cache placement and the most popular cache placement,the proposed strategy can reduce the user content acquisition latency and improve the cache hit ratio.
NETL
NSTL
万方数据
