Research On Cache Strategy Of Mobile Edge Network

With the increase of smart mobile applications and wireless network services,traditional network equipment and functions can't meet the rapidly increasing demand for data services and users demands for high-bandwidth and low-latency service experience quality.In order to continuously improve the quality of network functions and services,at the same times,meet various immediate needs of users,the research on mobile edge caching technology has attracted the attention of a large number of domestic and foreign researchers.Due to the limited cache resources on the mobile edge side,how to effectively use the edge side cache resources,coordinate the edge side cache resources,improve the cache hit rate at the edge,and improve the users experience quality has become a key issue of the edge cache technology.Therefore,this thesis focuses on the mobile edge cache placement and cache update strategies.The main contents are as follows:First,considering that the current video traffic has strong timeliness and obvious personalization characteristics.The thesis combines the average popularity of hot content and users interest to propose an edge caching strategy based on social relationship perception.First of all,estimate users need based on the average popularity of hot content and users interest,and reflect the strength of the users' s social relationship with the similarity of users need.Then,select the users device as the auxiliary cache location based on the strength of the users' s social relationship.Finally,comprehensively consider the cache location selection and users demand selection,design a joint caching strategy for base station cooperative caching and user-assisted caching.Simulation results show that the proposed cache placement strategy has a certain degree of improvement in cache hit rate and a decrease system cache delay compared to the comparison cache strategy.Second,with the development of smart terminals,user-generated video data has increased year by year,which greatly increases the randomness,dynamics,and uncertainty of video data,which will lead to untimely cache updates and the problem of premature deletion of popular content.In response to the above problems,the thesis proposes a cache update strategy based on regional popularity perception.First of all,it explores the social relationship between users,borrows the information dissemination model theory,and regards the regional popularity and dissemination of videos as a virus dissemination process.Then,according to the flashpoints and extinction points of information dissemination,popular content is selected for caching in time,and obsolete content is deleted.The simulation results show that the proposed cache update strategy has a higher cache hit rate and lower cache energy consumption than the comparison scheme.