Content centric networking (CCN) is a promising architecture for the future networks. In-networking caching of CCN can significantly improve the content delivery efficiency. Content popularity is one of the key factors considered in the design of the caching strategy. However, the existing research ignores the timeliness of content popularity statistics, which makes the changing of popular contents cached in the network lag behind the changing of the user preference. In this paper, a content clustering and popularity prediction based caching strategy (CPC) is proposed to solve this problem. Firstly, the massive contents are divided into different content types using cluster analysis. Then, the popularity of different content types is predicted by the autoregressive integrate moving average (ARIMA) model. Finally, based on the predicted content popularity, the process of the caching placement decision is given. The proposed caching strategy is a distributed caching management method, which can be implemented without centralized controller. Simulation results show that the proposed caching strategy can achieve better performance in terms of cache replacement rate, cache hit ratio and average hop count.
Content Clustering and Popularity Prediction Based Caching Strategy in Content Centric Networking
2017-06-01
218758 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Efficient caching strategy in content‐centric networking for vehicular ad‐hoc network applications
Wiley | 2018
|Efficient caching strategy in content-centric networking for vehicular ad-hoc network applications
IET | 2018
|