Content Retrieval At the Edge: A Social-aware and Named Data Cooperative Framework

TL;DR

This paper introduces sNDN, a social-aware named data framework for efficient content retrieval at the network edge, significantly reducing cellular traffic through device cooperation based on social and content similarities.

Contribution

It proposes a novel social-aware NDN framework, sNDN, utilizing Friendship Circles for cooperative content retrieval in mobile networks, improving efficiency over existing schemes.

Findings

sNDN greatly reduces cellular capacity usage

Outperforms other content retrieval schemes

Effective in leveraging social and content similarities

Abstract

Recent years with the popularity of mobile devices have witnessed an explosive growth of mobile multimedia contents which dominate more than 50\% of mobile data traffic. This significant growth poses a severe challenge for future cellular networks. As a promising approach to overcome the challenge, we advocate Content Retrieval At the Edge, a content-centric cooperative service paradigm via device-to-device (D2D) communications to reduce cellular traffic volume in mobile networks. By leveraging the Named Data Networking (NDN) principle, we propose sNDN, a social-aware named data framework to achieve efficient cooperative content retrieval. Specifically, sNDN introduces Friendship Circle by grouping a user with her close friends of both high mobility similarity and high content similarity. We construct NDN routing tables conditioned on Friendship Circle encounter frequency to navigate a content request and a content reply packet between Friendship Circles, and leverage social properties in Friendship Circle to search for the final target as inner-Friendship Circle routing. The evaluation results demonstrate that sNDN can save cellular capacity greatly and outperform other content retrieval schemes significantly.