Smart Collaborative Caching for Information-Centric IoT in Fog Computing

The significant changes enabled by the fog computing had demonstrated that Internet of Things (IoT) urgently needs more evolutional reforms. Limited by the inflexible design philosophy; the traditional structure of a network is hard to meet the latest demands. However, Information-Centric Networking...

Full description

Bibliographic Details
Main Authors: Fei Song, Zheng-Yang Ai, Jun-Jie Li, Giovanni Pau, Mario Collotta, Ilsun You, Hong-Ke Zhang
Format: Article
Language:English
Published: MDPI AG 2017-11-01
Series:Sensors
Subjects:
IoT
ICN
Online Access:https://www.mdpi.com/1424-8220/17/11/2512
Description
Summary:The significant changes enabled by the fog computing had demonstrated that Internet of Things (IoT) urgently needs more evolutional reforms. Limited by the inflexible design philosophy; the traditional structure of a network is hard to meet the latest demands. However, Information-Centric Networking (ICN) is a promising option to bridge and cover these enormous gaps. In this paper, a Smart Collaborative Caching (SCC) scheme is established by leveraging high-level ICN principles for IoT within fog computing paradigm. The proposed solution is supposed to be utilized in resource pooling, content storing, node locating and other related situations. By investigating the available characteristics of ICN, some challenges of such combination are reviewed in depth. The details of building SCC, including basic model and advanced algorithms, are presented based on theoretical analysis and simplified examples. The validation focuses on two typical scenarios: simple status inquiry and complex content sharing. The number of clusters, packet loss probability and other parameters are also considered. The analytical results demonstrate that the performance of our scheme, regarding total packet number and average transmission latency, can outperform that of the original ones. We expect that the SCC will contribute an efficient solution to the related studies.
ISSN:1424-8220