Caching in Named Data Networking for the wireless Internet of Things

Abstract

Named Data Networking (NDN) is a promising Information-Centric future Internet architecture. Besides its recognized potentialities as a content retrieval solution in wired domains, NDN has been also recently considered as an enabling technology for the Internet of Things (IoT), thanks to its innovative features like named-based routing and in-network caching. In particular, the possibility of caching at intermediate nodes can be especially useful to reduce the retrieval delay, and limit the network traffic and the load on data producer. However, unlike traditional Internet contents, IoT data are typically transient and periodically refreshed by the producer. At the same time, unlike Internet routers, IoT devices can be resource-constrained, with limitations in terms of energy, storage and processing capabilities. Therefore, caching algorithms designed for (intransient) Internet traffic and Internet routers do not well suit IoT domains. In this paper, we consider a wireless NDN-IoT network and propose a novel distributed probabilistic caching strategy that relies on the freshness of data and on potentially constrained capabilities of devices (energy level and storage capacity). The proposed solution has been evaluated through simulations with ndnSIM and results show that it outperforms traditional NDN caching mechanisms in terms of data retrieval and network energy efficiency.