MobCCN: a CCN-compliant protocol for data collection with opportunistic contacts in IoT environments

Abstract

In IoT environments, a significant fraction of services can be expected to be relevant and contextualised to the physical area where data is generated. This is due to the typical strong bond between IoT devices and the physical environment where they are located. Moreover, communication patterns may be largely content-centric rather than device-centric, as interest would be in getting the data, irrespective of where they are generated or stored. In this paper we propose MobCCN, a content-centric network protocol to support data delivery in presence of tiny IoT devices (such as sensors) and users' personal mobile devices (such as smartphones). MobCCN joins together opportunistic networking techniques (to support contextualised, proximity-based communications) and Information Centric Networking (ICN) to support content-centric communication patterns. MobCCN defines a new CCN-like routing and forwarding algorithm, which dynamically builds a gradient-based content-dissemination graph using estimates of the contact rates between nodes, and between nodes and the data that is produced and stored in the network. While preserving compatibility with the standard CCN mechanisms, the MobCCN algorithm makes CCN routing and forwarding suitable for opportunistic networking environments. We have implemented MobCCN in CCN-lite, a de-facto standard lightweight implementation of CCN, which is suitable for resource-constrained devices. Simulation results confirm the feasibility of the proposed approach both in terms of scalability and efficiency.