Internet of Things via Named Data Networking: The Support of Push Traffic Marica Amadeo, Claudia Campolo, Antonella Molinaro University Mediterranea of Reggio Calabria - DIIES Department, e-mail: {name.surname}@unirc.it Abstract—Named Data Networking (NDN) is a promising information-centric future Internet architecture. Its innovative concepts, like named content, name-based routing and in-network caching, particularly suit the requirements of the Internet of Things (IoT), interconnecting billions of heterogeneous con- strained objects. NDN supports receiver-driven communications and can easily enable IoT pull-based applications, e.g., to monitor sensors and to control actuators. However, NDN cannot natively allow devices to transmit unsolicited data, like alarms or status changes, unless to properly modify the semantics of exchanged packets and the forwarding strategy. In this paper we discuss how NDN can support reliable push-based IoT traffic, through the definition of three schemes conceived to be as much compliant as possible with the NDN tenets, while targeting efficiency in the usage of network and devices resources. A simple analytical framework provides preliminary quantitative insights into the proposed solutions and guidelines for future design. Index Terms—Named Data Networking, Internet of Things, data pushing, reliability I. I NTRODUCTION According to the Internet of Things (IoT) vision, tiny wireless devices will interact with the physical surrounding environment and with the Internet to support a wide range of services, like smart building automation, healthcare, smart energy management, which improve the quality of life of users. A suite of protocols has been proposed to bring IP, the standard de facto for computer communications, to the do- main of wireless resource-constrained devices, thus facilitating interoperability between entities in different domains [1]. At the same time, the research community is exploring cutting- edge and clean-slate approaches for the future Internet, where data retrieval is based on location-independent, unique, content names instead of IP addresses [2]. In the Information Centric Networking (ICN) model, data consumers specify what they search for and not where they expect it to be provided. Among several ICN proposals, Named Data Networking (NDN) [3] has the potential to become a key technology for data dissemination in the Future Internet. It leverages robust connection-less communications based on the exchange of two packet types: the Interest, to solicit the transmission of the content, and the Data, to actually transfer the content. In this paper, we consider NDN as a communication infras- tructure for IoT. NDN features natively bring several benefits to IoT systems in terms of easy and scalable data access, energy efficiency, security and mobility support [4]. However, IoT is also a very challenging and heterogeneous environment, especially if compared to the wired Internet. First, unlike high-performing routers in the core network, IoT devices are typically resource-constrained and data transfer performance may be very poor due to the unreliable wireless medium. Second, applications typically foresee the pull of small data (e.g., the temperature in a room requested by a heating system) instead of large file transfers and they can trigger a device to perform some tasks (e.g., to switch on a home appliance). In addition, push-based data transmission may be originated for periodic monitoring services (e.g., vehicles transmitting congestion updates) or for alarm propagation mechanisms. NDN natively supports pull traffic and there is always a 1-to-1 matching between Interest and Data. Vice versa, data pushing requires some modifications in the NDN service model and the forwarding fabric. In this context, the contributions of the paper are manifold. First, we focus on reliable push-based data delivery and present three mechanisms to support it via NDN, leveraging new packets and/or modifications in their native semantics. Their points of strength and weaknesses are discussed in terms of compliance with the NDN tenets, suitability for different applications and scope domains. Then, a simple analytical model is derived to characterize the behaviour of the proposed mechanisms, which accounts for the main parameters affecting the efficiency in network and devices usage. Indeed, the targeted IoT environments will encompass a high number of devices that periodically wake up for brief periods of time to transmit data, and it is crucial to keep the network overhead as small as possible, while reducing the time during which the battery-powered devices should be active. The remainder of the paper is organized as follows. Section II discusses NDN and related work. The proposed reliable pushing schemes are described in Section III, whereas Sec- tion IV presents the analytical model and results. Section V concludes the paper. II. NAMED DATA NETWORKING AND I OT NDN employs hierarchical URI-like content names that are carried in Interest and Data packets. When a NDN node receives an Interest, it first searches for a name prefix longest- match in its Content Store (CS); if a match is found, then the node sends the Data back. Otherwise, if there is a matching entry in the Pending Interest Table (PIT), which keeps track of the forwarded Interest(s) that are not yet satisfied, the Interest is discarded because an equal request has already been forwarded. Otherwise, a new PIT entry is created and the Interest is further forwarded to the interface stored in the Forwarding Information Base (FIB). The Data packet follows