Internet of Things (IoT) is a concept where the physical world is integrated into the digital world. In IoT, Internet connected devices are able to observe and act. Furthermore, data observations from these devices are linked with cloud based digital services to enable smart decision-making. IoT is seen as the underpinning for smart cities, with multiple services including intelligent traffic management, and effective resource management (e.g. energy or water). The benefits and potential of IoT are clear, resulting in significant uptake. However, numerous factors still hamper its growth. These factors include performance of networks and edge sensing devices, up to developing appropriate applications, all done in a secure and reliable manner. Data communication can typically occur over various networks and protocols, depending on the context of deployment and operation. This paper focuses on validating the Constrained Application Protocol (CoAP) in a low power personal area network to determine if CoAP is an effective application protocol in an IoT environment. Experiments were conducted in the context of the TRESCIMO testbed through which CoAP enabled devices and gateways communicate to IoT platforms. The experiments focused on measuring the efficiency of communication in an environment with poor connectivity. Experimental results show that CoAP is an efficient transport in low signal strength environments.

Introduction

Internet connected devices (with the ability to sense and also actuate) are becoming ubiquitous. Linking these devices with smart, cloud based decision-making services allow for sense-making that can create impact (either financially and/or socially). This fusion of technologies is commonly referred to as the Internet of Things (IoT) [1]. It is projected that the number of Internet devices will number into the billions and also that business stemming from IoT will be billions of dollars [2]. The benefits of IoT are clear. With contextual information acquired from the physical environment, smarter decisions spanning multiple domains can be made through a variety of applications [3]. However, the uptake of IoT has been slow and mostly limited to silos. A number of factors have impacted on the slow uptake. These include the challenge in providing a secure solution, the difficulty in choosing the “right” middleware (e.g. the market is flooded with a variety of middleware platforms, each providing its own approach [4]), and ensuring appropriate end-to-end network connectivity [5]. In a typical IoT architecture data observations are sent via a gateway over the Internet (“north” interface) to an implementation of an IoT middleware technology. From the middleware data is routed to applications. This process is depicted in Figure 1. Observations are often collected from a personal area network on the “south” interface of the gateway. These personal area networks can have low power requirements. The associated sensing device has to either harvest energy or is dependent on batteries. This in turn impacts on the amount of data that can be communicated within a timespan, and the distance the sensing device can be from the gateway as the energy consumption of the communication stack has to be limited, which in turn impacts on the transmission signal strength.