The world is rapidly getting connected. Commonplace everyday things are providing and consuming software services exposed by other things and service providers. A mash up of such services extends the reach of the current Internet to potentially resource constrained "Things", constituting what is being referred to as the Internet of Things (IoT). IoT is finding applications in various fields like Smart Cities, Smart Grids, Smart Transportation, e-health and e-governance. The complexity of developing IoT solutions arise from the diversity right from device capability all the way to the business requirements. In this paper we focus primarily on the security issues related to design challenges in IoT applications and present an end-to-end security framework.

Narrowband Internet of Things (NB-IoT) is gaining prominence as a key Low Power Wide Area Network (LPWAN) technology for IoT applications. Since it operates on licensed frequency spectrum it can provide guarantees to applications demanding Quality of Service (QoS). NB-IoT has emerged as a competitive rival for other LPWAN technologies such as LoRa and Sigfox, which work in the unlicensed frequency spectrum and are vulnerable to interference. Therefore, NB-IoT is the trivial fit for industries and other business companies that demand guaranteed services. In this paper the different features of the NB-IoT technology have been studied on the commercial Orange network in Belgium using the ublox SARA-N210 module [1] as the user equipment (UE). We focused on the device and network performance in terms of setup times, signal quality, throughput, latency, and reliability and studied the network dynamicity on signal strength. These observations are then compared with the theoretical defined limits of NB-IoT.

The wireless Internet of Things (IoT) landscape is quite diverse. For instance, Low-Power Wide-Area Network (LPWAN) technologies offer low data rate communication over long distance, whereas Wireless Personal Area Network (WPAN) technologies can reach higher data rates, but with a reduced range. For simple IoT applications, communication requirements can be fulfilled by a single technology. However, the requirements of more demanding IoT use cases can vary over time and with the type of data being exchanged. This is pushing the design towards multimodal approaches, where different wireless IoT technologies are combined and the most appropriate one is used as per the need. This paper considers the combination of Narrow Band IoT (NB-IoT) and Bluetooth Low Energy (BLE) as communication options for an IoT device that is running a Lightweight Machine to Machine/Constrained Application Protocol (LwM2M/CoAP) protocol stack. It analyses the challenges incurred by different protocol stack options, such as different transfer modes (IP versus non-IP), the use of Static Context Header Compression (SCHC) techniques, and Datagram Transport Layer Security (DTLS) security modes, and discusses the impact of handover between both communication technologies. A suitable end-to-end architecture for the targeted multimodal communication is presented. Using a prototype implementation of this architecture, an in-depth assessment of handover and its resulting latency is performed.

Bluetooth Low Energy (BLE) is a widely known short-range wireless technology used for various Internet of Things (IoT) applications. Recently, with the introduction of BLE mesh networks, this short-range barrier of BLE has been overcome. However, the added advantage of an extended range can come at the cost of a lower performance of these networks in terms of latency, throughput and reliability, as the core operation of BLE mesh is based on advertising and packet flooding. Hence, efficient management of the system is required to achieve a good performance of these networks and a smoother functioning in dense scenarios. As the number of configuration points in a standard mesh network is limited, this paper describes a novel set of standard compliant Quality of Service (QoS) extensions for BLE mesh networks. The resulting QoS features enable better traffic management in the mesh network, providing sufficient redundancy to achieve reliability whilst avoiding unnecessary packet flooding to reduce collisions, as well as the prioritization of certain traffic flows and the ability to control end-to-end latencies. The QoS-based system has been implemented and validated in a small-scale BLE mesh network and compared against a setup without any QoS support. The assessment in a small-scale test setup confirms that applying our QoS features can enhance these types of non-scheduled and random access networks in a significant way.

Current trend is being extended from the traditional Internet to the small, cheap, and low-power Internet of Things (IoT) in which the objects are being equipped with a device having computation and communication capabilities. As a result, all these objects can be connected to the Internet and have the capability to communicate among each other. This connection infrastructure among the objects would face different types of malicious attacks. Hence securing these objects is a primary goal. There are a lot of security mechanisms available today, but most of them are quite heavy in terms of computation and communication. As the IoT objects have very limited resources and mostly run on battery power, it is difficult to embed intensive computations on these resource-constrained devices. Datagram Transport Layer Security (DTLS) protocol has been standardized to work in cohesion with the CoAP protocol to provide security. But DTLS does not fit well for multicasting, though it is a quite common need for IoT environments. Indeed there are some adaptations for DTLS protocol to function in a multicast environment, but it consumes much communication and computation resources. We propose a mechanism called S-CPABE (Segregated Ciphertext Policy Attribute-Based Encryption) based on CPABE, particularly targeting the multicast needs and tailoring to the IoT framework. The novelty of S-CPABE lies on providing equivalent security as CPABE with reduced resource requirements at the low-power end devices. This mechanism perfectly meets the needs for secure multicast in an IoT environment and consumes much less resources as compared to DTLS.

A multitude of wireless communication technologies are available for building Internet of Things (IoT) solutions, ranging from Low Power Wide Area Network (LPWAN) technologies with large coverage and low data rates to Wireless Personal Area Network (WPAN) technologies with shorter range but higher data rates. To satisfy the requirements of more challenging and dynamic IoT use cases, a combination of different technologies might be the best choice. Leveraging on a communication architecture that supports such multi-modal IoT systems, this paper studies the energy efficiency of a prototype device that is able to switch between Narrowband Internet of Things (NB-IoT), an LPWAN technology, and Bluetooth Low Energy (BLE), a WPAN technology. Starting from the energy consumption of the individual technologies, the overall consumption of the multi-modal device is assessed, including the effects of switching, and a prediction of the battery lifetime is made for different application contexts. Our analysis shows that for energy-critical applications, BLE is preferable over NB-IoT wherever possible. However, in case BLE is unavailable, a switch to NB-IoT can be initiated. Based on our findings, a number of approaches to use the multi-modal system in an energy efficient way has been proposed.