Interoperability, Safety and Security in IoT

Opportunistic or device-to-device communications offer a great chance for straight-forward and cost-effective interoperability among various devices and manufacturers, from tiny sensors to end-used smartphones. However, their implementation is not trivial, as no standard communication technologies exist for their purposes. This paper explores the available options, qualitatively compares their properties, focusing especially on power consumption and user friendliness. We also offer an experimental comparison of their energy consumption and discuss further needed developments.

Interoperability and conformance testing are needed to ensure that systems behave as specified by the standards they implement. Today, interoperability testing is done through face-to-face “interop events”. Requiring physical presence of all parties impacts the scalability of the testing, and slows down the development of standards-based products.

In scenarios involving point-to-multipoint network traffic, transmitting to each destination individually with unicast may lead to poor utilisation of network bandwidth, excessive energy consumption caused by the high number of packets and suffers from low scalability as the number of destinations increases. An alternative approach, would be to use network-layer multicast, where packets are transmitted to multiple destinations simultaneously. In doing so, applications adopting a one-to-many communication paradigm may improve their energy efficiency and bandwidth utilisation. In this paper, we present Bi-directional Multicast Forwarding Algorithm (BMFA), a novel RPL-based multicast forwarding mechanism. BMFA improves its pre-predecessor SMRF in that it allows multicast traffic to travel both upwards as well as downwards in an RPL tree. At the same time, it retains SMRF’s low latency and very low energy consumption characteristics. Our performance evaluation results, conducted using the Contiki operating system, show that BMFA outperforms its rival Trickle Multicast/Multicast Protocol for Low power and Lossy Networks (TM/MPL), in terms of reducing both delay and energy consumption.

In this paper we argue that synchronization abstractions could be used as the glue that tie together the interactions between ‘things’ in an IoT environment. We also support that this is analog to what is used in distributed multimedia applications. Using this argument, we propose in this paper that IoT solutions, protocols and applications should benefit from standardized multimedia tools like specification languages and corresponding middleware support platforms as a means for harmonization and interoperability. Additionally, we extend our recent contributions in favor of a separation of concerns in multimedia systems, in which synchronization support can operate independently of other features. More specifically, the main contribution of this paper is the discussions about how media synchronization challenges can enroll the Internet of Things research area, where distributed sensors and actuators are specified as media objects and can be related to usual hypermedia objects, all synchronized in time and space, in what we call the “Synchronism of Things”.

In the IoT, data is exchanged and used by heterogeneous devices in machine-to-machine communications. Managing complex systems is at the core of autonomic computing and a key topic in the IoT. Therefore, interoperability is a central issue, at both the syntactic and the semantic level. To tackle syntactic and architectural interoperability, standards allow systems to connect and exchange structured data. However, for data to be used, semantic interoperability must be ensured to provide meaning and consistency. In this paper we provide syntactic and semantic interoperability solutions in a home automation autonomic system.

The constantly increased variety of available hardware and software solutions for the IoT sector is facilitating the development of novel applications, but at the same time the lack of standardized or widely accepted means of interaction, deployment and configuration is seriously hindering the IoT’s potential. The ARCADIA framework is an application development paradigm that enables the cooperation between software components designed and implemented independently and using various technologies and platforms, so that they can form sophisticated, distributed, cloud applications.

This article presents an initial set of results from the F-Interop European research project researching online platform for interoperability and performance tests for the Internet of Things. It presents some of the challenges faced by the IoT and online testing, and how F-Interop is addressing them, in order to provide an extensive experimental platform for online tests. It gives an overview of its overall architecture.

Time Slotted Channel Hopping (TSCH) is among the Medium Access Control (MAC) schemes defined in the IEEE 802.15.4-2015 standard. TSCH aims to guarantee high-level network reliability by keeping nodes time-synchronised. In order to ensure successful communication between a sender and a receiver, the latter starts listening shortly before the expected time of a MAC layer frame’s arrival. The offset between the time a node starts listening and the estimated time of frame arrival is called guard time and it aims to reduce the probability of missed frames due to clock drift. In this paper, we investigate the impact of the guard time length on network performance. We identify that, when using the 6TiSCH minimal schedule, the most significant cause of energy consumption is idle listening during guard time. Therefore, we perform empirical optimisations on the guard time to maximise the energy-efficiency of a TSCH link. Our experiments, conducted using the Contiki OS, show that optimal guard time configuration can reduce energy consumption by up to 40%, without compromising network reliability.

A core ingredient of the Internet of Things (IoT) is the use of deeply embedded resource constrained devices, often connected to the Internet over Low Power and Lossy Networks. These constraints compounded by the need for unsupervised operation within an untrusted environment create considerable challenges for the secure operation of these systems. In this paper, we propose a novel method to secure an edge IoT network using the concept of key pre-distribution proposed by Eschenauer and Gligor in the context of distributed sensor networks. First, we investigate the performance of the unmodified algorithm in the Internet of Things setting and then analyse the results with a view to determine its performance and thus its suitability in this context. Specifically, we investigate how ring size influences performance in order to determine the required ring size that guarantees full connectivity of the network. We then proceed to propose a novel RPL objective function and associated metrics that ensure that any node that joins the network can establish secure communication with Internet destinations.

The Internet of Things (IoT) promising a new generation of services been offered to a human being through a world of interconnected objects (called “things”) that may use different communication technologies. Objects, in IoT, are seamlessly connected on its owner/user behalf. To offer services, the service providers need to truly identify the effective actor/user rather than the communicated devices. Currently, users have relationships with multiple objects that can also be used to determine their user. These relationships between actors are changeable or may even vanish; however, they are important to distinguish the actual requester of the service. Hence, it is important to consider them when identifying the effective actor of the communicated object. This paper models these relationships, representing them in a general form, and proposes a new semantic identifier format that allows service providers to identify the service requester identity across domains based on those relationships.

With the rapid growth of information technology, more and more devices are connected to the network. Cyber security environment has become increasingly complicated. In the face of advanced threats, such as targeted attack and advanced persistent threat, traditional security measures of accumulating security devices to protect relevant systems and networks had been proved to be an unqualified failure. Aiming at this situation, this paper proposed a framework of cyber attack attribution based on threat intelligence. At first, after surveying and analyzing related academic research and industry solutions, this paper used the local advantage model to analysis the process of cyber attack. According to the definitions of seven steps in intrusion kill chains and six phases of F2T2EA model, this model proposed a method of collecting threat intelligence data and detecting and response to cyber attacks, so as to achieve the goals of early-warming, processing detection and response and posting attribution analysis, and finally to reverse the security situation. Then, this paper designed a framework of cyber attack attribution based on threat intelligence. The framework is composed by Start of analysis, Threat intelligence and Attribution analysis. The three main parts indicated the architecture of cyber attack attribution. Finally, we tested the framework by practical case. The case study shows that the proposed framework can provide some help in attribution analysis.

The openness of wireless communication and the unattended nature of sensor node deployment make it easy for an adversary to launch various attacks on wireless sensor networks. Cross-layer attack aims to achieve better attack effects, conceal attack behavior more better, reduce the cost of attack by using information from multiple protocol layers, or initiate attack at multiple layers cooperatively. There are now different understandings about cross-layer attack. In this paper, the definition of cross-layer attack is proposed and several cases of attacks are presented. In order to better understand their behaviors, the cases of cross-layer attack are modeled by utilizing unified modeling language, which helps to build more secure wireless sensor networks.

Nowadays the internet is considered as given in almost any consumer electronic application. Internet connections are now extended to physical objects and are able to connect the living environment with computers, laptops, tablets and smartphones. We are dealing here with the Internet of Things. However, it is only the beginning of the Internet of Things revolution and today the development process has entered a new stage, where Internet of Things includes more and more industrial devices. Of course, using Internet of Things in such application fields faces the challenge of balancing the flexibility of internet communication and the robustness of industrial applications. In this paper, a concept of the adoption of a miniaturized safety-related solution on a single chip for industrial Internet of Things applications is introduced. An example application is presented to prove the feasibility of the introduced concept.

Society is faced with the ever more prominent concerns of vulnerabilities including hacking and DoS or DDoS attacks when migrating to new paradigms such as Internet of Things (IoT). These attacks against computer systems result in economic losses for businesses, public organizations and privacy disclosures. The IoT presents a new soft surface for attack. Vulnerability is now found in a multitude of personal and private devices that previously lacked connectivity. The ability to trace back to an attack origin is an important step in locating evidence that may be used to identify and prosecute those responsible. In this theoretical research, IP traceback methods are compared and evaluated for application, and then consolidated into a set of metrics for potential use against attackers.