A Survey on Resilience in the IoT

Internet-of-Things (IoT) ecosystems tend to grow both in scale and complexity, as they consist of a variety of heterogeneous devices that span over multiple architectural IoT layers (e.g., cloud, edge, sensors). Further, IoT systems increasingly demand the resilient operability of services, as they become part of critical infrastructures. This leads to a broad variety of research works that aim to increase the resilience of these systems. In this article, we create a systematization of knowledge about existing scientific efforts of making IoT systems resilient. In particular, we first discuss the taxonomy and classification of resilience and resilience mechanisms and subsequently survey state-of-the-art resilience mechanisms that have been proposed by research work and are applicable to IoT. As part of the survey, we also discuss questions that focus on the practical aspects of resilience, e.g., which constraints resilience mechanisms impose on developers when designing resilient systems by incorporating a specific mechanism into IoT systems.

Internet of Things (IoT) Cloud Platform Using MQTT End-to-Cloud Architecture

IoT devices are constrained in computation and storage, therefore cannot store all long-term obtained data or perform complex computations. Shifting those jobs to cloud platform are feasible, yet rising heterogeneity and security issues. This study proposes an IoT cloud platform to facilitate communication among heterogeneous devices and the cloud while ensuring devices’ validity. It uses publish/subscribe paradigm with an end-to-cloud architecture and HTTP-based auth server. The proposed system has successfully addressed heterogeneity and security issues. Performance tests conclude that the fewer publishers publish data simultaneously, the smaller the delay. Moreover, the system performs better at up to 250 publishers as the average delay is under 1000 ms, compared to 500 publishers that has average delay above 1000 ms. On its scalability, in 250-concurrent-publishers experiment, the system affords 191 publishers responded in under one second with 100% success rate. In 500-concurrent-publishers one, 187 responded in under one second with 99% rate.

Implementation of the PointPillars Network for 3D Object Detection in Reprogrammable Heterogeneous Devices Using FINN

In this paper, we present a hardware-software implementation of a deep neural network for object detection based on a point cloud obtained by a LiDAR sensor. The PointPillars network was used in the research, as it is a reasonable compromise between detection accuracy and calculation complexity. The Brevitas / PyTorch tools were used for network quantisation (described in our previous paper) and the FINN tool for hardware implementation in the reprogrammable Zynq UltraScale+ MPSoC device. The obtained results show that quite a significant computation precision limitation along with a few network architecture simplifications allows the solution to be implemented on a heterogeneous embedded platform with maximum 19% AP loss in 3D, maximum 8% AP loss in BEV and execution time 375ms (the FPGA part takes 262ms). We have also compared our solution in terms of inference speed with a Vitis AI implementation proposed by Xilinx (19 Hz frame rate). Especially, we have thoroughly investigated the fundamental causes of differences in the frame rate of both solutions. The code is available at https://github.com/vision-agh/pp-finn.

Automating IoT Data Ingestion Enabling Visual Representation

The Internet of things has produced several heterogeneous devices and data models for sensors/actuators, physical and virtual. Corresponding data must be aggregated and their models have to be put in relationships with the general knowledge to make them immediately usable by visual analytics tools, APIs, and other devices. In this paper, models and tools for data ingestion and regularization are presented to simplify and enable the automated visual representation of corresponding data. The addressed problems are related to the (i) regularization of the high heterogeneity of data that are available in the IoT devices (physical or virtual) and KPIs (key performance indicators), thus allowing such data in elements of hypercubes to be reported, and (ii) the possibility of providing final users with an index on views and data structures that can be directly exploited by graphical widgets of visual analytics tools, according to different operators. The solution analyzes the loaded data to extract and generate the IoT device model, as well as to create the instances of the device and generate eventual time series. The whole process allows data for visual analytics and dashboarding to be prepared in a few clicks. The proposed IoT device model is compliant with FIWARE NGSI and is supported by a formal definition of data characterization in terms of value type, value unit, and data type. The resulting data model has been enforced into the Snap4City dashboard wizard and tool, which is a GDPR-compliant multitenant architecture. The solution has been developed and validated by considering six different pilots in Europe for collecting big data to monitor and reason people flows and tourism with the aim of improving quality of service; it has been developed in the context of the HERIT-DATA Interreg project and on top of Snap4City infrastructure and tools. The model turned out to be capable of meeting all the requirements of HERIT-DATA, while some of the visual representation tools still need to be updated and furtherly developed to add a few features.

Orchestrating Heterogeneous Devices and AI Services as Virtual Sensors for Secure Cloud-Based IoT Applications

The concept of the cloud-to-thing continuum addresses advancements made possible by the widespread adoption of cloud, edge, and IoT resources. It opens the possibility of combining classical symbolic AI with advanced machine learning approaches in a meaningful way. In this paper, we present a thing registry and an agent-based orchestration framework, which we combine to support semantic orchestration of IoT use cases across several federated cloud environments. We use the concept of virtual sensors based on machine learning (ML) services as abstraction, mediating between the instance level and the semantic level. We present examples of virtual sensors based on ML models for activity recognition and describe an approach to remedy the problem of missing or scarce training data. We illustrate the approach with a use case from an assisted living scenario.

Content Service Platform for Providing User-Centric Content in Smart Environments

Smart environments, such as smart cities and streets, contain various heterogeneous devices and content that provide information to users and interact with each other. In a smart environment, appropriate content should be provided based on the situations of users. Additionally, when a user is in motion, it is necessary to provide content in a seamless manner without interruption. A method for systematically controlling the delivery of such content is required. Therefore, we propose a content service platform to meet the needs discussed above. The content service platform supports the delivery of content and events between different devices, as well as the control of content. Context-aware technology can also be applied to support customized content. In this paper, we present an architectural model, a contextual reasoning process, and case study on applying content service platform to a smart street environment. The proposed content service platform applied as a base model to support the provision of user-specific content in smart environments.

Middleware Solutions for the Internet of Things: A Survey

The Internet of Things (IoT), along with its wider variants including numerous technologies, things, and people: the Internet of Everything (IoE) and the Internet of Nano Things (IoNT), are considered as part of the Internet of the future and ubiquitous computing allowing the communication among billions of smart devices and objects, and have recently drawn a very significant research attention. In these approaches, there are varieties of heterogeneous devices empowered by new capabilities and interacting with each other to achieve specific applications in different domains. A middleware layer is therefore required to abstract the physical layer details of the smart IoT devices and ease the complex and challenging task of developing multiple backend applications. In this chapter, an overview of IoT technologies, architecture, and main applications is given first and then followed by a comprehensive survey on the most recently used and proposed middleware solutions designed for IoT networks. In addition, open issues in IoT middleware design and future works in the field of middleware development are highlighted.

Hybrid Encryption for Messages’ Confidentiality in SOSE-Based IOT Service Systems

Internet of Things (IOT) is an essential paradigm where devices are interconnected into network. The operations of these devices can be through service-oriented software engineering (SOSE) principles for efficient service provision. SOSE is an important software development method for flexible, agile, loose-coupled, heterogeneous and interoperable applications. Despite all these benefits, its adoption for IOT services is slow due to security challenges. The security challenge of integration of IOT with service-oriented architecture (SOA) is man-in-the-middle attack on the messages exchanged. The transport layer security (TLS) creates a secured socket channel between the client and server. This is efficient in securing messages exchanged at the transport layer only. SOSE-based IOT systems needs an end-to-end security to handle its vulnerabilities. This integration enables interoperability of heterogeneous devices, but renders the system vulnerable to passive attacks. The confidentiality problem is hereby addressed by message level hybrid encryption. This is by encrypting the messages by AES for efficiency. However, to enable end-to-end security, the key sharing problem of advanced encryption standard (AES) is handled by RSA public key encryption. The results shows that this solution addressed data contents security and credentials security privacy issues. Furthermore, the solution enables end-to- end security of interaction in SOSE-based IOT systems.

An Era of Recommendation Technologies in IoT: Categorisation by techniques, Challenges and Future Scope

The evolution of the Internet of Things (IoT) accelerates the augmentation of data present on the Internet and possibilities for connections to the more dynamic and heterogeneous devices to the Internet. Recommendation technologies have proven their capabilities of digging the personalised information by proactive filtering in many application domains and can also be a backbone platform in IoT for identifying personalised things, services and relevant artefacts by prevailing over information overload problems. This paper is a comprehensive literature review that categorises IoT recommender systems by exploring the literature’s different IoT based recommendation techniques. We conclude the paper by discussing the challenges and future scope for IoT based recommendations techniques to advancing and widening the frontiers of this research area.

Role of artificial intelligence in cloud computing, IoT and SDN: Reliability and scalability issues

Information technology fields are now more dominated by artificial intelligence, as it is playing a key role in terms of providing better services. The inherent strengths of artificial intelligence are driving the companies into a modern, decisive, secure, and insight-driven arena to address the current and future challenges. The key technologies like cloud, internet of things (IoT), and software-defined networking (SDN) are emerging as future applications and rendering benefits to the society. Integrating artificial intelligence with these innovations with scalability brings beneficiaries to the next level of efficiency. Data generated from the heterogeneous devices are received, exchanged, stored, managed, and analyzed to automate and improve the performance of the overall system and be more reliable. Although these new technologies are not free of their limitations, nevertheless, the synthesis of technologies has been challenged and has put forth many challenges in terms of scalability and reliability. Therefore, this paper discusses the role of artificial intelligence (AI) along with issues and opportunities confronting all communities for incorporating the integration of these technologies in terms of reliability and scalability. This paper puts forward the future directions related to scalability and reliability concerns during the integration of the above-mentioned technologies and enable the researchers to address the current research gaps.