Machine learning algorithms for IoT applications

The correct implementation and behavior of Internet of Things (IoT) applications are seldom investigated in the literature. This paper shows how the simulation mechanism can be integrated well into an IoT application development platform for correct implementation and behavior investigation. We use an IoT application development platform called IoTtalk as an example to describe how the simulation mechanism called SimTalk can be built into this IoT platform. We first elaborate on how to implement the simulator for an input IoT device (a sensor). Then we describe how an output IoT device (an actuator) can be simulated by an animated simulator. We use a smart farm application to show how the simulated sensors are used for correct implementation. We use applications including interactive art (skeleton art and water dance) and the pendulum physics experiment as examples to illustrate how IoT application behavior investigation can be achieved in SimTalk. As the main outcome of this paper, the SimTalk simulation codes can be directly reused for real IoT applications. Furthermore, SimTalk is integrated well with an IoT application verification tool in order to formally verify the IoT application configuration. Such features have not been found in any IoT simulators in the world.

Widespread adoption of Internet of Things (IoT) applications assumes a sustainable ecosystem of firms creating, distributing, and using these applications. Such an ecosystem often forms around a platform that implements commonly used functionality and offers it to the members of the ecosystem for creating applications and services. In IoT, such a platform may be materialized in a form of a cloud platform that supports the management of connected devices, and the development and use of the applications based on them. A number of cloud platforms targeting IoT applications are available in the market; to prosper, they need to assist the providers of the IoT applications in designing and developing the applications, as well as in deploying and operating the application software. In this paper, a framework for evaluating the IoT platforms from the perspective of how widely they cover the potential needs of the application providers is introduced. Based on the framework, the maturity of the available IoT platforms is evaluated, by using the publicly available information about the platforms' features and supporting services. As the results of this initial analysis suggest, none of the platforms today offers the comprehensive support. This can be seen as a factor contributing to a slow adaption of the IoT platforms, and therefore to the slower-than-expected takeoff of the IoT ecosystem in general.

Internet of Things (IoT) applications are generally developed in a vertical manner, i.e., each IoT application is developed for a certain scenario which generally does not share data with IoT applications for recommendation services. This leads to an inefficient exploitation of other IoT service applications. In fact, such recommendation services can be achieved with the help of Social IoT (SIoT) by using data generated by various IoT applications. SIoT builds a profile of objects based on IoT applications data that can be exchanged with SIoT network for accessible to other IoT applications. In this manner, the SIoT network provides recommendation services for reusability of IoT applications' data among various IoT applications, as well as adapting IoT services according to users' needs which improve user experience. Additionally, the profiles built by a SIoT network can also help a single IoT application by looking for similar conditions that have been addressed in the past for the same IoT application. We propose a concept for exploiting the SIoT for recommendation services among various IoT applications with the help of a sample application scenario that highlights how the SIoT can help by providing recommendations. We also provide some implementation challenges for this concept.

An Internet of Things (IoT)-based system is smart system that provides automated and intelligent services. However, its intelligence still depends on the pre-ruled functions, which are only actuated according to rules without any predictions. To improve the intelligence of IoT and to train it to predict the future, it needs to be integrated with Artificial Intelligence (AI) technologies - either Machine Learning (ML) or Deep Learning (DL). Applying AI techniques in IoT involve some challenges and difficulties. In this paper, we go through some of challenges related to both systems of applying ML in IoT, as well as the issues that resulted from applying some of ML techniques in IoT. Moreover, this paper surveyed the background of IoT and ML and their components, and the usage of ML in IoT with possible applications.

The rise of successful cutting-edge Internet of Things (IoT) applications for urban development has inspired the industry and research community. Although the industry verticals such as automotive, robotics, e-health, and entertainment applications create new business opportunities for the service providers, they pose challenges in terms of deployment cost, reliability, and latency requirements. For example, self-driving vehicles require ultralow latency and reliable data processing to make split-second decisions. Unfortunately, cloud-based solutions for such IoT applications are not suitable due to the end-to-end latency and connection reliability. To meet these challenging requirements, edge computing paradigm has emerged as a solution where computational resources are brought to the proximity of the end users. Now, advanced data analytics, machine learning, and cognitive techniques can also be deployed at the edge of the network. Together, edge computing, data analytics, and machine learning empower service providers with true and intelligently automated infrastructure for IoT applications. In this chapter, we investigate the relevant use cases of edge-enabled IoT applications for smart urban development. To this end, we also provide a comprehensive study of the recent trends and the state of the art in accommodating all these emerging technologies in the acceleration of smart urban development.

The internet of things (IoT) transforms the world in many ways. It combines many types of hardware and software with a variety of communication technologies to enable the development of innovative applications. A typical IoT system consists of IoT device, IoT gateway, IoT platform, and IoT application. Developing these elements and delivering an IoT system for fulfilling business requirements encompasses many activities to be executed and is not straightforward. To expedite these activities, some major vendors provide software development kits (SDK), integrated development environments (IDE), and utility tools for developing software to be executed on IoT devices/gateways. Moreover, these vendors utilize their cloud platforms to provide fundamental services, such as data storage, analytics, stream processing, for developing IoT systems. These vendors also developed IoT specific cloud-based services, such as connectivity and device management, to support IoT system development. This chapter presents an overview of tools and platforms provided by five major vendors.

The rapid development of low-cost sensors, smart devices, communication networks, and learning algorithms has enabled data-driven decision making in large-scale systems. However, the development platforms for such Internet of Things (IoT) applications and collecting the data in a cohesive, yet simple manner is not very well understood. MIT App Inventor [1] is an open-source, user-friendly interface to develop mobile applications and has been used by over ten million users worldwide. We have added IoT capability to the MIT App Inventor platform where people can build applications using various sensors and IoT platforms such as Raspberry Pi [2] and Android Things [3] to collect the data through mobile applications. This poster paper presents a realization of easy to program IoT applications in the mobile application space with a focus on collecting data from the connected sensors. We have integrated the Android Things platform with the MIT App Inventor and introduced the existing MIT App Inventor community to the IoT based App development. This integration allows not only the seasoned developers, but also the novice developers to make interesting IoT applications with minimal programming knowledge.

As a major agricultural province in China, Shandong Province has always played an important role. At the same time, it is also closely related to the strong support of science and technology. As a new technology, the Internet of things (IoT) has made positive contributions. Agricultural IoT is a mode that uses information technology, wireless communication technology and radio frequency automatic identification technology to share the identified agricultural information to the network. The application of agricultural IoT is a major change in agricultural development, which provides convenient conditions for the development of agriculture, provides a favorable environment for the agricultural products, promotes the transformation of agricultural technology, realizes the informatization of agricultural production, improves the efficiency of rural land, integrates agricultural resources and provides a foundation for food safety. In this paper, the application of IoT in the field of agriculture in Shandong Province is described from two practical cases. One is about how to use wireless sensor network technology to transmit data in the big data platform based on the Bohai granary project. The other describes how to use IoT for food quality and safety traceability. It takes pigs as the research object, runs through all links of the whole industrial chain of pig breeding, breeding, slaughtering, processing, logistics and sales, collects the data of the whole industrial chain with the help of IoT, and realizes the information tracking and traceability of the whole industrial chain. At the end of the paper, it discusses the countermeasures.

The Internet of Things (IoT) is a hot topic in the engineering and technology community, and IoT-enabled devices are becoming increasingly common due to advancements in technology. IoT platforms provide valuable and specific services for IoT applications, and the number of platforms used by IoT applications is growing rapidly commensurate with the increasing number of IoT devices. However, fierce competition between IoT platforms is making it difficult for companies to select a suitable type of IoT platform for their business needs. This paper attempts to identify the important factors of IoT platforms that can be consider before selecting an IoT platform. In this paper, Twenty-one important IoT platform factors are identified from the literature and these factors were then verified by the Delphi method and categorized. This paper aims to help companies choose an appropriate IoT platform from the huge number and variety of IoT platforms available in the market.

SummaryIn recent years, applying Internet of Things (IoT) applications has significantly increased to facilitate and improve quality of human life activities in various fields such as healthcare, education, industry, economics, etc. The energy aware cloud‐edge computing paradigm has developed as a hybrid computing solution to provide IoT applications using available cloud service providers and fog nodes for the smart devices and mobile applications. Since the IoT applications are developed in the form of several IoT services with various quality of service (QoS) metrics which can deploy on the cloud‐edge providers with different resource capabilities, finding an efficient placement solution as one of challenging topics to be measured for IoT applications. The service placement issue arranges IoT applications on the cloud‐edge providers with various capabilities of atomic services though sufficient different QoS factors to support service level agreement (SLA) contracts. This paper presents a technical analysis on the cloud‐edge service placement approaches in IoT systems. The key point of this technical analysis is to identify substantial studies in the service placement approaches which need additional consideration to progress more efficient and effective placement strategies in IoT environments. In addition, a side‐by‐side taxonomy is proposed to categorize the relevant studies on cloud‐edge service placement approaches and algorithms. A statistical and technical analysis of reviewed existing approaches is provided, and evaluation factors and attributes are discussed. Finally, open issues and forthcoming challenges of service placement approaches are presented.

Chapter 3 - Integrated machine learning techniques for preserving privacy in Internet of Things (IoT) systems

The internet of things (IoT) is focused in offering new technologies which is changing the business environment by showing the potential in different sectors like retail, transportation, healthcare etc. Through IoT, the managers can gather and process data and information in an effective manner, also they can use them to automate in order to enhance productivity and efficiency. With the usage of such technologies, human resource managers can access information for possible candidates which can be accessed through IoT. The overall behaviour of potential candidate tends to help them in shortlisting and choosing the desirable ones. The application of IoT has emerged in enhancing the recruitment of candidates in search of candidates based on large volume of information, assist in screening the candidates, selecting the most suitable ones. The application of IoT technology can modify the HR roles, enable in changing the company perspective in enhancing recruitment process, source the candidates effectively and select them so as to meet the organisational requirements. The implementation of IoT driven technologies support in removing the biases in screening the candidates and aids in candidate selection with the support of other technologies like artificial intelligence, predictive analytics etc. Furthermore, IoT is also involved in better inventory management as it supports the organisation in converting the data which are sourced through RFID (Radio Frequency Identification) into more insights related to inventory data, location of the products, tracking the movement of goods until it reaches the hands of the customers. With the application of IoT inventory managers possess better control in managing the inventory through automated ordering request and thereby reduce manual time and cost for the business. This paper is focused in presenting a conceptual understanding on the application of environmental IoT in recruitment process and also the overall impact created by the technology in managing the inventory efficiently.KeywordsInternet of thingsRecruitment processInventory managementPredictive analytics

Under the background of prosperity of Internet of things (IoT), in order to accelerate the training of talented persons of China for IoT, we propose to design and develop a new course for Chinese students: Essential Technologies, Applications, and Commercial Patterns of Internet of Things. Aiming at three-dimensional target, which has dimensionalities of attitude and thoughts, knowledge and skills, process and methods, the contents of the course include the following parts: system and application architectures, key technologies of IoT, applications of IoT, commercial patterns of IoT, IoT industry and social significance etc. The course highlights the training of abilities of innovation, creativity, comprehensive analysis and practical operations. The construction of teaching resources, such as textbooks, lecture notes, teaching equipments, practice bases, are also emphasized in the process of development of this course.

Purpose – The purpose of this paper is to study lightweight security scheme for Internet of Things (IoT) applications using Constrained Application Protocol (CoAP). Resource-constrained characteristics of IoT systems have ushered in compelling requirements for lightweight application protocol and security suites. CoAP has already been established as the candidate protocol for IoT systems. However, low overhead security scheme for CoAP is still an open problem. Existing security solutions like Datagram Transport Layer Security (DTLS) is not suitable, particularly due to its expensive handshaking, public key infrastructure (PKI)-based authentication and lengthy ciphersuite agreement process. Design/methodology/approach – This paper proposes a lightweight security scheme in CoAP using Advanced Encryption Standard (AES) 128 symmetric key algorithm. The paper presents an object security (payload embedded)-based robust authentication mechanism with integrated key management. The paper introduces few unique modifications to CoAP header to optimize security operation and minimize communication cost. Findings – It is resilient to number of security attacks like replay attack, meet-in-the-middle attack and secure under chosen plaintext attack. This scheme is generic in nature, applicable for gamut of IoT applications. The paper proves efficacy of our proposed scheme for vehicle tracking application in emulated laboratory setup. Specifically, it compares with DTLS-enabled CoAP to establish the lightweight feature of our proposed solution. Research limitations/implications – This paper mainly focuses on implementing in-vehicle tracking systems as an IoT application and used CoAP as the application protocol. Practical implications – Such a lightweight security scheme would provide immense benefit in IoT systems so that resource constraint-sensing devices and nodes can be made secure. This would impact IoT eco systems to a large extent. Originality/value – Such kind of security suite that provides both robustness and lightweight feature is hitherto not known to the authors, particularly in CoAP for IoT applications.

Anomaly detection is an important security mechanism for the Internet of Things (IoT). Existing works have been focused on developing accurate anomaly detection models. However, due to the resource-constrained nature of IoT networks and the requirement of real-time security operation, cost efficient (regarding computational efficiency and memory-consumption efficiency) approaches for anomaly detection are highly desirable in IoT applications. In this paper, we investigated machine learning (ML) enabled anomaly detection models for the IoT with regard to multi-objective optimization (Pareto optimization) that minimizes the detection error, execution time, and memory consumption simultaneously. Making use of well-known datasets consisting of network traffic traces captured in an IoT environment, we studied a variety of machine learning algorithms through the world-class H2O AI platform. Our experimental results show that the Gradient Boosting Machine, Random Forest, and Deep Learning models are the most accurate and fastest anomaly detection models; the Gradient Boosting Machine and Random Forest are the most accurate and memory-efficient models. These ML models form the Pareto-optimal set of anomaly detection models. Our results can be used by the industry to facilitate their selection of ML models for anomaly detection on various IoT networks based on their security requirements and system constraints.