Intelligent Internet Of Things Devices Research Articles

Combinative model compression approach for enhancing 1D CNN efficiency for EIT-based Hand Gesture Recognition on IoT edge devices

Tiny Machine Learning is rapidly evolving in edge computing and intelligent Internet of Things (IoT) devices. This paper investigates model compression techniques with the aim of determining their compatibility when combined, and identifying an effective approach to improve inference speed and energy efficiency within IoT edge devices. The study is carried out on the application scenario of Hand Gesture Recognition (HGR) based on Electrical Impedance Tomography (EIT), which involves complex signal processing and needs real-time processing and energy efficiency. Therefore, a customized 1-Dimensional Convolutional Neural Network (1D CNN) HGR classification model has been designed. An approach based on strategically combining model compression techniques was then implemented resulting in a model customized for faster inference and improved energy efficiency for IoT embedded devices. The model size became compact at 10.42 kB, resulting in a substantial size reduction of 98.8%, and an inference gain of 94.73% on a personal computer with approximately 8.56% decrease in accuracy. The approach of combinative model compression techniques was applied to a wide range of edge-computing IoT devices with limited processing power, resulting in a significant improvement in model execution speed and energy efficiency for these devices. Specifically, there was an average power consumption gain of 52% for Arduino Nano BLE and 34.05% for Raspberry Pi 4. Inference time was halved for Arduino Nano BLE Sense, Nicla Sense, and Raspberry Pi 4, with a remarkable gain of 94% for ESP32.

Progressive Adoption of RINA in IoT Networks: Enhancing Scalability and Network Management via SDN Integration

Thousands of devices are connected to the Internet as part of the Internet of Things (IoT) ecosystems. The next generation of IoT networks is expected to support this growing number of Intelligent IoT devices and tactile Internet solutions to provide real-time applications. In view of this, IoT networks require innovative network architectures that offer scalability, security, and adaptability. The Recursive InterNetwork Architecture (RINA) is a clean slate network architecture that provides a scalable, secure, and flexible framework for interconnecting computers. SDN technology is becoming a de facto solution to overcome network requirements, making RINA adoption difficult. This paper presents an architecture for integrating RINA with SDN technologies to lower the barriers of adopting RINA in IoT environments. The architecture relies on a RINA-based distributed application facility (DAF), a RINA southbound driver (SBI), and the RINA L2VPN. The RINA-based DAF manages RINA nodes along the edge–fog–cloud continuum. The SBI driver SDN enables the hybrid centralized management of SDN switches and RINA nodes. Meanwhile, the RINA L2VPN allows seamless communication between edge nodes and the cloud to facilitate the data exchange between network functions (NFs). Such integration has enabled a progressive deployment of RINA in current IoT networks without affecting their operations and performance.

Smart urban planning: Intelligent cognitive analysis of healthcare data in cloud-based IoT

In recent years, there has been a growing recognition among competent city developers and researchers regarding the critical importance of integrating healthcare data analysis into smart urban planning. This research study focuses on enhancing the healthcare system using a cognitive-based cloud framework. We employ the Internet of Things (IoT) to collect patient data and utilize a cloud-based cognitive framework for real-time patient monitoring. We aim to achieve cost-effective and high-quality healthcare. Deep learning techniques are applied to assess the health status and obtain experimental results. Specifically, we conduct pathology detection and classification using Electroencephalography (EEG). Multimodal patient health data, including EEG signals, are recorded using an EEG sensor. Intelligent IoT devices transmit the EEG signals from patients to the cloud, where they undergo processing and are forwarded to a cognitive module. The system tracks various sensor readings, such as facial expressions, speech, EEG, movements, and gestures, to determine the patient's condition. To classify the EEG signals as pathological or normal, our proposed method employs a Transfer Learning-based Convolutional Neural Network with Long Short-Term Memory (CNN-LSTM) and the Kruskal-Wallis (KW) method. We demonstrate the effectiveness of our approach, outperforming existing methods on the same dataset, achieving an impressive accuracy of 95.13% in identifying EEG pathologies.

Energy-Aware, Device-to-Device Assisted Federated Learning in Edge Computing

The surging of deep learning brings new vigor and vitality to shape the prospect of intelligent Internet of Things (IoT), and the rise of edge intelligence enables provisioning real-time deep neural network (DNN) inference services for mobile users. To perform efficient and effective DNN model training in edge computing environments while preserving training data security and privacy of IoT devices, federated learning has been envisioned as an ideal learning paradigm for this purpose. In this paper, we study energy-aware DNN model training in edge computing. We first formulate a novel energy-aware, Device-to-Device (D2D) assisted federated learning problem with the aim to minimize the global loss of a training DNN model, subject to bandwidth capacity on an edge server and energy capacity on each IoT device. We then devise a near-optimal learning algorithm for the problem when the training data follows the i.i.d. data distribution. The crux of the proposed algorithm is to explore using the energy of neighboring devices of each device for its local model uploading, by reducing the problem to a series of weighted maximum matching problems in corresponding auxiliary graphs. We also consider the problem without the assumption of the i.i.d. data distribution, for which we propose an efficient heuristic algorithm. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results show that the proposed algorithms are promising.

A provably secure and PUF-based authentication key agreement scheme for cloud-edge IoT

With the exponential growth of intelligent Internet of Things (IoT) applications, Cloud-Edge (CE) paradigm is emerging as a solution that facilitates resource-efficient and timely services. However, it remains an underlying issue that frequent end-edge-cloud communication is over a public or adversary-controlled channel. Additionally, with the presence of resource-constrained devices, it's imperative to conduct the secure communication mechanism, while still guaranteeing efficiency. Physical unclonable functions (PUF) emerge as promising lightweight security primitives. Thus, we first construct a PUF-based security mechanism for vulnerable IoT devices. Further, a provably secure and PUF-based authentication key agreement scheme is proposed for establishing the secure channel in end-edge-cloud empowered IoT, without requiring pre-loaded master keys. The security of our scheme is rigorously proven through formal security analysis under the random oracle model, and security verification using AVISPA tool. The comprehensive security features are also elaborated. Moreover, the numerical results demonstrate that the proposed scheme outperforms existing related schemes in terms of computational and communication efficiency.

Federated Learning Based Privacy Ensured Sensor Communication in IoT Networks: A Taxonomy, Threats and Attacks

Our daily lives are significantly impacted by intelligent Internet of Things (IoT) application, services, IoT gadgets, and more intelligent industries. Artificial Intelligence (AI) is anticipated to have a substantial impact on training machine learning algorithms on IoT devices without sharing data. As data privacy has become a serious societal concern, Federated Learning (FL) has emerged as a hot research area for enabling the collaborative training of machine learning models across many smart IoT devices while adhering to privacy constraints. Although, FL is utilized for preserving privacy in IoT networks, but it is also facing some challenges such as privacy, effectiveness, and efficiency. In this article, a taxonomy of FL-based IoT systems is proposed and an analysis of related works on FL-based IoT systems is presented. Further, a comprehensive study of various types of threats, attacks, and frameworks is done. In addition to this, a taxonomy on privacy-preserving FL techniques for IoT networks is also devised. Finally, the study is concluded by highlighting the various open research challenges in FL-based IoT networks.

Intelligent Resource Optimization for Blockchain-Enabled IoT in 6G via Collective Reinforcement Learning

Artificial intelligence (AI)-enabled Internet of Things (IoT) has attracted great interests. The accuracy of data training model in AI is vital for further development of IoT. In addition, with the increasing number of intelligent IoT devices, the amounts of data available for transmission, learning and training can lead to serious communication burdens and data reliability issues. In order to address these issues, we study novel network architectures in future 6G networks to support the intelligent IoT. Moreover, inspired by the collective learning of humans, we introduce and adopt a novel method named as collective reinforcement learning (CRL) in the intelligent IoT to realize the sharing of learning and training results. To ensure security and privacy, as well as improve computing efficiency, blockchain, mobile edge computing (MEC) and cloud computing are applied to protect data security and enrich computing resources. On this basis, we formulate an optimization problem in the intelligent IoT based on the proposed framework to optimize transmission latency and energy consumption. Simulation results demonstrate that the system performance has improved significantly. At last, some research challenges and open issues are pointed out to the intelligent IoT in future networks.

A Secure Group Data Encryption Scheme in Intelligent Manufacturing Systems for IIoT

In recent years, there are many industries that imported intelligent systems to help them make the intelligent factory and get product record analysis to evaluate product rate. These intelligent systems could generate product records, store them to the on-line database, and provide product rate analysis from these records. Due to the rapid development of internet of things (IoT), the stockholder can construct its own smart factory with the smart intelligent system to develop its own industrial internet of things (IIoT) architecture. With the help of IIoT, the smart intelligence system can collect data information with IoT sensors embedded into each machine in the production line. However, there are some security issues arising between smart intelligent systems and IoT devices. In addition, the authors also discovered that there are fewer methodologies to talk about the data security during the machine transmitting its censored data to the other machines under the same network environment.

Secured control systems through integrated IoT devices and control systems

The Internet of Things (IoT) provides customers with seamless integration and interoperability between gadgets, technologies, or products, and across multiple networks focus on control schemas. Regardless of the integration node, whether it is a smartphone, tablet, laptop, TV, automobile, or other Internet-compatible gadget, end-users want a fast and clear connection. Intelligent IoT devices can act as flexible connectors with variable structures, allowing rapid responses and possibly universal access. Consequently, gadgets could pose serious security issues. When they are used to manage accounts receivable at IoT terminals, these methods are likely to replay, provoke human attacks in the middle, or by denial of service. Responding to the real-time challenge for safe control operations, this study provides a methodology that blends the characteristics of intelligent IoT systems with power system interfaces. A proposed method combines computation, cryptography, signal/image processing, and communication networks of authentication and authorization tasks, using both terminals and gateways. Unlike conventional methods, they provide a safer, more adaptable, and resilient setting for real-time performance. The researchers also describe Intel's IoT framework for power system inputs that are used in management, industrial, transport, and detailed monitoring contexts.

ES-PPDA: an efficient and secure privacy-protected data aggregation scheme in the IoT with an edge-based XaaS architecture

In an Internet of Things (IoT) system based on an anything as a service (XaaS) architecture, data are uploaded from heterogeneous nodes in a nonstandardized format and aggregated on the server side. Within this data-intensive architecture, privacy preservation is one of the most important issues. In response to this concern, there are numerous privacy-protection data aggregation (PPDA) solutions available for various IoT applications. Because of the limited resources of intelligent IoT devices, traditional PPDA cannot meet practical privacy and performance needs. To tackle this challenge, we provide a more efficient and secure PPDA solution that guarantees data security and integrity through Paillier homomorphic encryption and online/offline signing technology. Detailed security analysis shows that our system is unpredictable under a chosen message attack, and the data integrity may be guaranteed under the assumption of q-strong Diffie-Hellman (Q-SDH). We choose an M/G/1 priority queue model to maximize system performance. M/G/1 enhances queuing efficiency and accelerates channel access, thus reducing waiting time and increasing reliability. The experimental results show that our data aggregation scheme is reliable with low latency.

Multiagent Bayesian Deep Reinforcement Learning for Microgrid Energy Management Under Communication Failures

Microgrids (MGs) are important players for the future transactive energy systems where a number of intelligent Internet of Things (IoT) devices interact for energy management in the smart grid. Although there have been many works on MG energy management, most studies assume a perfect communication environment, where communication failures are not considered. In this article, we consider the MG as a multiagent environment with IoT devices in which AI agents exchange information with their peers for collaboration. However, the collaboration information may be lost due to communication failures or packet loss. Such events may affect the operation of the whole MG. To this end, we propose a multiagent Bayesian deep reinforcement learning (BA-DRL) method for MG energy management under communication failures. We first define a multiagent partially observable Markov decision process (MA-POMDP) to describe agents under communication failures, in which each agent can update its beliefs on the actions of its peers. Then, we apply a double deep <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -learning (DDQN) architecture for <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -value estimation in BA-DRL, and propose a belief-based correlated equilibrium for the joint-action selection of multiagent BA-DRL. Finally, the simulation results show that BA-DRL is robust to both power supply uncertainty and communication failure uncertainty. BA-DRL has 4.1% and 10.3% higher reward than Nash deep <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -learning (Nash-DQN) and alternating direction method of multipliers (ADMM), respectively, under 1% communication failure probability.

Energized IOT Sensor through RF Harvesting Energy

Wireless Power Collecting (WPC) present the future in powering and energizing intelligent Internet of Things (IoT) electronics devices. This chapter studies and utilizes a circuit to powering wirelessly IoT devices. The WPC offer a best technique to help researchers and engineers of modern societies to build cell blocks. The concept is to energy any IoT devices and sensors wirelessly from Radio Frequency (RF) power strength in the same areas that may be hard to achieve or potentially hazardous. We implemented the RF harvesting technology with IoT devices to increase the efficiency of sensors. The idea of this system is to power up and self-energize any IoT sensors wirelessly. This work studies two different topologies of a rectangular patch antenna and different RF harvesting circuit voltage multiplier configurations using a microwave power station as the input RF source. This work aims to utilize the wireless power transmission technique in the smart house solution. The proposed prototype gets all technical parameters to generate enough electricity to power up the bulbs 5 W wirelessly at a gap distance around a five meters. Finally, we test the RF rectifier circuit coupled with a twin patch antenna that can self-energize the bulb, eventually devices work without batteries.

IoT intelligence empowered by end-edge-cloud orchestration

To support intelligent Internet of Things (IoT) applications, such as autonomous driving, smart city surveillance, and virtual reality (VR)/augmented reality (AR), cloud services are expected to be pushed to the proximity of IoT devices for quality performance. For instance, to facilitate safe autonomous driving, the service delay of most vehicular applications is required to be within milliseconds, and any information delay may result in dangerous on-road conditions. Edge intelligence aims at processing data/computing-intensive IoT tasks at the edge of network, where a set of IoT devices can work cooperatively for data collection, processing, model training, caching, and data analytics via edge caching, edge training, edge offloading, etc. It empowers intelligent IoT services at the network edge. However, it is challenging to achieve satisfying performance due to the following reasons. On the one hand, as the edge nodes are constrained by storage/computing resources, it is essential to conduct end-edge-cloud resource orchestration and resource sharing, taking into account device mobility, burst & stochastic service requests, and heterogeneous resources. On the other hand, better quality of service/learning of IoT applications is difficult to be guaranteed as the task execution/model training is contributed by distributed IoT devices. As a result, individual quality characterization, participating node selection, multi-level collaboration, robustness against malicious attacks are crucial. At last, to enable IoT intelligence, frequent communication and coordination among IoT devices, edge nodes and cloud servers are required, which can raise significant overhead, delay, and potential disclosure of sensitive information. Overcoming those challenges calls for further in-depth research.

Learning the Optimal Partition for Collaborative DNN Training With Privacy Requirements

With the growth of intelligent Internet of Things (IoT) applications and services, deep neural network (DNN) has become the core method to power and enable increased functionality in many smart IoT devices. However, DNN training is difficult to carry out on end devices because it requires a great deal of computational power. The conventional approach to DNN training is generally implemented on a powerful computation server; nevertheless, this approach violates privacy because it exposes the training data to curious service providers. In this article, we consider a collaborative DNN training system between a resource-constrained end device and a powerful edge server, aiming at partitioning a DNN into a front-end part running on the end device and a back-end part running on the edge server to accelerate the training process while preserving the privacy of the training data. With the key challenge being how to locate the optimal partition point to minimize the end-to-end training delay, we propose an online learning module, called learn-to-split (L2S), to adaptively learn the optimal partition point on the fly. This approach is unlike existing efforts on DNN partitioning that relies heavily on a dedicated offline profiling stage. In particular, we design a new contextual bandit learning algorithm called LinUCB-E as the basis of L2S, which has provable theoretical learning performance and is ultralightweight for easy real-world implementation. We implement a prototype system consisting of an end device and an edge server, and experimental results demonstrate that L2S can significantly outperform state-of-the-art benchmarks in terms of reducing the end-to-end training delay and preserving privacy.

Intelligent energy and ecosystem for real-time monitoring of glaciers

The United Nations is deeply concerned about global warming and its impacts on natural resources. Simultaneously, it has been recommended that cutting-edge technologies be employed to predict the impact of climate change on natural reservoirs such as glaciers. With the motivation of the above facts, this study investigates the impact and significance of emerging and cutting-edge technologies like Remote Sensing, the Internet of Things (IoT), Artificial Intelligence (AI), Unmanned Aerial Vehicles (UAVs), and robots implementation for the digitalization in glaciers. The study identified that convergence of AI, Machine Learning (ML), and Deep Learning approaches with Spatio-temporal data empowers to detect non-linear characteristics, especially in high mountainous regions due to their diversity and unpredictable nature. The article suggested valuable recommendations such as establishing an intelligent eco-system in glaciers, low-cost intelligent IoT devices with intelligent energy algorithms, ML empowered edge devices, glacier-resistant rescue robots, and Wearable IoT-based safety guide devices.

A Traceable and Anonymous Data Aggregation Scheme with Fog Computing Devices for Smart Grid

As an extension of the Internet of Things (IoT), smart city is a new aim for applications used in different industrial aspects. Intelligent IoT devices are used everywhere in smart cities to implement the functions such as monitoring and managing. Smart grid is one of the critical parts. Obtaining the quantity and cost of electricity usage is the most critical task of the smart grid. But such data in plaintext may leak the user privacy. So, it is an emergent aim to protect the private information of the user. Thus, we present a secure scheme for the smart grid, which not only protects the user’s information including identity and power consumption in the communications but also tracks the correct electricity cost for each user. Also, electricity consumption data from users could be aggregated in fog devices and then analyzed by the utility provider. The formal proof shows that the message transmission reaches the security level of the chosen plaintext attack (CPA). Also, the security properties of the scheme express the robustness. Finally, the performance study demonstrates that the proposed protocol is acceptable in practice.

Intelligent Load-Balancing Framework for Fog-Enabled Communication in Healthcare

The present technological era significantly makes use of Internet-of-Things (IoT) devices for offering and implementing healthcare services. Post COVID-19, the future of the healthcare system is highly reliant upon the inculcation of Artificial-Intelligence (AI) mechanisms in its day-to-day procedures, and this is realized in its implementation using sensor-enabled smart and intelligent IoT devices for providing extensive care to patients relative to the symmetric concept. The offerings of such AI-enabled services include handling the huge amount of data processed and sensed by smart medical sensors without compromising the performance parameters, such as the response time, latency, availability, cost and processing time. This has resulted in a need to balance the load of the smart operational devices to avoid any failure of responsiveness. Thus, in this paper, a fog-based framework is proposed that can balance the load among fog nodes for handling the challenging communication and processing requirements of intelligent real-time applications.

Novel Image Encryption and Compression Scheme for IoT Environment

Latest advancements made in the processing abilities of smart devices have resulted in the designing of Intelligent Internet of Things (IoT) environment. This advanced environment enables the nodes to connect, collect, perceive, and examine useful data from its surroundings. Wireless Multimedia Surveillance Networks (WMSNs) form a vital part in IoT-assisted environment since it contains visual sensors that examine the surroundings from a number of overlapping views by capturing the images incessantly. Since IoT devices generate a massive quantity of digital media, it is therefore required to save the media, especially images, in a secure way. In order to achieve security, encryption techniques as well as compression techniques are employed to reduce the amount of digital data, being communicated over the network. Encryption Then Compression (ETC) techniques pave a way for secure and compact transmission of the available data to prevent unauthorized access. With this background, the current research paper presents a new ETC technique to accomplish image security in IoT environment. The proposed model involves three major processes namely, IoT-based image acquisition, encryption, and compression. The presented model involves optimal Signcryption Technique with Whale Optimization Algorithm (NMWOA) abbreviated as ST-NMWOA. The optimal key generation of signcryption technique takes place with the help of NMWOA. Besides, the presented model also uses Discrete Fourier Transform (DFT) and Matrix Minimization (MM) algorithm-based compression technique. Extensive set of experimental analysis was conducted to validate the effective performance of the proposed model. The obtained values infer that the presented model is superior in terms of both compression efficiency and data secrecy in resource-limited IoT environment.

Design Guidelines for Cooperative UAV-supported Services and Applications

Internet of Things (IoT) systems have advanced greatly in the past few years, especially with the support of Machine Learning (ML) and Artificial Intelligence (AI) solutions. Numerous AI-supported IoT devices are playing a significant role in providing complex and user-specific smart city services. Given the multitude of heterogeneous wireless networks, the plethora of computer and storage architectures and paradigms, and the abundance of mobile and vehicular IoT devices, true smart city experiences are only attainable through a cooperative intelligent and secure IoT framework. This article provides an extensive study on different cooperative systems and envisions a cooperative solution that supports the integration and collaboration among both centralized and distributed systems, in which intelligent AI-supported IoT devices such as smart UAVs provide support in the data collection, processing and service provisioning process. Moreover, secure and collaborative decentralized solutions such as Blockchain are considered in the service provisioning process to enable enhanced privacy and authentication features for IoT applications. As such, user-specific complex services and applications within smart city environments will be delivered and made available in a timely, secure, and efficient manner.

The convergence of IoT and distributed ledger technologies (DLT): Opportunities, challenges, and solutions

Digital revolution is characterized by the convergence of technologies — from cloud computing to edge/fog computing, Artificial Intelligence (AI), big data, Intelligent Internet of Things (IoT), and Distributed Ledger Technology (DLT) — that is blurring the lines between physical and digital worlds. In this context, the IoT tsunami, the ubiquitous adoption of intelligent connected devices, and the public embracement of DLT, of which blockchain is a popular example, are increasingly becoming an integral feature of many modern systems, particularly, IoT-based smart and connected healthcare. Although the IoT and DLT/blockchain are two very different technologies and distinct from each other, the fusion of blockchain and IoT technologies is an unprecedented paradigm shift that is expected to disrupt both current and future systems in various fields. Blockchain has exactly what is needed to fix the weaknesses and vulnerabilities of IoT. It solves the security fault line among intelligent IoT where most of the IoT devices are connected to each other through the public trustless environment. Moreover, its distributed peer-to-peer nature can address the shortcomings of client/server models in Cloud-IoT solutions. Although the convergence of IoT and blockchain (Blockchain-IoT) can potentially tackle major shortcomings of today's solutions, its adoption is still in infancy, suffering from various issues and thus there is a necessity to address significant challenges including scalability, consensus algorithms, data privacy, efficiency, availability, storage, interoperability, and standardization, among others. In addition, there is no consensus towards any reference model or best practices that specify how blockchain should be utilized in IoT. This paper aims to present a holistic reference architecture as well as fundamentals, recent advancements, promises, and challenges in order to foster the investigations on cutting-edge research and allowing one to contribute to advancing the convergence of blockchain and IoT.