Two-Stage Client Selection Scheme for Blockchain-Enabled Federated Learning in IoT

The Internet of Things (IoT) has revolutionized data collection and analysis from diverse sources. IoT-based monitoring systems are now widespread in manufacturing, healthcare, and smart cities. These systems gather vast amounts of data from sensors and devices, enabling the detection of anomalies and patterns. The IoT has become an integral part of our lives, transforming various industries by enabling seamless connectivity between devices and increasing automation and efficiency. However, the reliability of IoT systems is often compromised due to the complexity and scale of these networks, making them vulnerable to failures and security breaches. To mitigate this problem, anomaly detection using efficient learning in IoT is a promising solution to identify unusual patterns that could indicate system faults or threats. This paper implements an efficient learning-driven approach for anomaly detection classification in IoT-based monitoring systems. The study aims to improve the accuracy and speed of identifying unusual patterns or behaviours in IoT data streams. By using advanced machine learning techniques, the proposed method can effectively detect anomalies in real-time, enhancing the overall security and reliability of IoT monitoring systems. The proposed framework enhances IoT system reliability by reducing downtime, improving security, and ensuring consistent performance of connected devices. Experimental results demonstrate the efficiency of machine learning techniques and their capabilities in anomaly detection. Empirical findings show that Decision Trees (DT) and Random Forests (RF) outperform other competitive models like Logistic Regression (LR), Naïve Bayes (NB), K-Nearest Neighbour (KNN), Extreme Gradient Boosting (XGB), and Adaptive Boosting (AB) and Voting Classifier (VC) for network intrusion detection in the context of anomaly detection.

Fast advancements in equipment, programming, and correspondence advances have permitted the rise of internet-associated tangible gadgets that give perception and information estimation from the physical world. It is assessed that the aggregate number of internet-associated gadgets being utilized will be in the vicinity of 25 and 50 billion. As the numbers develop and advances turn out to be more develop, the volume of information distributed will increment. Web-associated gadgets innovation, alluded to as internet of things (IoT), keeps on broadening the present internet by giving network and cooperation between the physical and digital universes. Notwithstanding expanded volume, the IoT produces big data described by speed as far as time and area reliance, with an assortment of numerous modalities and changing information quality. Keen handling and investigation of this big data is the way to creating shrewd IoT applications. This chapter evaluates the distinctive machine learning techniques that deal with the difficulties in IoT information.

The Internet of Things (IoT) will be a main data generation infrastructure for achieving better system intelligence. However, the extensive data collection and processing in IoT also engender various privacy concerns. This paper provides a taxonomy of the existing privacy-preserving machine learning approaches developed in the context of cloud computing and discusses the challenges of applying them in the context of IoT. Moreover, we present a privacy-preserving inference approach that runs a lightweight neural network at IoT objects to obfuscate the data before transmission and a deep neural network in the cloud to classify the obfuscated data. Evaluation based on the MNIST dataset shows satisfactory performance.

The Raspberry Pi is being increasingly adopted as a suitable platform in both research and applications of the Internet of Things (IoT). This study presents a novel project-based teaching and learning approach devised in an Internet of Things course for undergraduate students in the computer science major, where the Raspberry Pi platform is used as an effective vehicle to greatly enhance students’ learning performance and experience. The devised course begins with learning simple hardware and moves to building a whole prototype system. This paper illustrates the outcome of the proposed approach by demonstrating the prototype IoT systems designed and developed by students at the end of one such IoT course. Furthermore, this study provides insights and lessons regarding how to facilitate the use of the Raspberry Pi platform to successfully achieve the goals of project-based teaching and learning in IoT.

The Internet of Things (IoT) is becoming a buzz word due to the extensive application of it. It can be found anywhere nowadays, Healthcare, Smart cities, Smart energy, smart buildings and so on. The internet of thing has been a subject of improvement in terms of the price, size and persistence, but one of the main challenges in understanding IoT is in dealing with the device and data heterogeneities and establishing the interoperability between devices, information, and services built on top of it. Therefore, Ontologies can bring the change we are aiming in establishing a standard inter-devices communication and enable interoperability in m2m communication that allows data and data semantics to be described in application-independent ways.

Hyperdimensional Computing (HDC) developed by Kanerva is a computational model for machine learning inspired by neuroscience. HDC exploits characteristics of biological neural systems such as high-dimensionality, randomness and a holographic representation of information to achieve a good balance between accuracy, efficiency and robustness. HDC models have already been proven to be useful in different learning applications, especially in resource-limited settings such as the increasingly popular Internet of Things (IoT). One class of learning tasks that is missing from the current body of work on HDC is graph classification. Graphs are among the most important forms of information representation, yet, to this day, HDC algorithms have not been applied to the graph learning problem in a general sense. Moreover, graph learning in IoT and sensor networks, with limited compute capabilities, introduce challenges to the overall design methodology. In this paper, we present GraphHD - a baseline approach for graph classification with HDC. We evaluate GraphHD on real-world graph classification problems. Our results show that when compared to the state-of-the-art Graph Neural Networks (GNNs) the proposed model achieves comparable accuracy, while training and inference times are on average <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$14.6\times$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$2.0 \times$</tex> faster, respectively.

The development of Internet of Things (IoT) makes human life more intelligent, and the interconnection of all things has become a reality. However, the surge in the number of devices and centralized management brings severe challenges to IoT, such as single point of failure, poor security, privacy leakage, and low reliability. Due to the decentralization, verifiability and privacy protection of blockchain federated learning, some blockchain federated learning schemes have been proposed to solve these problems, but bring new challenges, such as device privacy leakage and heavy security computing load. In this paper, we propose a new decentralized, secure and verifiable consortium blockchain federated learning privacy protection scheme, named LPBFL, which realizes lightweight computing while ensuring the privacy of the local model and dataset of the device. To achieve lightweight privacy protection, LPBFL adopts the Paillier encryption and the newly designed lightweight digital signature and batch verification algorithm. Additionally, considering that devices upload invalid or even toxic local models intentionally or unintentionally, we design a device reputation selection mechanism to make blockchain federated learning more efficient. Finally, the theoretical analysis proves the security of LPBFL and verifies the unforgeability of the proposed digital signature. Comprehensive comparisons and extensive experiments demonstrate that our LPBFL has significant advantages in multiple aspects.

To address the asynchronous challenges stemming from user resource constraints and intermittent network connections in distributed Internet of Things (IoT) systems, asynchronous federated learning (AFL) has been extensively explored in both academic and industrial domains. However, existing AFL approaches still struggle to effectively tackle the issue of model aggregation information loss due to delayed users, which leads to degraded model accuracy. To alleviate the impact of delayed or unavailable model updates on model aggregation, we propose a novel model update enhancement method to compensate for the loss of model aggregation information. Specifically, we utilize delayed model updates as an update agent for this user and correct the weights of these updates within the delayed rounds threshold to ensure the delayed user’s contribution to the model aggregation. Additionally, we integrate symmetric homomorphic encryption (SHE) into AFL to ensure user privacy while simultaneously minimizing computational overhead. Lastly, we conduct extensive experiments to demonstrate that our scheme improves model accuracy by 6.53% compared to state-of-the-art solutions. Our code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/MiaoGroup-XDU/acc-improved-ppafl</uri>.

The proliferation of the Internet of Things (IoT) and widespread use of devices with sensing, computing, and communication capabilities have motivated intelligent applications empowered by artificial intelligence. Classical artificial intelligence algorithms require centralized data collection and processing, which are challenging in realistic intelligent IoT applications due to growing data privacy concerns and distributed datasets. Federated Learning (FL) has emerged as a privacy-preserving distributed learning framework, which enables IoT devices to train global models through sharing model parameters. However, inefficiency due to frequent parameter transmissions significantly reduces FL performance. Existing acceleration algorithms consist of two main types including local update and parameter compression, which considers the trade-offs between communication and computation/precision, respectively. Jointly considering these two trade-offs and adaptively balancing their impacts on convergence have remained unresolved. To solve the problem, this paper proposes a novel efficient adaptive federated optimization (FedEAFO) algorithm to improve the efficiency of FL, which minimizes the learning error via jointly considering two variables including local update and parameter compression. The FedEAFO enables FL to adaptively adjust two variables and balance trade-offs among computation, communication, and precision. The experiment results illustrate that compared with state-of-the-art algorithms, the FedEAFO can achieve higher accuracies faster.

Network Intrusion Detection System (NIDS) has become more important as a large number of diverse devices connect to the Internet of Things (IoT). Generally, training an effective NIDS requires a large amount of high-quality and centralized attack data. However, in real-world scenarios, it is difficult to centralize the distributed data for training NIDS in the IoT due to the privacy concerns and data format heterogeneity. To solve this problem, a novel NIDS combining federated learning and cost-sensitive learning is proposed, named FIDS-CL. Specifically, federated learning with dynamic weights aggregation tackles the problem of non-clusterable data, where multiple clients collaboratively enhance the overall performance while dynamically aggregating weights to maximize the retention of high-performing client models. Moreover, cost-sensitive learning is employed to alleviate the problem of class imbalance in NIDS by dynamically adjusting the gradient descent weights of different classes in the loss function, thereby emphasizing the importance of minority classes. Therefore, our method can effectively handle data imbalance while safeguarding data privacy of clients, which is more effective to detect network intrusions. The experiments conducted across various scenarios validate the superior detection capabilities and computational efficiency of FIDS-CL when compared to other state-of-the-art NIDSs.

Federated learning (FL) framework enables user devices to collaboratively train a global model based on their local data sets without privacy leak. However, the training performance of FL is degraded when the data distributions of different devices are incongruent. Fueled by this issue, we consider a clustered FL (CFL) method where the devices are divided into several clusters according to their data distributions and are trained simultaneously. Convergence analysis is conducted, which shows that the clustered model performance depends on cosine similarity, device number per cluster, and device participation probability. Besides, to quantify the training performance, the utility of clustered model training is defined based on the analysis results. Then, aiming at optimizing the system utility, a joint problem of resource allocation and device clustering is formulated, which is solved by decoupling it into two subproblems. First, given the results of device clustering, a low-complexity iterative algorithm based on the convex optimization theory is proposed to make the bandwidth allocation and the transmit power control. Then, according to the individual stability, a coalition formation algorithm is proposed for the device clustering. Finally, the real-data experiments on the classification tasks (e.g., MNIST, CIFAR-10, and CIFAR-100) validate the results of convergence analysis and advantages of the proposed algorithm in terms of the test accuracy.

The Internet of Vehicles (IoV) is a promising branch of the Internet of Things. IoV simulates a large variety of crowdsourcing applications, such as Waze, Uber, and Amazon Mechanical Turk, etc. Users of these applications report the real-time traffic information to the cloud server which trains a machine learning model based on traffic information reported by users for intelligent traffic management. However, crowdsourcing application owners can easily infer users' location information, traffic information, motor vehicle information, environmental information, etc., which raises severe sensitive personal information privacy concerns of the users. In addition, as the number of vehicles increases, the frequent communication between vehicles and the cloud server incurs unexpected amount of communication cost. To avoid the privacy threat and reduce the communication cost, in this article, we propose to integrate federated learning and local differential privacy (LDP) to facilitate the crowdsourcing applications to achieve the machine learning model. Specifically, we propose four LDP mechanisms to perturb gradients generated by vehicles. The proposed Three-Outputs mechanism introduces three different output possibilities to deliver a high accuracy when the privacy budget is small. The output possibilities of Three-Outputs can be encoded with two bits to reduce the communication cost. Besides, to maximize the performance when the privacy budget is large, an optimal piecewise mechanism (PM-OPT) is proposed. We further propose a suboptimal mechanism (PM-SUB) with a simple formula and comparable utility to PM-OPT. Then, we build a novel hybrid mechanism by combining Three-Outputs and PM-SUB. Finally, an LDP-FedSGD algorithm is proposed to coordinate the cloud server and vehicles to train the model collaboratively. Extensive experimental results on real-world data sets validate that our proposed algorithms are capable of protecting privacy while guaranteeing utility.

Integrating Internet of Things (IoTs) devices with secure smart home networks assisted by the cloud signifies a cutting-edge and potent tool for contemporary home automation. This allows various appliances and devices in a home remotely controlled by the internet to communicate and share data. The typical smart home system depends on the cloud service or centralized server, which makes them further vulnerable to potential security breaches and single points of failure. As a decentralized nature, Blockchain (BC) distributes the control and storage of data across the network, preventing unauthorized attacks. Integrating BC technology into the protected smart home network boosts the system's dependability, safety, and privacy. In addition, machine learning (ML) and analytics offer behaviour analysis and predictive maintenance for optimized energy consumption. Finally, combining IoT with cloud-assisted security transforms homes into smart, connected ecosystems, offering convenience without integrating confidentiality or dependability. Accordingly, this study presents a BC-based Deep Learning in the Secure Smart Home Network (BPDL-SSHN) methodology in the IoT-cloud platform. In the BPDL-SSHN methodology, BC technology permits secret proficient data from the smart home network. Furthermore, the BPDL-SSHN method follows a series of processes to detect malicious activities such as Binary Fox Optimization Algorithm (BFOA) based feature selection, Attention-based Long Short-Term Memory (ALSTM)-based classification, and Harbor Seal Whiskers Optimization (HSWO)-based hyperparameter tuning. The HSWO method's design helps better the hyperparameter choice of the ALSTM method, significantly enhancing the recognition performance. The comparative outcome of the BPDL-SSHN methodology reported the proficient solution of the smart home network to detect and monitor malicious or harmful activities. The experimental outcome implied that the BPDL-SSHN methodology accomplishes a maximum accuracy performance of 98.91% over other approaches.

‘Machine Learning (ML) in Internet of Things (IoT)’ delves deeply into the complex and dynamic interplay between ML and IoT, revealing the game-changing effects of their association across a wide range of technological fields. To lay a solid beginning, the core ideas of IoT like data transfer and acquisition, and basic ML are reviewed. Readers will acquire a thorough understanding of the foundations of both ML and IoT by going over these key ideas. The potential and inherent issues that come with managing IoT data are explored. The main challenges include managing massive and quickly created data streams, guaranteeing data quality, and preserving data security. An investigation is done to exhibit how IoT data presents unseen prospects for industry-wide optimization, predictive analytics, and data-driven decision-making. The investigation of ML strategies adapted to the specifics of IoT become the core of this chapter. Supervised, unsupervised, and reinforcement learning techniques are analyzed to see whether they are applicable to specific IoT applications. Distributed ML and edge computing, which are crucial for making decisions in real-time in IoT settings with constrained computational resources are also dealt with. To reinforce the usefulness of the discussion, a few interesting case studies and actual implementations are included. These demonstrate how ML may be used to improve predictive maintenance procedures in industrial settings, reduce energy usage in smart homes, and transform healthcare through wearable IoT devices. ML being used in IoT ecosystems for anomaly detection, failure prediction, and environmental monitoring is also illustrated. This chapter is a technical reserve and a thorough guide for anybody looking to leverage the revolutionary power of IoT and ML in an increasingly linked world.

The increasing Internet-of-Things (IoT) devices have produced large volumes of data. A deep learning technique is widely used to analyze the potential value of these data due to its unprecedented performance in both the academic and industrial communities. However, the data generated from the IoT devices are distributed among different users. Directly combining these data to a central server will cause privacy leakage, especially for personal sensitive data. Rather than centralized training by getting access to all these raw data, an alternative is to collaboratively learn a model in a distributed manner. However, there exist two main challenges in a distributed learning setting. The first one is how to preserve the privacy of users. The second one is to reduce the communication burden (e.g., mobile users have limited bandwidth) due to high-frequent data exchange. To address these two challenges, we design a communication efficient and privacy-preserving framework to enable different participants to distributively learn a model with a privacy protection guarantee. In particular, we develop a differentially private approximate mechanism for the distributed deep learning. In addition, we design a new gradient sparsification method to, at the first time, reduce both upload and download communication costs. The performance of the proposed framework is tested under different neural network structures for different data sets including, image classification and mobile sensor data. The experimental results demonstrate that we can reduce the communication up to only 2% compared to the full gradients exchange and achieve up to 16% accuracy increase compared to the previous works.