The integration of Tiny Machine Learning (TinyML) with Human Behavior Analysis (HBA) represents a significant advancement in the field of Artificial Intelligence (AI), enabling real-time, efficient, and privacy-preserving analysis on resource-constrained devices. This paper provides the first comprehensive survey exploring this integration, presenting a detailed overview of TinyML, including its definitions, key concepts and advantages. The survey proposes a systematic taxonomy of TinyML applications in HBA, categorizing state-of-the-art implementations based on their use cases and specific methodologies. Furthermore, the challenges and limitations of integrating TinyML in HBA are thoroughly discussed, including technical constraints, data quality issues, and ethical considerations. Finally, future research directions and open issues are outlined, emphasizing the potential advancements and emerging trends in this field. This survey serves as a foundational resource, guiding researchers and practitioners in harnessing the capabilities of TinyML to advance HBA.

Multi-modal learning leverages data from diverse perceptual media to obtain enriched representations, thereby empowering machine learning models to complete more complex tasks. However, recent research results indicate that multi-modal learning still suffers from " modality imbalance '': Certain modalities' contributions are suppressed by dominant ones, consequently constraining the overall performance enhancement of multimodal learning. To tackle this issue, current approaches attempt to mitigate modality competition in various ways, but their effectiveness is still limited. To this end, we propose an Euler Representation Learning-based Modality Rebalance (ERL-MR) strategy, which reshapes the underlying competitive relationships between modalities into mutually reinforcing win-win situations while maintaining stable feature optimization directions. Specifically, ERL-MR employs Euler's formula to map original features to complex space, constructing cooperatively enhanced non-redundant features for each modality, which helps reverse the situation of modality competition. Moreover, to counteract the performance degradation resulting from optimization drift among modalities, we propose a Multi-Modal Constrained (MMC) loss based on cosine similarity of complex feature phase and cross-entropy loss of individual modalities, guiding the optimization direction of the fusion network. Extensive experiments conducted on four multi-modal multimedia datasets and two task-specific multi-modal multimedia datasets demonstrate the superiority of our ERL-MR strategy over state-of-the-art baselines, achieving modality rebalancing and further performance improvements.

Federated learning (FL) has emerged as a crucial approach in the realm of distributed machine learning, providing a framework for training models on decentralized data while preserving data privacy. This paradigm has established itself as an effective solution for deploying artificial intelligence technology in scenarios associated with the Internet of Things (IoT). Despite its potential, FL faces encounters several challenges, particularly the limited computational and communication capabilities of some local clients, which can hinder further advancement. Such constraints limit the effective implementation and utilization of deep neural networks (DNNs) with numerous parameters on IoT devices. Our study tackles this issue by utilizing Group Lasso for model sparsification and pruning, aimed at lowering the computational and communication demands on IoT devices. Moreover, this article proposes a Group Lasso-enabled FL model pruning strategy specifically tailored for IoT, designed to reduce the size of model parameters, and provides theoretical guarantees of FL convergence. Empirical analysis across multiple models and data sets demonstrates that our method effectively halved the parameters in fully connected layers during federated training. This substantial reduction is achieved with minimal impact on accuracy, thus preserving the integrity of model performance and providing a competitive edge over existing methodologies.

Intrusion Detection Systems (IDS) integrated with Machine Learning (ML) techniques have proven to be effective defenses against the increasing cybersecurity attacks in the Artificial Intelligence of Things (AIoT) domain. Privacy concerns have prompted the emergence of Federated Learning (FL) as a promising solution for AIoT intrusion detection. Despite their potential, FL-based IDSs still face challenges related to class-imbalanced data and Non-Independent and Identically Distributed (non-IID) data among AIoT devices. These challenges hinder FL from learning meaningful features from the data, thus impeding the convergence of the learning process. To tackle these issues, this paper proposes a Clustering-enabled Federated Meta-Training (CFMT) framework for AIoT intrusion detection. The proposed CFMT framework effectively addresses the negative impact of imbalanced and non-IID data. Specifically, we design a data-and model-agnostic meta-sampler that adaptively balances local datasets, thereby mitigating the data imbalance problem. Additionally, we propose a dynamic clustering algorithm that selectively eliminates the local models affected by the training state bias caused by non-IID data, thereby addressing the non-IID data issue. Extensive case studies on two real-world datasets demonstrate the superior performance of the proposed CFMT framework compared to existing solutions, including federated non-IID algorithms and federated imbalanced learning algorithms, in terms of IDS performance. Our code and data are available at https://gitee.com/mindspore/models/tree/master/research/cv/HDFL-IDS-Meta.

Federated learning (FL) is an emerging technology for empowering various applications that generate large amounts of data in intelligent cyber–physical systems (ICPS). Though FL can address users’ concerns about data privacy, its maintenance still depends on efficient incentive mechanisms. For long-term incentivization to participants in data federation under dynamic environments, deep reinforcement learning as a promising technology has been extensively studied. However, the nonstationary problem caused by the heterogeneity of ICPS devices results in a serious effect on the convergence rate of existing single-agent reinforcement learning. In this article, we propose a multiagent learning-based incentive mechanism to capture the stationarity approximation in FL with heterogeneous ICPS. First, we formulate the secure communication and data resource allocation problem as a Stackelberg game in FL with multiple participants. Then, to tackle the heterogeneous problem, we model this multiagent game as a partially observable Markov decision process. In particular, a multiagent federated reinforcement learning algorithm is proposed to learn the allocation policies efficiently by dwindling variances in policy evaluation caused by interaction among multiple devices without the requirement of sharing privacy information. Moreover, the proposed algorithm is proved to attain convergence at an expected rate. Finally, extensive experimental results demonstrate that our proposed algorithm significantly outperforms baseline approaches.

Transient stability assessment (TSA) plays an important role to ensure the safe operation of the power system in Internet of Energy (IoE). Many time-domain simulation (TDS)-based and transient energy function (TEF)-based methods have been proposed to assess the transient stability of the power system. With the wide area measurement system (WAMS) and the phasor measure units (PMUs) applied to observe the real-time data, methods of TSA based on the machine learning and data-driven are continuously studied. These kinds of methods can only assess the transient stability of the power system when subjected to large disturbances. However, these kinds of methods cannot infer the type of event which leads to the collapse of the power system. In this article, the gated graph neural network (GGNN) is applied to assess the power system transient stability and infer the type of event leading the instability of the power system. First, conditional generative adversarial network (CGAN) is applied to generate unstable samples making the training data more balanced. With the balanced data graph-structured and used to train the GGNN-based TSA model, the GGNN-based TSA model achieves better performances. Finally, the real-time data is input into the trained TSA model and the transient stability of the power system is assessed. When the power system is considered unstable, the proposed TSA model can also infer the type of event leading the instability of the power system, classifying the unstable state to the corresponding event. Simulations performed on the New England 39-bus system verify the effectiveness of the proposed method.

In the era of Industry 4.0, the Industrial Internet of Things (IIoT) has been applied to help physical entities to access real-time data in the network and share critical information. However, both the decentralization of IIoT and the heterogeneity of different devices constituting IIoT also pose a serious threat to secure communication between entities. Although the encryption technologies can ensure the confidentiality of communication data, the security of key becomes more important for devices in IIoT. Recently, some asymmetric group key agreement (AGKA) protocols have been proposed to negotiate a common encryption key for group members, and each group member holds its decryption key. However, the existing AGKA protocols fail to effectively control the access of agreement participants in IIoT. To solve this problem, in this article, we propose an AGKA protocol based on blockchain and attribute for IIoT. The proposed protocol not only realizes the access control of agreement participants but also realizes the automation of access control, the tamper resistance, and the nonrepudiation of the agreement process. The security and performance analysis of the proposed protocol show that it is more secure and effective compared with the existing AGKA protocols.

Federated Edge Learning (FEL) allows edge nodes to train a global deep learning model collaboratively for edge computing in the Industrial Internet of Things (IIoT), which significantly promotes the development of Industrial 4.0. However, FEL faces two critical challenges: communication overhead and data privacy. FEL suffers from expensive communication overhead when training large-scale multi-node models. Furthermore, due to the vulnerability of FEL to gradient leakage and label-flipping attacks, the training process of the global model is easily compromised by adversaries. To address these challenges, we propose a communication-efficient and privacy-enhanced asynchronous FEL framework for edge computing in IIoT. First, we introduce an asynchronous model update scheme to reduce the computation time that edge nodes wait for global model aggregation. Second, we propose an asynchronous local differential privacy mechanism, which improves communication efficiency and mitigates gradient leakage attacks by adding well-designed noise to the gradients of edge nodes. Third, we design a cloud-side malicious node detection mechanism to detect malicious nodes by testing the local model quality. Such a mechanism can avoid malicious nodes participating in training to mitigate label-flipping attacks. Extensive experimental studies on two real-world datasets demonstrate that the proposed framework can not only improve communication efficiency but also mitigate malicious attacks while its accuracy is comparable to traditional FEL frameworks.

Ultrareliable and low‐latency connection (URLLC) is one of the novel features in 5G networks and subsequent generations, in which it targets to fulfill stringent requirements on data rates, reliability, and availability. Moreover, the multiconnectivity concept is introduced to meet these requirements, where multiple different technologies are connected simultaneously, and the data packet is duplicated and transmitted from multiple transmitters. To this end, in this paper, we present an analysis, model, and method to ensure the reliability of data delivery when organizing URLLC in 5G networks. In addition, a new approach based on the organization of multiple connections (multiconnectivity) and duplication of transmitted data is considered. Further, an analytical model is presented for assessing the probability of failure, taking into account the traffic intensity, the probability of failure of elements, and the number of used connections. Moreover, an efficient method is proposed for increasing the reliability of data delivery by optimizing the number of connections. Further, a multiconnectivity‐based URLLC model has been built for evaluating the proposed method and verifies that the optimal number of routes for data delivery between the user and the point of service can be obtained, where the probability of losses and equipment reliability are jointly considered. Finally, detailed analysis of results shown that with “equal” routes in terms of load (with an equally probable traffic distribution) and the probability of equipment failure, the optimal number of routes can be found, at which the minimum probability of losses is achieved.

In the explosive growth of time-series data (TSD), the scale of TSD suggests that the scale and capability of many Internet of Things (IoT)-based applications has already been exceeded. Moreover, redundancy persists in TSD due to the correlation between information acquired via different sources. In this article, we propose a cohort of dominant data set selection algorithms for electricity consumption TSD with a focus on discriminating the dominant data set that is a small data set but capable of representing the kernel information carried by TSD with an arbitrarily small error rate less than$\varepsilon $. Furthermore, we prove that the selection problem of the minimum dominant data set is an NP-complete problem. The affine transformation model is introduced to define the linear correlation relationship between TSD objects. Our proposed framework consists of the scanning selection algorithm with$O({n^{3}})$time complexity and the greedy selection algorithm with$O({n^{4}})$time complexity, which are, respectively, proposed to select the dominant data set based on the linear correlation distance between TSD objects. The proposed algorithms are evaluated on the real electricity consumption data of Harbin city in China. The experimental results show that the proposed algorithms not only reduce the size of the extracted kernel data set but also ensure the TSD integrity in terms of accuracy and efficiency.

Generative Adversarial Network (GAN) and its variants serve as a perfect representation of the data generation model, providing researchers with a large amount of high-quality generated data. They illustrate a promising direction for research with limited data availability. When GAN learns the semantic-rich data distribution from a dataset, the density of the generated distribution tends to concentrate on the training data. Due to the gradient parameters of the deep neural network contain the data distribution of the training samples, they can easily remember the training samples. When GAN is applied to private or sensitive data, for instance, patient medical records, as private information may be leakage. To address this issue, we propose a Privacy-preserving Generative Adversarial Network (PPGAN) model, in which we achieve differential privacy in GANs by adding well-designed noise to the gradient during the model learning procedure. Besides, we introduced the Moments Accountant strategy in the PPGAN training process to improve the stability and compatibility of the model by controlling privacy loss. We also give a mathematical proof of the differential privacy discriminator. Through extensive case studies of the benchmark datasets, we demonstrate that PPGAN can generate high-quality synthetic data while retaining the required data available under a reasonable privacy budget.

Biometric binarization is the feature-type transformation that converts a specific feature representation into a binary representation. It is a fundamental issue to transform the real-valued feature vectors to the binary vectors in biometric template protection schemes. The transformed binary vectors should be high for both discriminability and privacy protection when they are employed as the input data for biometric cryptosystems. In this paper, we propose a novel binarization scheme based on random projection and random Support Vector Machine (SVM) to further enhance the security and privacy of biometric binary vectors. The proposed scheme can generate a binary vector of any given length as an ideal input for biometric cryptosystems. In addition, the proposed scheme is independent of the biometric feature data distribution. Several comparative experiments are conducted on multiple biometric databases to show the feasibility and efficiency of the proposed scheme.