TinyML enables efficient on-device machine learning for resource-constrained edge devices, addressing privacy and computational challenges in real-world applications. This paper presents a TinyML-based Edge-Cloud system using the FOMO (Faster Objects, More Objects) model deployed on a constrained edge device to analyze visitor engagement in Smart Museums. Unlike conventional camera-based tracking, our approach ensures privacy-preserving analytics by processing visual data locally, extracting only anonymized metrics (e.g., dwell time) to allow for holding power (HP) calculations. We optimize FOMO via data augmentation (DA) and quantization, significantly improving performance; with the optimal configuration, we achieved 86.5% F1-score at 155ms inference latency (a 9.2× speedup over the baseline configuration) while maintaining 88% precision. Our results demonstrate that TinyML-enhanced edge deployment offers a robust, low-latency, and privacy-compliant solution for museum analytics, outperforming traditional cloud and wearable-based approaches.

Infrared image generation is essential in scenarios with low illumination or complex environments, but the scarcity of aligned visible–infrared data and the lack of physical realism in generated results remain key challenges. However, existing generative models often overlook the thermodynamic principles underlying infrared imaging, resulting in synthetic images that are visually plausible yet physically inaccurate. In this paper, we propose Infrared Physics-guided Latent Diffusion (IPLD), a novel framework that integrates physics-guided modeling into a latent diffusion process for high-fidelity synthesis of infrared images. Central to IPLD is the Temperature–Emissivity–Environmental Radiance (TeR) decomposition, which decomposes thermal signals into temperature, emissivity, and environmental radiance components, governed by the laws of blackbody radiation. To enhance the environmental radiance modeling, we introduce Environmental Radiance Map Estimation (ERME), a hybrid local–global estimation mechanism that preserves both spatial detail and thermal consistency. Furthermore, a novel Skip Connection Diffusion Transformer (SCDT) is proposed to strengthen and balance semantic structure and fine-grained details during image reconstruction. Extensive experiments on public datasets demonstrate that IPLD outperforms state-of-the-art Generative Adversarial Network (GAN)-based and diffusion-based methods, achieving superior results in Structural Similarity Index Measure (SSIM), Peak Signal-to-Noise Ratio (PSNR), Learned Perceptual Image Patch Similarity (LPIPS), and Fréchet Inception Distance (FID) metrics. Ablation studies validate the complementary value of TeR, ERME, and SCDT in improving radiative realism. Our approach establishes a new paradigm for physically grounded image translation, offering enhanced generalization and reliability for downstream perception tasks such as target detection.

Digital Twins (DTs) have emerged as essential tools for virtualizing and enhancing Cyber-Physical Systems (CPS) by providing synchronized digital counterparts that enable monitoring, control, prediction, and optimization. Initially conceived as passive digital shadows, DTs are increasingly evolving into intelligent and proactive entities, enabled by the integration of Artificial Intelligence (AI). Among these advancements, Opportunistic Digital Twins (ODTs) represent a novel class of DTs: living, AI-aided, and actionable models that opportunistically exploit edge-cloud resources to deliver enriched and adaptive representations of physical entities and processes. However, despite their promise, current research lacks systematic engineering methods to ensure reliable coordination, determinism, and real-time responsiveness of ODTs in distributed and resource-constrained CPS. This article addresses this gap by introducing an engineering approach to build dependable and efficient ODTs by leveraging the deterministic concurrency, explicit timing semantics, and disciplined event handling of Lingua Franca (LF). The approach is exemplified through a Smart Traffic Management case study centered on Emergency Vehicle Preemption (EVP), where the ODT dynamically selects AI models based on runtime conditions while ensuring deterministic coordination across distributed nodes. Experimental results confirm the feasibility and effectiveness of our methodology, underscoring the potential of LF-based ODT engineering to enhance reliability, adaptability, and scalability in intelligent and distributed CPS deployments.

In recent years, machine learning (ML) and deep learning (DL) have emerged as key technologies for balancing energy efficiency and thermal comfort in Internet of Things (IoT) based Smart Buildings (SBs). Among the various components of SBs, the heating, ventilation, and air conditioning (HVAC) system plays a critical role, as it significantly influences both energy consumption and occupant comfort. In this context, accurately predicting indoor temperatures is essential for optimizing HVAC operations, resulting in enhanced energy efficiency, improved comfort, and lower energy costs. To address this challenge, this paper proposes Advanced Temperature Management of Smart Building (ATM-SB), a hybrid DL approach that combines Long-Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks to predict indoor temperature at 15, 30, and 60 minute intervals using multivariate sensor data. Using the ATM-SB approach, a prototype was developed at AEI S.r.l., Italy. It was trained and validated on a real-world dataset collected from a smart office building and laboratory across both summer and winter seasons, and further tested on four public datasets from diverse environments. ATM-SB achieves competitive performance and demonstrates robust generalization across diverse real world and public datasets, including comparisons with LSTM, GRU, CNN-LSTM, and gradient boosting, achieving an MAE as low as 0.0098 and an R 2 of up to 0.98 on real data. Statistical validation through 5-fold cross-validation and significance testing confirms the robustness and generalization of the model. By enabling accurate forecasting across diverse scenarios, ATM-SB provides an effective solution for intelligent HVAC control in next-generation SBs.

Brain tumour segmentation is a key application of AI in neuroimaging. Recently, federated learning (FL) has emerged as a strategic and increasingly relevant paradigm in neural computing due to its ability to address key challenges in large-scale neural network training, such as data access, privacy, collaborative learning, and model robustness. However, its adoption is currently hindered by high communication costs and the heterogeneity of client data. In this study, we investigated an efficient FL framework for brain tumour segmentation based on communication-aware optimization. We evaluated FedWSOComp, which integrates sparsification, quantization, and entropy-based encoding, in combination with a 3D U-Net architecture under both homogeneous and heterogeneous data distributions. The multi-institutional FeTS 2024 dataset was employed and partitioned into independent and identically distributed (IID) and non-IID settings, with an independent test set of 67 patients. An overall of 18 configurations combined sparsification rates and quantization levels. Performance was measured using Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff Distance (HD95). Experimental results demonstrated that aggressive compression caused severe degradation in segmentation quality, with HD95 exceeding 60 mm. In contrast, higher retention with finer quantization achieved the best balance between efficiency and accuracy, reaching a DSC and HD95 mm on the test set under non-IID conditions. The findings demonstrated that, when configured with moderate-to-fine quantization and high sparsification retention, FedWSOComp enabled accurate and communication-efficient federated brain tumour segmentation. This study provides quantitative evidence and practical guidance for the deployment of FL-based segmentation models in privacy-sensitive and bandwidth-constrained clinical settings. • Analyse the impact of FedWSOComp, an integrated strategy combining top-k sparsification, quantization, and entropy-based encoding. • The performance of 18 different configurations (including IID and non-IID) was evaluated systematically. • High retention (60%) with fine quantization (64 clusters) optimizes performance.

Federated Learning (FL), Continual Learning (CL), and Digital Twins (DTs) have emerged as key paradigms for the development of intelligent, adaptive, and privacy-aware systems in various domains. FL enables collaborative model training across decentralized data sources without sharing raw data, thus ensuring privacy. CL allows models to continuously learn from evolving data streams and adapt to dynamic environments, reducing the need for retraining from scratch. DTs provide accurate virtual representations of physical systems, supporting real-time monitoring, simulation, and predictive maintenance. Combining these paradigms is a recent strategy for building physical systems that are decentralized, adaptive, and continuously improve using real-time data in various contexts. For example, in industries the integration of FCL with DTs can enable factories to learn from new sensor data across distributed sites while preserving sensitive data, adapting to equipment changes, and optimizing maintenance cycles. In mobile edge computing, this combination can enhance service reliability and user experience by updating models based on fresh data and dynamic user behavior. However, their combination also amplifies the inherent challenges, such as model drift, system complexity, and resource constraints, that need to be managed. Despite its promising potential, no existing surveys offer a focused and structured analysis of their intersection. This survey presents the first structured and comprehensive analysis of these three paradigms, highlighting not only existing approaches but also discussing their potential synergies and conflicts, outlining open research questions that must be addressed to unlock their full potential in real-world applications. • Identifies key challenges and outlines future research directions in FCL+DT. • Taxonomy and analysis of FCL methods applied to real-world DT systems. • First survey on the convergence of FCL and Digital Twin technologies.

Smart contracts are increasingly being adopted in modern supply chain (SC) systems, offering a transformative shift from traditional centralized models to decentralized, automated, and trustless processes. However, vulnerabilities in the chain of smart contract represent a critical threat to the reliability and security of supply chain ecosystems, often arising from intricate logical flaws, unintended inter-contract interactions, or improper handling of user input—issues that remain difficult to uncover through conventional testing or manual auditing. This paper introduces an innovative deep learning–based methodology that transforms smart contract bytecode into image representations, enabling precise and efficient classification of diverse vulnerability patterns. Unlike existing approaches that rely on source code availability or dynamic execution, the proposed framework operates independently of source code and circumvents the limitations inherent to dynamic analysis, thereby offering a versatile and system-agnostic solution. Experimental evaluations conducted on a newly curated dataset collected from multiple publicly available repositories demonstrate the robustness of the proposed method, achieving 92.24% accuracy and an 89.06% F1-score. Beyond its strong empirical performance, the framework ensures reproducibility, data transparency, and adaptability across heterogeneous blockchain environments. Collectively, these contributions establish a comprehensive and accessible foundation for enhancing the detection, mitigation, and overall resilience of blockchain smart contracts.

Digital Twins (DT) have evolved from static digital mirrors into executable cyber-physical counterparts that predict, optimize, and control complex systems. However, the practical deployment of DT in Internet of Things (IoT) environments suffers from limited data fidelity, model brittleness, and resource constraints across the edge–cloud continuum. Generative DT (GDT) is DT augmented with Generative AI (GenAI). They enable the synthesis of high-fidelity data, bridge model-driven and data-driven paradigms, and provide adaptive decision support under uncertainty. This paper systematically reviews the research progress on GDT in the Manufacturing Internet of Things (MIoT), covering system architectures, key enabling technologies, and representative application scenarios. It also summarizes the main limitations of existing studies and outlines future research directions.

With the rapid development of Industry 5.0, the Industrial Internet of Things (IIoT) plays an increasingly important role as a key information infrastructure supporting data collection, transmission, and decision-making. Federated Learning-enabled IIoT (FL-enabled IIoT) systems deploy artificial intelligence (AI) models at the network edge and utilize model aggregation mechanisms to facilitate efficient data processing and intelligent decision-making. However, anomalies occurring in local nodes can propagate through the aggregation process, leading to model contamination and performance degradation, thereby compromising overall system reliability. To address this issue, this paper proposes a reliability model for FL-enabled IIoT systems. In this model, we systematically describe the entire process of model performance degradation caused by node failures and its impact on system task reliability. This includes the effects of multiple functional failures induced by node faults (i.e., data loss, communication interruption, and computational resource degradation) on model performance, as well as the failure propagation process caused by model contamination and data quality deterioration. Additionally, to evaluate the impact of model performance variations on practical production tasks, a task-oriented reliability metric is proposed. Simulation and experimental results demonstrate that the proposed modeling approach effectively characterizes the model performance degradation process and task reliability under node failure conditions.

Human activity prediction is crucial for enabling intelligent smart home services, yet it is often hindered by the scarcity of high-quality, multi-dimensional datasets. Existing datasets are typically fragmented, capturing either long-term activity sequences or short-term device interactions, but rarely both in a unified manner. Traditional data collection methods are costly and time-consuming, while conventional simulation techniques struggle to generate diverse and logically coherent behavior sequences. To address these limitations, we propose SmartLLM, a novel Large Language Model (LLM)-based simulation framework for automated generation of multi-dimensional smart home datasets. SmartLLM simulates simulated agents with distinct profiles (e.g., old man, remote worker, holiday maker) performing daily activities within configurable home environments, generating temporally aligned sequences across Activity-Device-Sensor dimensions. We generate two months of simulated data for three user profiles and validated their plausibility through activity distribution visualization, statistical perplexity analysis, and case studies. Multi-dimensional feature validation experiments further demonstrate that our multi-dimensional data significantly enhances the accuracy of activity prediction models compared to using single-dimensional features. This work successfully addresses key bottlenecks in smart home data acquisition and provides a scalable, high-quality data foundation for advancing smart home algorithm research. The code is available at https://github.com/HuankeZheng/SmartLLM.

The rapid proliferation of Internet of Vehicles (IoV) technology has significantly enhanced traffic efficiency and driving safety, yet it has also introduced severe security challenges due to malware and cyberattacks. This paper proposes a novel Fractional-Order Attack-Defense Game model (FADG-IoV) to address dynamic malware propagation in IoV environments. By integrating fractional-order dynamics, the model accounts for communication delays, traffic density heterogeneity, and channel fading, capturing memory-dependent behaviors inherent in IoV systems. We introduce the Fractional-Order Attack-Defense Game Sparse Attention Multi-Agent Soft Actor-Critic (FADG-SMASAC) algorithm, a model-free reinforcement learning approach that leverages sparse attention mechanisms to achieve adaptive and robust control without requiring a known system model. Through multi-baseline experiments, we validate the FADG-IoV model and FADG-SMASAC algorithm, demonstrating superior convergence, scalability, and robustness compared to existing methods. Our findings highlight the effectiveness of fractional-order game-theoretic strategies in enhancing IoV security against dynamic malware threats, paving the way for future research in adaptive defense mechanisms.

This study investigates how heterogeneous information flow in the modular large language model (LLM) pipeline architectures influences hallucination behavior in multidocument and multimodal analysis tasks. Past research in this subject focused on building solutions by consolidating documents first or processing them separately; however, measuring the architectural effect was missing. This research work examines how different LLM architecture pipelines impact hallucination behavior when dealing with multiple heterogeneous documents. To enable a rigorous comparison of downstream reasoning strategies, we introduce an explicit unstructured-to-structured transformation layer incorporating a data stratification framework (DSF) that categorizes heterogeneous inputs into standardized structural classes and applies modality-specific preprocessing for temporal, spatial, visual, and dense narrative data. The experiments are conducted on approximately 3400 multimodal data files across multiple financial industry sectors and educational documents. The results exhibit that parallel extraction with reconciliation consistently achieves the lowest hallucination rates, reducing errors by up to 33% relative to baseline pipelines and yielding the lowest normalized semantic entropy (0.38). The findings are compared with state-of-the-art retrieval-augmented generation (RAG) model used by the Fin-tech industry and further evaluated by human experts, demonstrating that hallucination behavior is strongly shaped by pipeline structure rather than model scale alone. Industry validation reveals that pipeline 3 (parallel extraction + reconciliation) achieves a hallucination score of 0.112, compared to the industry production RAG system (0.135). The results achieved through systematic architectural comparison substantiate that pipeline design can independently determine reliability even when base models are held constant.

Alzheimer’s disease (AD) is considered to be a significant health challenge that affects the cognitive ability of elderly people. The effects can only be slowed down if the disease is detected at an early stage. Researchers have extensively explored the use of machine learning algorithms to ensure early detection and prediction. However, effective models are complex, hence limiting their interpretability and privacy. Federated learning (FL) approaches have also been proposed to add privacy aspect to the machine learning models, however, FL methods are vulnerable to model related attacks. To address this we propose Agentic ElderFedLearn, a novel framework that proceeds in the following steps: 1) model healthcare institutions as autonomous artificial intelligence (AI) agents training local models on multimodal data [electronic health record (EHR) and synthetic magnetic resonance imaging (MRI)]; 2) apply personalized differential privacy (DP) to gradients, adapting budgets based on dataset size and sensitivity; 3) use multiagent reinforcement learning (MARL) to optimize agent interactions, such as privacy adjustments and communication; and 4) perform effective aggregation via weighted trimmed mean to defend against attacks. This innovation ensures privacy, handles heterogeneity, and achieves 94% accuracy with 0.93 F1-score, outperforming centralized approaches while using synthetic data.

In recent years, AI models have demonstrated a growing ability to interpret complex human behavior, including emotional states. While vision-language models (VLMs) have shown significant capabilities in this area, their application to group emotion recognition remains a challenging task. This challenge is compounded by the limitations of traditional evaluation metrics, which, due to their binary logic, fail to capture the nuanced semantic relationships between different emotional states. To address this evaluation gap, our work pioneers a systematic method for evaluating cohesive group emotion, moving beyond the simple aggregation of individual states. We achieve this within azero-shot paradigm, establishing a lightweight baseline that eliminates the need for model fine-tuning. We present GEM-AI: a group-based Emotion recognition method, a novel group emotion detection framework based on semantic similarity. We apply this framework to a comparative study of several open-source VLMs, includingLLaVA,MiniCPM,Deepseek-VL, andQwen-VL, by reformulating precision, recall, and F1-score into “soft metrics” for a more semantically coherent assessment. Our results demonstrate that, when assessed with the GEM-AI framework, these models achieve high performance, reaching semantic accuracy of approximately 80%. This reveals a sophisticated comprehension of group emotion that is significantly underestimated by conventional metrics, which achieved only around 60% accuracy in our experiments.

The proliferation of hate speech on social media presents an escalating threat to both public discourse and individual mental well-being. Traditional strategies that prioritize detection and removal often neglect to engage directly with hate speakers or address the underlying causes of their hostility. This article proposesHateMediator, a dialogue-based intervention framework that fine-tunes large language models (LLMs) to generate persuasive, context-aware counter-hate speech. The framework emphasizes two core aspects: the generation of effective counter-hate responses and their evaluation through multiturn dialogues. Our fine-tuning approach integrates tutorial-based learning with critical token guidance, enabling LLMs to recognize and reproduce strategic rhetorical patterns observed in expert interventions. To support training and evaluation, we introduce theMedHatedataset, grounded in social science theory, comprising complete dialogue records from 85 real-world hate incidents (including 255 dialogues), expert-crafted counter-responses, and feedback from the original hate speakers. Experimental results show thatHateMediatorconsistently outperforms baseline LLMs across multiple evaluation dimensions. This study advances both the technical frontier of hate speech intervention and the ethical deployment of LLMs in addressing complex social issues.

In a socially tense environment with rising emotional pressure, understanding the spread patterns of group emotions-particularly negative emotions-is crucial for identifying social risks. Extensive research has explored emotion contagion, often using propagation models where node state transitions rely on preset probabilities. However, these methods introduce randomness, making them less reflective of real-world dynamics by failing to capture individual node behaviors and interactions in emotional networks. To address this, our study introduces a novel approach integrating text-based emotion recognition with propagation models, reconstructing emotion contagion at an individual level. This model enhances traditional nodes with multihop agents driven by text emotion analysis, where agents record and respond to neighbors' emotional states. As a result, emotion spread becomes a deterministic process, with individualized infection rates reflecting node variability. We categorized nodes based on emotional states, creating corresponding agent types to form the dynamic agent-based emotion model (AEmo). Tests on real-world and scale-free networks show this method effectively predicts group negative emotion spread and provides insight into individual emotion evolution, validating the model's effectiveness.

This study explores the innovative application of a nearable solution (i.e., mattress) based on fiber Bragg grating (FBG) technology for continuously monitoring of critical sleep-related biomarkers. Based on biocompatible silicone compounds, the mattress embeds thirteen strategically positioned FBG sensors to detect bed occupancy, sleeping posture, respiratory rate (RR), and heart rate (HR). Our experimental protocol involves ten participants who underwent simulated sleeping conditions to evaluate the mattress's performance across different postures and respiratory patterns. Employing traditional machine learning algorithms, including decision tree, support vector machine (SVM), and Naïve-Bayes classifiers, the mattress achieves 100% accuracy in bed occupancy detection. It also effectively distinguishes between axial and lateral sleeping positions, with SVM achieving the highest accuracy of 78.4% for axial versus lateral differentiation and convolutional neural networks achieving 75.9% in distinguishing left from right positions. Additionally, for most participants, the system successfully estimates RR and HR with mean absolute errors of less than 0.7 breaths per minute and 4 bpm, respectively, across various breathing patterns in terms of frequencies and amplitudes employing different algorithms (frequency and time-domain approaches). The promising findings highlight the potential of the proposed system for a comprehensive evaluation of sleep-related breathing disorders in clinical and home settings.

In this paper, the problem of controlling autonomous vehicles in a Flexible Manufacturing System is addressed in order to optimize logistic operations. To this end, vehicles are required to navigate between machines and from/to the Load/Unload station. The core contribution of this paper is to propose a set-theoretic distributed Model Predictive Control in charge of controlling the autonomous vehicles properly integrated with a Reinforcement Learning scheme to address the routing problem. In addition, vehicles are organized as platoons in order to improve the efficiency of the overall architecture. The numerical simulation shows the effectiveness of the proposed approach.

Federated Learning (FL) facilitates decentralized model training while prioritizing data privacy. However, its effective implementation faces significant challenges, primarily related to high communication overhead and energy consumption. Quantization, a key optimization technique, plays a crucial role in enhancing energy efficiency by reducing the bit precision of model updates, thereby lowering computational and transmission costs. This paper explores the impact of quantization on energy-efficient FL, focusing on techniques such as lattice quantization and stochastic gradient quantization methods. By compressing gradient updates and model parameters, quantization significantly reduces bandwidth requirements, enables efficient model aggregation, and prolongs battery life in resource-constrained edge devices. Furthermore, we discuss the trade-offs between quantization levels, model accuracy, and energy savings, emphasizing strategies to mitigate performance degradation while maintaining robust learning. Experimental results demonstrate that quantized FL can achieve up to 50% reduction in energy consumption while maintaining competitive model accuracy. This study highlights the importance of quantization in scalable, sustainable, and energy-aware FL, paving the way for its widespread adoption in real-world applications such as smart healthcare, IoT, and edge AI.

The growing demand for computation-intensive artificial intelligence (AI) and machine learning (ML) applications necessitates carbon-neutral edge computing to enhance resource efficiency, reduce energy consumption, and promote sustainability in Industrial Internet of Things (IIoT) systems. However, reducing service latency and energy consumption while ensuring execution accuracy and a predictable carbon footprint and its associated cost remains a critical research challenge. Existing works in the literature experience significant challenges for task offloading due to a lack of edge collaboration and ineffective management of Carbon Emission Rights (CER) credits. In this paper, we have developed an optimization framework leveraging Mixed Integer Linear Programming (MILP), namely GRELMON, to jointly minimize service latency and energy consumption while maximizing task accuracy in carbon-neutral collaborative edge and cloud computing for IIoT environments. Moreover, a carbon emission forecasting model using a hybrid deep learning approach is also developed to prevent unnecessary CER purchases. The experimental results demonstrate that GRELMON outperforms state-of-the-art methods by reducing latency and energy consumption while improving the accuracy of the execution of ML tasks.

The Internet of Medical Things (IoMT) has emerged as a transformative technology platform in the healthcare sector, enabling real-time monitoring and intelligent decision-making through connected devices. However, prioritizing and offloading the massive volume of computational tasks generated by IoMT devices while minimizing latency and energy consumption poses significant challenges. Existing approaches often overlook dynamic real-time factors such as task urgency and data freshness, as well as the integration of local task processing via Device-to-Device (D2D) communication with offloading to Mobile Edge Computing (MEC) servers. In this paper, we develop a priority- and Age of Information (AoI)-Aware task offloading framework for latency and energy optimization in healthcare IoT systems, namely PRALEIT, exploiting Mixed Integer Linear Programming (MILP) problem. The developed PRALEIT system introduced probabilistic classification of IoMT tasks based on vital signs and AoI value by leveraging a Bayesian classifier. The experimental results depict that the PRALEIT system significantly reduces task execution delay and energy consumption compared to state-of-the-art models, ensuring reliable and sustainable healthcare services.

Chronic stress significantly undermines cognitive function and well-being, yet its reliable detection remains elusive due to inter-individual psychobehavioral heterogeneity. To the best of our knowledge, this study is the first to introduce a novel multimodal framework synergizing psychometric traits (conscientiousness, neuroticism) and 64-channel electroencephalography (EEG) for precision-driven chronic stress classification. We collected psychophysiological data from 21 subjects, including Big Five personality inventories, Perceived Stress Scale (PSS) scores, and EEG recordings. Statistical analysis revealed a strong inverse correlation between conscientiousness and perceived stress (r = -0.59), while neuroticism showed a weaker positive association (r = 0.12). Power spectral density (PSD) features from artifact-corrected EEG were fused with trait metrics to train SVM and KNN classifiers. The proposed framework achieved state-of-the-art accuracy (SVM: 94.7%; KNN: 89.4%), with neuroticism emerging as a critical predictor alongside beta/gamma-band spectral markers—a novel finding underscoring its role in stress pathophysiology. This work pioneers the integration of personality-aware analytics with neurophysiological biomarkers for stress phenotyping, establishing a new paradigm for personalized mental healthcare. By demonstrating the feasibility of trait-guided machine learning, our contributions advance scalable, individualized interventions, addressing a critical gap in precision psychiatry. These results lay the groundwork for adaptive digital health systems that leverage multimodal data to mitigate chronic stress and enhance quality of life.

Human Activity Recognition (HAR) represents a significant domain within pervasive computing, facilitating a diverse array of applications ranging from healthcare to smart environments. Traditional HAR models suffer from several challenges, including data privacy and the distributed participation of heterogeneous resource-constrained devices. To mitigate these challenges, the research community popularly uses federated learning (FL). However, selecting clients in FL is a critical issue, mainly when there is a combination of resource-constrained heterogeneous devices. This paper proposes a resource-and performance-aware client selection algorithm for HAR, namely EAPD-CS, that amalgamates the benefits of FL with energy efficiency. The framework allows for the training of machine learning models across multiple devices without the necessity of sharing raw data, thereby preserving user privacy. Additionally, it employs energy-aware strategies to diminish the carbon footprint and reduce the computational costs typically linked to traditional cloud-based HAR systems. Experimental results indicate that the proposed framework achieves more than 90% accuracy, comparable to centralized models, while significantly lowering energy consumption and improving the robustness of the model. This work contributes to the evolving field of green AI, delivering an effective, privacy-preserving, and environmentally sustainable approach for HAR applications.

Task failures in decentralized Internet of Things (IoT)-edge computing environments not only lead to inefficiencies, increased latency, and resource wastage but can also introduce system instability and cause application malfunctions. These failures may arise due to network disruptions, resource constraints, or inefficient task scheduling, ultimately affecting the overall reliability and performance of IoT-edge systems. This study presents a novel Long Short-Term Memory (LSTM)-based Federated Learning (FL) framework for proactive task failure prediction, ensuring adaptive scheduling and efficient resource utilization. Unlike existing conventional methods, our approach personalizes failure prediction per device, addressing heterogeneous execution characteristics while preserving data privacy. By integrating LSTM with FL, we improve the failure detection accuracy and reduce unnecessary task executions. We first trained all models using Federated Learning (FL) and then conducted a comparative analysis of Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and LSTM. Our findings show that LSTM achieves the highest accuracy and F1 score, while CNN excels in recall and energy efficiency. These insights validate the effectiveness of our FL-based failure prediction framework and highlight the advantages of model personalization for dynamic decentralized IoT-edge environments.

Industrial Wireless Sensor Networks (IWSNs) play a critical role in real-time monitoring and data collection in smart factories. However, energy constraints in sensor nodes significantly limit the network lifespan. In addition, traditional simulation methods overlook the impact of industrial noise, reducing the truthfulness of experimental results. To address these challenges, we propose an Automated Guided Vehicle-Integrated Noise-Aware Adaptive Clustering (A-INAC) algorithm. The algorithm incorporates an Industrial Wireless Noise Model (IWNM) to reflect noise characteristics in the factory environment and optimizes the selection of cluster directors to achieve more balanced energy consumption. In addition, a hierarchical transmission strategy leveraging the mobility of AGVs is designed to meet large-scale network transmission needs. Simulation results demonstrate that the A-INAC algorithm can effectively reduce network energy consumption and extend network lifetime by 39% and 118% compared to LEACH and LEACH-C, respectively.

The increasing diffusion of Smart Environments enabled by the Internet of Things (IoT) technologies has evidenced the limitations of traditional Internet Protocol (IP), thus pushing for a paradigm shift from host-centric to Information-Centric Networking (ICN). The Named Data Networking (NDN) is a particular ICN implementation that prospects more efficient and effective communication and service provision, reason why it is widely considered as an enabler towards Future Internet. Driven by the PRISMA methodology, in this work we systematically survey the current literature and analyze opportunities and limitations of NDN adoption within Smart Environments, targeted application areas, adopted technologies and research gaps. In particular, by means of a research framework, we highlight how, by shifting from the traditional IP-based to NDN, Smart Environments can benefit from unseen degrees of mobility, scalability, security and performance, paving the way to innovative and cutting-edge cyberphysical services.

Many-objective computation offloading (MOCO) has emerged as a critical research issue in vehicular edge computing. A key challenge in the MOCO problem is how to optimize task offloading under limited edge computing resources to effectively balance multiple objectives, such as latency, energy consumption, and load balancing. To address this challenge, we formulate the MOCO problem by modeling the task computation and offloading procedure of vehicle terminals based on queuing theory, which aims to minimize the average delay time, average energy consumption, and average offloading cost for each vehicle terminal task, as well as the average load variance of edge resources. To tackle the MOCO problem, we propose a novel evolutionary algorithm based on Bayesian Maximum Entropy and incremental learning (BMEILEA) for efficient optimization of all objectives. A novel many-objective fitness evaluation mechanism based on Bayesian maximum entropy is proposed to evaluate and select solutions in the evolving population. An adaptive dynamic reference point strategy based on incremental learning is developed to effectively guide the evolutionary process. Extensive experimental results show that BMEILEA outperforms other well-known many-objective algorithms in solving the MOCO problem and achieves better convergence and diversity in the obtained nondominated solutions.

After geohazards occur, conducting rapid and sustainable secondary geohazard monitoring plays a crucial role in reducing secondary geohazard risks. However, geohazard situations vary across different areas and dynamically change with the development of geohazards. Therefore, ensuring timely data collection and the ability to dynamically adjust to changes in geohazards poses significant challenges in geohazard monitoring scenarios. This article proposes a low-latency data collection scheme considering data importance levels (LLDCL), which prioritizes data collection from high-importance sensor nodes (SNs) while still collecting data from lower importance SNs. Given the potential for sudden events in geohazard monitoring scenarios that may require adjustments to the emergency levels of monitoring points, this article introduces a deep reinforcement learning (DRL) algorithm for unmanned aerial vehicles (UAVs) path planning based on weighted age of information (DRL-WAoI). This algorithm enables UAVs to respond quickly to dynamic environments by adjusting their flight paths in real time. Furthermore, considering the limited battery capacity of UAVs, this article establishes a token-based energy trading model between UAVs and the base station (BS) to facilitate UAV recharging. Simulation experiments show that the LLDCL scheme can effectively adapt to the dynamically changing conditions of geohazard monitoring scenarios, providing a viable solution for UAV data collection and transmission.

The combination of wearable sensors and competitive sports provides quantitative information for scientific training, effectively assisting athletes in improving their athletic performance. This study presents a technical framework for athletic sports assessment in competitive swimming based on body-area sensor networks. In our approach, wearable inertial sensor nodes are placed on specific body parts of the athletes to capture motion data during different competitive swimming strokes. Multiwearable inertial sensor nodes are worn on specific body parts of athletes for real-time monitoring motion data during training sessions. A motion intensity detection-based error-state-Kalman-filter algorithm is proposed for multisensor data fusion. Additionally, through kinematic statistical analysis, the characteristics of joint motion during training are clearly explained. Furthermore, a deep learning network that fuses sensor time series and human skeleton graphs is proposed for different stroke phase segmentation, enabling quantitative measurement of motion phases, and several baseline classifiers are chosen for comparison to validate the robustness of our phase segmentation method. We also investigate the sensor combination selection issue during the phase segmentation process to determine the optimal sensor configuration. Our approach provides a scientific solution for the integration of wearable sensors and competitive sports, contributing to the high-quality development of the next generation of smart sports.

Abstract In the past, most strategies were mainly designed to focus on stocks or futures as the trading target. However, due to the enormous number of companies in the market, it is not easy to select a set of stocks or futures for investment. By investigating each company’s financial situation and the trend of the overall financial market, people can invest precisely in the market and choose to go long or short. Moreover, how to determine the position size of the transaction is also a problematic issue. In the past, many money management theories were based on the Kelly criterion. And they put a certain percentage of their total funds into the market for trading. Nonetheless, three massive problems cannot be overcome. First, futures are leveraged transactions, and extra funds must be deposited as margin. It causes that the position size is hard to be estimated by the Kelly criterion. The second point is that the trading strategy is difficult to determine the winning rate in the financial market and cannot be brought into the Kelly criterion to calculate the optimal fraction. Last, the financial data are always massive. A big data technique should be applied to resolve this issue and enhance the performance of the framework to reveal knowledge in the financial data. Therefore, in this paper, a concept of converting the original futures trading strategy into options trading is proposed. An LSTM (long short-term memory)-based framework is proposed to predict the profit probability of the original futures strategy and convert the corresponding daily take-profit and stop-loss points according to the delta value of the options. Finally, the proposed framework brings the results into the Kelly criterion to get the optimal fraction of options trading. The final research results show that options trading is closer to the optimal fraction calculated by the Kelly criterion than futures trading. If the original futures trading strategy can profit, the benefits after converting to options trading can be further superior.

In this paper, autonomous vehicles are considered for addressing logistic operations in manufacturing systems. The starting idea consists in organizing a given group of autonomous robots/vehicles in a finite set of platoons in charge to accomplish prescribed job(s) within the manufacturing system. Three aspects are then needed to be formally outlined: task scheduling, routing decisions and command inputs computations. Here, a new distributed multi-layer architecture has been conceived by using three methodologies: timed colored Petri nets, deep reinforcement learning and model predictive control. Roughly speaking, timed colored Petri nets are exploited to formally model the manufacturing system so that an optimal scheduling task complying with the required jobs and the available vehicles is derived; then, run-time routing decisions are obtained by using a distributed reinforcement learning algorithm which exploits the available information provided by the vehicle sensor module; finally, the distributed model predictive control algorithm is built by resorting to a set-theoretic approach where most of the computations are off-line performed. A flexible manufacturing system consisting of four machines and a Load/Unload station is used for simulation purposes. Specifically, five jobs are hypothesized and some scenarios with an increasing number of available vehicles are simulated. In order to evaluate the benefits of the proposed approach, a time criterion based on the completion of all the jobs is considered with the aim to put in light that increasing the number of vehicles improves the control performance until congestion phenomena become unavoidable. Note to Practitioners—This paper proposes an innovative methodology for addressing the logistic operations within flexible manufacturing systems (FMSs) by using a team of autonomous ground vehicles. Unlike existing approaches, the core of this framework consists in combining along a hierarchical structure the capabilities of timed colored Petri nets and the deep reinforcement learning techniques to determine a near-optimal scheduling and run-time routing decisions that are provided to the distributed model predictive units in charge to accomplish the prescribed task. This multi-layer architecture has two main merits: a single platoon, completely disconnected from the others, is devoted to perform its own job; computational burdens are affordable during the on-line operations because most of the computations are moved in the off-line phase.

This scoping review paper redefines the Artificial Intelligence-based Internet of Things (AIoT) driven Human Activity Recognition (HAR) field by systematically extrapolating from various application domains to deduce potential techniques and algorithms. We distill a general model with adaptive learning and optimization mechanisms by conducting a detailed analysis of human activity types and utilizing contact or non-contact devices. It presents various system integration mathematical paradigms driven by multimodal data fusion, covering predictions of complex behaviors and redefining valuable methods, devices, and systems for HAR. Additionally, this paper establishes benchmarks for behavior recognition across different application requirements, from simple localized actions to group activities. It summarizes open research directions, including data diversity and volume, computational limitations, interoperability, real-time recognition, data security, and privacy concerns. Finally, we aim to serve as a comprehensive and foundational resource for researchers delving into the complex and burgeoning realm of AIoT-enhanced HAR, providing insights and guidance for future innovations and developments.

This paper delves into the resilience challenges encountered by platoons of self-governing agents navigating city streets, particularly focusing on the impact of misleading data infiltrating neighborly communication channels. The core issue centers around defining the trajectory paths of vehicles through routing decisions that must adhere to traffic fl ow constraints. To address this, we employ a decentralized model predictive control (MPC) strategy, leveraging deep reinforcement learning (DRL) abilities for effective decision-making in managing the regulation tasks. Additionally, the paper explores cybersecurity concerns by developing an effective anomaly detection system and practical attack countermeasures to enhance the resilience of the vehicle platoons against data tampering and other cyber threats. Through simulations and experiments, we demonstrate the efficacy of our approach in maintaining smooth traffic flow and ensuring secure communication within autonomous vehicle networks.

An important challenge in edge computing service management is maintaining good quality of service and low latency for end-users. Edge services must be hosted near the end user, necessitating sophisticated management of virtualized resources in the edge infrastructure. In the context of edge computing, users' location and mobility patterns are essential components of resource allocation and service transfer. As a result, it is of the utmost importance to assess the effectiveness of the proposed management solutions in a city environment with realistic mobility. The purpose of this study is to investigate the simulation of edge computing and to present an integrated solution environment that makes use of two validated simulators for the modeling of urban mobility and edge computing services.

Artificial Intelligence of Things (AIot) has made Stream Processing (SP) an essential element for data analytics. However, the dynamic nature of AIoT data streams incurs significant challenges in SP systems, especially when real-time data processing is required. While traditional approaches rely on resource scaling to meet the AIoT application requirements, our work explores a complementary approach by optimizing in real-time the configuration of SP systems to balance operational and resource efficiencies with data throughput and latency considerations. At the core of our work is the application of a Q-learning algorithm, enabling dynamic and responsive tuning of the SP system configuration parameters to adapt to the fluctuating demands of real-time data processing. Through empirical testing, our approach’s ability to improve SP efficiency in various data environments is demonstrated, providing insights into its practical application and effectiveness. The research concludes by highlighting the potential of extending this optimization approach to broader real-time data processing scenarios, suggesting avenues for future exploration in SP optimization.

The current surge in AI trends has catalyzed a strong inclination among organizations to transition towards AI-driven solutions. However, a significant challenge arises from the prevalent monolithic nature of existing applications, which often impedes scalability and limits the potential for enhancement through agent-based interventions. This paper aims to investigate strategies for transitioning from monolithic applications to microservices-based architectures and explore the utilization of agents for control within microservices environments. Subsequently, drawing from existing literature and our own insights, we endeavor to formulate a comprehensive strategy for transforming original monolithic applications into intelligently controlled microservices-based systems. We conclude with an IoT use case in order to illustrate the application of this strategy and highlight the advantages that can be achieved.

Human activity prediction has become increasingly prevalent in a plethora of time-critical applications. To realize accurate identification and prediction of human behaviour, we propose a situation-aware wearable computing system. A wearable computing system has the capability to perceive, comprehend and project situations by analyzing the human behavioral patterns in different environments. In particular, this work proposes a situation-aware human activity prediction (SA-HAP) approach based on sequential pattern mining that aims to anticipate future activities and tailor its responses according to situations by analyzing frequent sequential patterns and their correlations to understand how these situations are interrelated. The approach not only improves prediction accuracy but also provide the foundation for a more informed decision-making process, as the projected situations can be explained using the identified behavioral patterns. The approach is compared with other traditional techniques for activity prediction (LSTM and HMM), achieving better performance on the Extrasensory dataset.

Magnetic resonance imaging (MRI) with the high-resolution in computer-aided diagnostic technology is widely used to provide doctors with diagnostic advice, especially in brain tumor segmentation. In addition, MRI multi-sequence images of brain tumors also provide better image data support for studying brain tumor segmentation. In this paper, an adaptive threshold segmentation numerical optimization spiking neural P system (ATONSNPS or ATONSN P system) is designed to dynamically adjust the threshold quantity. In addition, the ATONSN P system and connectivity algorithm are combined to finish multi-sequence brain tumor segmentation. Experimental results on BraTS2019 show that the multi-sequence brain tumor segmentation approach can achieve more effective segmentation of brain tumor images comparing with several benchmark algorithms.

Federated learning (FL) is a distributed machine learning (ML) approach that enables models to be trained on client devices while ensuring the privacy of user data. Model aggregation, also known as model fusion, plays a vital role in FL. It involves combining locally generated models from client devices into a single global model while maintaining user data privacy. However, the accuracy and reliability of the resulting global model depend on the aggregation method chosen, making the selection of an appropriate method crucial. Initially, the simple averaging of model weights was the most commonly used method. However, due to its limitations in handling low-quality or malicious models, alternative techniques have been explored. As FL gains popularity in various domains, it is crucial to have a comprehensive understanding of the available model aggregation techniques and their respective strengths and limitations. However, there is currently a significant gap in the literature when it comes to systematic and comprehensive reviews of these techniques. To address this gap, this paper presents a systematic literature review encompassing 201 studies on model aggregation in FL. The focus is on summarizing the proposed techniques and the ones currently applied for model fusion. This survey serves as a valuable resource for researchers to enhance and develop new aggregation techniques, as well as for practitioners to select the most appropriate method for their FL applications.

The healthcare industry faces challenges due to rising treatment costs, an aging population, and limited medical resources. Remote monitoring technology offers a promising solution to these issues. This article introduces an innovative adaptive method that deploys an ultrawideband (UWB) radar-based Internet of Medical Things (IoMT) system to remotely monitor elderly individuals’ vital signs and fall events during their daily routines. The system employs edge computing for prioritizing critical tasks and a combined cloud infrastructure for further processing and storage. This approach enables monitoring and telehealth services for elderly individuals. A case study demonstrates the system’s effectiveness in accurately recognizing high-risk conditions and abnormal activities, such as sleep apnea and falls. The experimental results show that the proposed system achieved high accuracy levels, with a mean absolute error (MAE) ± standard deviation of absolute error (SDAE) of 1.23± 1.16 bpm for heart rate (HR) detection and 0.22 ± 0.27 bpm for respiratory rate (RR) detection. Moreover, the system demonstrated a recognition accuracy of 90.60% for three types of falls (i.e., stand, bow, squat to fall), one daily activity, and No Activity Background. These findings indicate that the radar sensor provides a high degree of accuracy suitable for various remote monitoring applications, thus enhancing the safety and well-being of elderly individuals in their homes.

The emergence of the Internet of Things (IoT) has revolutionized various domains by enabling seamless connectivity and real-time data exchange between connected IoT devices. However, in sparse deployment scenarios where sensor nodes are sparsely distributed, ensuring low data delivery latency becomes a significant challenge. Our research aims to address this issue by utilizing unmanned aerial vehicles (UAVs) to support IoT networks. In the existing UAV-aided IoT systems, all UAVs are required to return to the base station to deliver data, which results in significant data delivery latency. To overcome this limitation, we propose a collaborative data acquisition model that uses air-to-air data relay between UAVs. By leveraging the mobility and agility of UAVs, the proposed system facilitates efficient data relay between sensor nodes and the base station. To further optimize the performance of the system, we present a time-balancing scheduling data acquisition (TSDA) scheme. This scheme combines a centripetal-based relay pairing method for UAVs to achieve seamless data relay and a joint scheduling scheme to minimize the hovering time during data delivery. Through extensive simulations, we demonstrate that the proposed TSDA scheme can achieve lower data delivery latency in sparse deployment scenarios compared to existing data acquisition schemes. In addition, the joint scheduling scheme can significantly reduce the hovering time of UAVs so that the collaborative relaying advantage can be better exploited.

Deep learning and big data analysis are among the most important research topics in the fields of biomedical applications and digital healthcare. With the fast development of artificial intelligence (AI) and Internets of Things (IoT) technologies, deep learning (DL) for big data analytics—including affective learning, reinforcement learning, and transfer learning—are widely applied to sense, learn, and interact with human health. Examples of biomedical applications include smart biomaterials, biomedical imaging, heartbeat/blood pressure measurement, and eye tracking. These biomedical applications collect healthcare data through remote sensors and transfer the data to a centralized system for analysis. With an enormous amount of historical data, DL and big data analysis technologies are able to identify potential linkage between features and possible risks, raise important decision for medical diagnosis, and provide precious advice for better healthcare treatment and lifestyle. Although significant progress has been made with AI, DL, and big data analytic technologies for medical and healthcare research, there remain gaps between the computer-aided treatment design and real-world healthcare demands. In addition, there are unexplored areas in the fields of healthcare and biomedical applications with cutting-edge AI and DL technologies. Hence, exploring the possibility of DL and big data analytics in the fields of biomedical applications and digital healthcare is in high demand.

Numerous sensors on smart devices have made it possible to automatically recognize human movement, which might be helpful for intelligent applications like elder care, smart homes, and health monitoring. Nevertheless, implementing an activity recognition model in practical situations faces two main obstacles. First, machine learning models use a large number of labeled data to recognize human activities, which is not always feasible in real scenarios. Second, existing human activity recognition (HAR) systems cannot dynamically adapt to a new action. Furthermore, current methods fail to separate short-term activities from heterogeneous smart devices with varying positions and orientations that have similar sensory reading patterns. To address these issues, we propose Flexi-HAMR, an intelligent adaptive human activity monitoring and recognition system that dynamically recognizes activities using online, real-time activity signals. Many empirical findings show that the suggested flexible activity recognition model performs competitively on multiindividual activity identification tasks and has a comparatively more vital generalization ability.

In this article, the problem of generating multimodel state space descriptions in a data-driven context to embed the dynamic behavior of nonlinear systems is addressed. The proposed methodology takes advantage of three ingredients: 1) linear time-invariant system behavior; 2) data-driven modeling; and 3) reinforcement learning (RL) technicalities. These elements are properly combined to develop a data-driven algorithm capable to derive an accurate outer convex approximation of the nonlinear evolution. In particular, an actor-critic RL scheme is designed to efficiently comply with the exhaustive research on the whole parameter space. At each iteration, the effectiveness of the obtained uncertain polytopic model is tested by a probabilistic approach based on a confidence level metrics. As the main merits of the proposed approach are concerned, the following aspect clearly stands up: the development of an interdisciplinary methodology that takes advantage of system theory, probabilistic arguments and RL capabilities giving rise to an harmonized architecture in charge to deal with a vast class of nonlinear systems. Finally, the validity of the proposed approach is tested by resorting to benchmark examples that allow to quantify the level of accuracy of the computed convex hull.

This paper presents a set-theoretic receding horizon control strategy for platoons of autonomous vehicles driving in smart cities context. In order to reduce traffic and$CO_{2}$emissions, we propose a path planer based on Deep Reinforcement Learning (DRL). The advantages of this solution is the ability to deal with the actual traffic congestion, while driving the autonomous vehicles to their destination and fulfilling the constraints. In particular, the high-level routing decisions are translated into set-points for the receding horizon controllers, making the control actions on the vehicle dynamics more computational efficient. In order to show the effectiveness of the overall architecture, a campaign of simulations on a platoon of eight vehicles, moving in the city center of Bologna in Italy, is provided.

The Metaverse is an emerging concept that aims to integrate the physical and virtual worlds, creating a shared 3D virtual world where users can interact and immerse in new experiences. With the rise of Metaverse-driven Cyber-Physical-Social Systems (CPSSs), integrating drones as a critical technology in the Metaverse has become increasingly important. CPSSs have become proliferating and integral to our daily lives. This paper proposes a conceptual framework for Metaverse-driven drone edge intelligence, which integrates drone-enabled sensing, communication, and computation to enable real-time decision-making in CPSSs. We present a detailed analysis of the challenges and opportunities for integrating drones in the Metaverse and discuss the potential impact of our framework on various application domains. Our work contributes to advancing the Metaverse and CPSSs by providing a novel approach for empowering real-time decision-making and enabling new user experiences through integrating drones and the Metaverse. The proposed framework has the potential to revolutionize the way we approach data-driven decision-making in various industries and applications, including precision agriculture, transportation, emergency response, smart cities, healthcare, manufacturing, and energy.

Cities have undergone significant changes due to the rapid increase in urban population, heightened demand for resources, and growing concerns over climate change. To address these challenges, digital transformation has become a necessity. Recent advancements in Artificial Intelligence (AI) and sensing techniques, such as synthetic sensing, can elevate Digital Twins (DTs) from digital copies of physical objects to effective and efficient platforms for data collection and in-situ processing. In such a scenario, this paper presents a compre-hensive approach for developing a Traffic Monitoring System (TMS) based on Edge Intelligence (EI), specifically designed for smart cities. Our approach prioritizes the placement of intelligence as close as possible to data sources, and leverages an “opportunistic” interpretation of DT (ODT), resulting in a novel and interdisciplinary strategy to re-engineering large-scale distributed smart systems. The preliminary results of the proposed system have shown that moving computation to the edge of the network provides several benefits, including (i) enhanced inference performance, (ii) reduced bandwidth and power consumption, (iii) and decreased latencies with respect to the classic cloud -centric approach.

Postural monitoring in wheelchair users is a topic of growing interest. The detection of changes in the sitting patterns of these patients may serve to detect changes in their functional status and be able to adapt rehabilitation early. For this reason, this paper presents a methodology for the detection of specific postural anomalies that, unlike previous works, adopts unsupervised learning. The proposed methodology involves data dimensionality reduction using Principal Component Analysis, and the application of K-means clustering to group different normal posture states. The anomalies are detected using a threshold approach, where data points that fall outside a certain threshold are considered as anomalies. The results show that the methodology is effective in identifying anomalies with a high degree of accuracy (around 90%).

The blockchain (BC) paradigm is currently modifying all the major Internet of Things (IoT) application domains, giving the opportunity of constructing decentralized environments in which trustful and anonymous activities can be efficiently performed. BC proposes an approach which assures data saved on a distributed ledger (DL) would be continuously synchronized and in such a way the ledger will remain consistent. The DL has the responsibility to maintain a higher level of consistency. Given a connected network, the optimum neighbor selection (ONS) of paths can be obtained by finding the minimum spanning tree of the network. However, none of past approaches proposed in the literature to construct ONS considered the problem of having nodes with low levels of reliability in the network and even malicious or fraudulent nodes, that is a situation very common in an IoT environment. In this article, we propose an optimized BC ONS algorithm, called trust-based ONS (TONS), which allows the miners to communicate with a globally optimized selection of neighbors. The algorithm can also guarantee that these nodes are the most reliable miners. We also describe an experimental simulation campaign we have performed to evaluate the effectiveness and the efficiency of our approach.

The expeditious propagation of Internet of Things (IoT) technologies implanted in different smart devices such as smartphones and smartwatches have a ubiquitous consequence on the modern population. These devices are employed to collect data and to aid in tracking and analyzing the users’ daily activities using various human activity monitoring and recognition (HAR) techniques. However, most current HAMR approaches rely on exploratory case-based shallow feature learning architectures, which endeavor to recognize activities correctly in real-world situations. To address this issue, we offer a unique strategy for HAMR that leverages the attention mechanism with multi-head convolutional neural networks (CNNs) and Long-Short-Term-Memory (LSTM). The accuracy of activity detection is improved in the presented method by integrating attention into multi-head CNNs followed by LSTM for better feature extraction and selection. Verification investigations are carried out using data from the University of California (UCI) repository, which is publicly available. The results show that our proposed framework is more accurate than current frameworks using both the 10-fold and leave-one-subject-out cross-validation. Finally, the proposed method can recognize human activity in real-time, regardless of the type of smart device.

The introduction of the IoT technology and its pervasive penetration in our daily life implies that IoT smart objects can often participate in social events, and in this case their paradigm of interaction is commonly denoted as social IoT (SIoT). To make reliable transactions into a SIoT scenario, in this article, we introduce a multi-agent SIoT architecture, which integrates a reputation system based on a clustering of the SOs. In our framework, when a SO looks for a resource and detects a reliable partner having that resource, then the two SOs can interact to make a transaction, and at the end of the transaction each of them provides a feedback about the partner to the local reputation system of the edge server, to update the SOs’ reputation scores. Moreover, each edge server sends these reputation scores to the cloud, which updates the associated reputation values of the SOs, deriving from experiences coming from all the edge domains. This architecture provides a given object, moving from an edge domain to another one, with the possibility to have an updated value of its reputation, represented by the value stored in the cloud. We have validated our approach by a campaign of simulations, whose results seem particularly promising.

More data are generated through interaction between cyber space, physical space, and social space thanks to mobile network technology, giving birth to the so-called cyber–physical social intelligent ecosystem (C&P-SIE). This survey studies the development of physical social intelligence. First, it classifies and discusses the behavior modeling, learning, and adaptation applications of C&P-SIE from intelligent transportation, healthcare, public service, economy, and social networking. Then, it prospects the application of behavior modeling in the C&P-SIE from the perspectives of information security, data-driven techniques, and modeling learning under cooperative artificial intelligence technologies. The research provides a theoretical basis and new opportunities for the digital and intelligent development of smart cities and social systems.

The integration of artificial intelligence (AI) with cyber–physical–social systems (CPSS) creates new research opportunities and challenges with major societal implications. The behavioral and cognitive enhancement of intelligent systems promotes a productive and creative partnership and collaboration between humans and machines. Advancements in these areas enable adaptability, scalability, resiliency, safety, security, and usability that expand the horizons of CPSS.

Artificial intelligence-driven automation has gradually become the technical trend of the new automation era. At present, many artificial intelligence technologies have been applied to improve the intelligence level in the field of automation. Among them, convolutional neural network (CNN) technology is one of the most representative, which is used in the detection of defective products in industrial automation, robot human tracking has been widely used in the field of machine vision driven automation. However, the high dependence of the current neural network application leads to the potential failure of the defective product detection system. In this article, we model the learning and decision-making process of CNN with a statistical physical percolation model. Based on the differentiation degree and vulnerability of percolation, we propose the concept of CNN differentiation degree and summarize the empirical formula to quantify it. The relationship between the differentiation degree and vulnerability is analyzed from both adversarial attack and adversarial training perspectives to explain the decision-making mechanism of CNN and classification reliability. The physical model can approach the essence of things and finally guide the reliable CNN for industrial automation.

In this paper, a sustainable routing algorithm for vehicle platoons operating in smart urban networks is presented. The proposed approach makes use of deep reinforcement learning (DRL) and set-theoretic model predictive control (MPC). In particular, the learning process aims at reducing traffic congestion and$CO_{2}$emissions, whereas the MPC unit allows to adequately track the assigned path by using real-time traffic data. To adequately analyze the performance of the resulting control architecture, the SUMO and MATLAB environments are used to implement complex operating scenarios where road maps data and vehicle state trajectories can be shared and exchanged. Finally, numerical studies are provided by resorting to the SUMO environment and considering a platoon of five vehicles. The resulting simulation campaign puts in light the capability of the training process to significantly mitigate the$CO_{2}$emissions of the whole platoon: from a minimum of 3.7 % to a maximum of 13% with respect to the use of the well-known Dijkstra algorithm.

Cognitive networking is a valuable enabler to improve the capability of intelligent transportation system (ITS) by analyzing and utilizing the heterogeneous traffic information. However, the significant increase in the amount of decision-making tasks makes it difficult to guarantee real-time performance of decision response. This paper focuses on the problem of the quality and real-time assurance of collaborative decision-making response in large-scale ITS during multi-task parallelism execution. First, a collaborative decision architecture with cognitive networking is developed, which introduces the advanced 6G communication technology to enhance information interaction capability of vehicle-road-cloud collaboration, and lays the foundation for multi-task real-time decision-making with inevitable fuzzy information in the perception process. Then, a multi-task parallel multi-granularity collaborative decision model (MPMCD) is designed to improve knowledge discovery ability for decision-making process by building multi-granularity information structures. An AI-driven cognitive networking collaborative decision-making (ACNCD) algorithm is further proposed based on MPMCD model to support multi-task parallel vehicle-road-cloud collaborative real-time decision. Extensive simulation experiments are carried out to evaluate ACNCD algorithm in terms of several performance criteria including decision response time, accuracy, and accident rate. The obtained results show that the comprehensive decision-making performance of ACNCD outperforms other relevant existing algorithms.

Edge AI technologies have been used for many Intelligent Transportation Systems, such as road traffic monitor systems. Neural Architecture Search (NAS) is a typcial way to search high-performance models for edge devices with limited computing resources. However, NAS is also vulnerable to adversarial attacks. In this paper, A One-Shot NAS is employed to realize derivative models with different scales. In order to study the relation between adversarial robustness and model scales, a graph-based method is designed to select best sub models generated from One-Shot NAS. Besides, an evaluation method is proposed to assess robustness of deep learning models under various scales of models. Experimental results shows an interesting phenomenon about the correlations between network sizes and model robustness, reducing model parameters will increase model robustness under maximum adversarial attacks, while, increasing model paremters will increase model robustness under minimum adversarial attacks. The phenomenon is analyzed, that is able to help understand the adversarial robustness of models with different scales for edge AI transportation systems.

In practical cyber-manufacturing systems (CMS), the node component is the forwarder of information and the provider of services. This dual role makes the whole system have the typical physical-services interaction characteristic, making CMS more vulnerable to cascading failures than general manufacturing systems. In this work, in order to reasonably characterize the cascading process of CMS, we first develop an interdependent network model for CMS from a physical-service networking perspective. On this basis, a realistic cascading failure model for CMS is designed with full consideration of the routing-oriented load distribution characteristics of the physical network and selective load distribution characteristics of the service network. Through extensive experiments, the soundness of the proposed model has been verified and some meaningful findings have been obtained: (1) attacks on the physical network are more likely to trigger cascading failures and may cause more damage; (2) interdependency failures are the main cause of performance degradation in the service network during cascading failures; and (3) isolation failures are the main cause of performance degradation in the physical network during cascading failures. The obtained results can certainly help users to design a more reliable CMS against cascading failures.

This paper focuses on the use of machine learning (ML) algorithms for greenhouses temperature forecasting. We use a new approach called TinyML, an emerging technique with high potential for a low-power environments. Our focus in this work was to understand which are the limits of this approach and how much complexity can the ML models have with the used devices. Our results reveal that a TinyML-based device such as an Arduino Nano 33 BLE Sense can execute easily low computational constrained ML models while consuming very low power compared to a high-end device. In particular, a simple multilayer perceptron (MLP) is targeted, offering accurate predictions in terms of RMSE, MAE and R2, which is remarkable considering that the models under study are extremely light and simple.

In this paper, the path planning problem under routing decisions is considered for platoons of autonomous vehicles moving in urban road networks. The key idea is to exploit and adequately combine arguments coming from two research fields: deep reinforcement learning and model predictive control. Along these lines, a novel control architecture is proposed and its feasibility formally proved. In particular, the high-level routing decisions arising from the distributed deep reinforcement learning operations are translated into manipulable set-points for the underlying bank of receding horizon controllers by making more computational affordable and efficient the action on the vehicle dynamics. Finally, some simulations on platoon, consisting of three agents described by double-integrator models, are provided to show the effectiveness of the overall architecture.

Intelligent driving technology plays a key role in reducing road traffic accidents and ensuring driving, where the vehicle behavior decision making capability largely determines the driving performance of intelligent vehicles. At this point, more research is focused on enhancing vehicle environment identification and vehicle control capabilities, with decision-making systems receiving less attention. In order to improve the accuracy of intelligent vehicle behavior decision making and ensure the active safety of path planning, this paper firstly establishes an edge intelligence based software-defined Internet of Vehicles (ESIOV) architecture. Then, a POMDP-based intelligent vehicle behavior decision model is designed using the time-series iterative property of partially observable Markov decision process (POMDP). Finally, a reinforcement learning vehicle behaviour decision (ERVBD) algorithm based on edge intelligence is proposed to ensure the accuracy of intelligent vehicle behavior decision results and improve the decision speed. The simulation results show that ERVBD can successfully implement vehicle behavioural decision making, enabling intelligent vehicles to anticipate collision risks and adopt reasonable avoidance strategies in real time.

Research trends are pushing artificial intelligence (AI) across the Internet of Things (IoT)–edge–fog–cloud continuum to enable effective data analytics, decision making, as well as the efficient use of resources for QoS targets. Approaches for collective adaptive systems (CASs) engineering, such as aggregate computing, provide declarative programming models and tools for dealing with the uncertainty and the complexity that may arise from scale, heterogeneity, and dynamicity. Crucially, aggregate computing architecture allows for “pulverization”: applications can be decomposed into many deployable micromodules that can be spread across the ICT infrastructure, thus allowing multiple potential deployment configurations for the same application logic. This article studies the deployment architecture of aggregate-based edge services and its implications in terms of performance and cost. The goal is to provide methodological guidelines and a model-based toolchain for the generation and simulation-based evaluation of potential deployments. First, we address this subject methodologically by proposing an approach based on deployment code generators and a simulation phase whose obtained solutions are assessed with respect to their performance and costs. We then tailor this approach to aggregate computing applications deployed onto an IoT–edge–fog–cloud infrastructure, and we develop a corresponding toolchain based on Protelis and EdgeCloudSim. Finally, we evaluate the approach and tools through a case study of edge multimedia streaming, where the edge ecosystem exhibits intelligence by self-organizing into clusters to promote load balancing in large-scale dynamic settings.

Nonorthogonal multiple access (NOMA) with a grant-free access has received a lot of attention due to its support to massive machine-type communication (mMTC) devices. The devices in grant-free systems are allowed to transmit information without undergoing an authentication process. Therefore, in such systems, the base station needs to distinguish between active and nonactive devices, called the active user detection (AUD) process. This process is challenging as the active device needs to be detected from the received signals that are superimposed. Furthermore, the identification of the Internet of Things (IoT) devices from these signals also poses a great challenge, which could help allocate resources in future generation communication systems. Motivated from the aforementioned facts, this article proposes a device detection and identification (DDI) architecture for joint AUD and IoT device identification from the received superimposed signals. The architecture extracts the Fourier patterns as the representative feature vector, which results in an improved detection and identification process. Experimental results show that the architecture not only outperforms the conventional schemes and deep neural network-based approaches in terms of success probability for the AUD task but also yields lower computational complexity. The evaluation of the DDI architecture for IoT device identification problems has also been performed and compared to various shallow learning methods to prove its efficacy.

Motion capture technology has been widely used in the sport analysis to improve their performance and reduce the injury risk. Kayak, a popular outdoor sport, employs the coordination of multiple muscles and skeletons, especially those of upper limbs that must be investigated carefully. The fine-time phase segmentation of rowing cycle plays an important role in analyzing kayaker’s technique. Aiming at the problem of laborious manual phase labeling in the traditional video analysis method, an automatic phase segmentation method for kayak rowing is proposed combined with a machine learning algorithm. In this article, inertial sensors and a data fusion algorithm are used to calculate the joint angles between arm and trunk, left elbow and right elbow when the athlete is rowing. According to the permutation and combination principle, the angle sequence is combined in nine different ways, and four machine learning algorithms (decision tree, support vector machine,$k$-nearest neighbor, bagging ensemble learning) are used to study the effects of different combinations on rowing phase division. Among them, the precision of phase segmentation becomes higher with the increase of motion information. The combination of arm to trunk joint angle only needs three data collection nodes; thus, the computational cost is smaller; moreover, all the four algorithms show good classification accuracy (up to 98.1%). The results indicating that the combination of arm to trunk joint angle and support vector machine algorithm could better complete the task of the phase segmentation for kayak rowing.

Critical infrastructures across many industries, such as smart water treatment and distribution networks (SWTDNs) and power generation and public transport networks, depend on the supervisory control and data acquisition (SCADA) system. However, being the core component of the critical infrastructures, it has made the SCADA-based SWTDN system an attractive target for cyberattacks. A successful attack on the SCADA will have a devastating impact on an SWTDN in terms of proper operations; therefore, safeguarding the SCADA from cyberattacks is of paramount. With the increasing cyberattacks on SWTDN, both in number and sophistication, the need to detect these attacks early has become a subject of great interest among practitioners and researchers. To this end, we propose a novel strategy, based on a semisupervised approach. Two semisupervised approaches, including unsupervised learning and deep learning-based approaches, have been proposed. The proposed approaches can involve learning dynamic cyberattack patterns from unlabeled data in an SWTDN. We validate the proposed semisupervised approach experimentally using an operational water treatment plant testbed. The proposed approach achieved almost 100% accuracy and substantially outperforms the existing baseline approaches used in this article. The outcome of the experiment is encouraging and demonstrates the potential use of the semisupervised approach for security control in smart water distribution.

Cryptocurrencies are the new form of trade that has revolutionized how we look into our financial institutions. Bitcoin dominates the industry with the highest market share among the hundreds of other cryptocurrencies. However, high energy consumption leading to increasing carbon emission, prioritizing high-value transactions, and long waiting times are some of the flaws preventing it from reaching its full potential. Owing to the block rewards getting halved every four years, miners and researchers are fearful that this would be the breaking point of Bitcoin’s success. This article proposes an Industry-4.0-compliant next-generation Bitcoin architecture by introducing a dynamic and sustainable block concept. Along with our modified knapsack algorithms, i.e., priority-based 0/1 knapsack and advanced-priority-based 0/1 knapsack, we can ensure a balanced transaction selection, quicker verification, higher transaction throughput, reduced carbon emission, and increased earnings for the miners. Moreover, with the addition of only one of our proposed sustainable blocks, we can cut down verification times by 50% and increase throughput by 39%. We can also reduce carbon emissions per transaction by 61.3%, which would help reduce Bitcoins’ large carbon footprint, enabling us to approach greener digital transactions.

With the rise of the Industrial Internet of Things (IIoT), more and more industrial devices can be connected via the network. Data collection, processing, analysis, task execution, and other devices that can product network traffic volume are gradually being deployed to IIoT. However, under the limited spectrum resources and low-cost and low-energy production requirements of enterprises, how to ensure the interconnection and intercommunication of industrial networks while realizing the effective use of network communication resources is currently a hot topic. Among them, network traffic prediction is considered to be a very important task. The time variability and interpretability, especially the time-varying features of traffic sequences, greatly challenge this task. To address those, this article proposes a method calledFlow2graphto predict network traffic in IIoT. Specifically, some key segments, i.e., shapelets are extracted from the network traffic sequence according to time-varying traffic; then uses the relationship between the traffic sequence and shapelets to convert the flow into a shapelets conversion graph; Subsequently, the graph isomorphism network are used to learn the specificity of the flow sequence from different devices, thereby to predict its traffic value for a period of time in the future; finally, we conduct extensive experiments on real data to verify the effectiveness of the proposed method.

With the widespread application of artificial intelligence and the Internet of Things, the intellectualization of the industrial Internet of Things (IIoT) has received more and more attention. However, in the application scenario with numerous sensors, the contradiction between massive requests of computing tasks and high requirements of inference quality affects the operation efficiency and service reliability. Moreover, due to the heterogeneity of computing resources and the randomness of communication environments of the cloud-edge system, how to compute and deploy deep learning models in a cloud-edge collaborative environment has also become a challenging problem. Therefore, this article presents a collaborative cloud-edge service cognitive framework for deep neural network (DNN) model service configuration to provide dynamic and flexible computing services. In order to adapt to different service requirements, we explored the tradeoffs between accuracy, latency, and energy consumption indicators, and a revenue target is established, which considers the quality of service experience and the system energy consumption to improve resource utilization efficiency. By transforming the optimization of the revenue target into a partially observable DNN configuration reinforcement learning problem, a dueling deep Q-learning network-based self-adaptive DNN configuration algorithm is proposed. Experimental results show that the proposed mechanism can effectively learn from external experience, adapt to the dynamic network environment, and reduce delay and energy consumption while meeting the service requirements.

The development of activity recognition based on multi-modal data makes it possible to reduce human intervention in the process of monitoring. This paper proposes an efficient and cost-effective multi-modal sensing framework for activity monitoring, it can automatically identify human activities based on multi-modal data, and provide help to patients with moderate disabilities. The multi-modal sensing framework for activity monitoring relies on parallel processing of videos and inertial data. A new supervised adaptive multi-modal fusion method (AMFM) is used to process multi-modal human activity data. Spatio-temporal graph convolution network with adaptive loss function (ALSTGCN) is proposed to extract skeleton sequence features, and long short-term memory fully convolutional network (LSTM-FCN) module with adaptive loss function is adapted to extract inertial data features. An adaptive learning method is proposed at the decision level to learn the contribution of the two modalities to the classification results. The effectiveness of the algorithm is demonstrated on two public multi-modal datasets (UTD-MHAD and C-MHAD) and a new multi-modal dataset H-MHAD collected from our laboratory. The results show that the performance of the AMFM approach on three datasets is better than the performance of the video or the inertial-based single-modality model. The class-balanced cross-entropy loss function further improves the model performance based on the H-MHAD dataset. The accuracy of action recognition is 91.18%, and the recall rate of falling activity is 100%. The results illustrate that using multiple heterogeneous sensors to realize automatic process monitoring is a feasible alternative to the manual response.

Recent years have witnessed an increasing popularity of wireless body area network (WBAN), with which continuous collection of physiological signals can be conveniently performed for healthcare monitoring. Energy consumption is a critical issue because it directly affects the duration of the equipped sensors. In this article, we propose a low-cost and confidential electrocardiogram (ECG) acquisition approach for WBAN. The compressed sensing (CS) is employed for low-cost signal acquisition, and its cryptographic features are exploited for promoting the framework's confidentiality. In particular, the RIPless measurement matrix is used to give CS the resistance against plaintext attack, while the first-order Σ∆ quantizer is employed to embed the cryptographic diffusion feature into the whole system. Two chaotic systems are employed for generating the required secret elements for the acquisition and encryption. Experiment results well demonstrate the signal reconstruction and security performance of the proposed framework.

In the Cognitive Internet of Vehicles (CIoV), vehicles, road side units (RSU) and other key nodes have been equipped with more and more software to support intelligent transportation system (ITS), vehicle automatic control and intelligent road information services. Additionally, technological innovation forces the software in the CIoV to update and upgrade in time. However, escalation is critical to the safety, stability, and maintenance cost of transportation systems. It can be assumed that when the intelligent services supporting CIoV can realize self-perception and escalation, the cognitive ability and coordination ability of the entire CIoV will be greatly improved. To address this, we first propose a deep learning-based method for Software Escalation Prediction (SEP) in CIoV. Specifically, the pretraining mechanism of transformers in the field of natural language processing is combined with software upgrade-related events to dynamically model software sequence activities. To capture the event association in the software activities, we use graph modeling software’s state log and utilize a graph neural network (GNN) to learn the complex life activity rule of software. Finally, the above characteristics are deeply integrated. The proposed method has a 6%–8% improvement over the RoBERTa methods.

Limited by the computational capabilities and battery energy of terminal devices and network bandwidth, emotion recognition tasks fail to achieve good interactive experience for users. The intolerable latency for users also seriously restricts the popularization of emotion recognition applications in the edge environments such as fatigue detection in auto-driving. The development of edge computing provides a more sustainable solution for this problem. Based on edge computing, this article proposes a multi-modal multi-layer emotion-aware service (MULTI-EASE) architecture that considers user’s facial expression and voice as a multi-modal data source of emotion recognition, and employs the intelligent terminal, edge server and cloud as multi-layer execution environment. By analyzing the average delay of each task and the average energy consumption at the mobile device, we formulate a delay-constrained energy minimization problem and perform a task scheduling policy between multiple layers to reduce the end-to-end delay and energy consumption by using an edge-based approach, further to improve the users’ emotion interactive experience and achieve energy saving in edge computing. Finally, a prototype system is also implemented to validate the architecture of MULTI-EASE, the experimental results show that MULTI-EASE is a sustainable and efficient platform for emotion analysis applications, and also provide a valuable reference for dynamic task scheduling under MULTI-EASE architecture.

The recent widespread pandemic of COVID-19 has put tremendous pressure on the healthcare system. The deployment of telehealth technology is crucial in solving this problem when patients are mildly ill and need to self-isolate at home or in a specific location. This paper proposes using a single radar sensor to continuously contact-less monitor the patients' vital signals in their daily lives. We use edge computing to handle high-priory tasks and combined cloud infrastructure for further process and storage to provide monitoring and telehealth services. A case study is presented to show how the approach can continuously monitor and recognize high-risk diseases and abnormal activity (e.g., sleep apnea). While an accident occurs, the system could provide fast and accurate emergency services. The work has been compared with a good standard. And the experimental results show that the proposed approach for heart rate (HR) and respiratory rate (RR) detection achieved a Mean Absolute Error (MAE) ± Standard Deviation of Absolute Error (SDAE) of 0.09±1.43 bpm and 0.23±3.23 bpm, respectively. This indicates the radar sensor can provide a high recognition accuracy to meet the requirements for a range of cardiopulmonary function monitoring. This kind of telemedicine service facilitates monitoring the self-isolated subjects to detect and recognize human physical and physiological activities.

Summary form only given, as follows. The complete presentation was not made available for publication as part of the conference proceedings. Gartner estimates the global smart wearable computing systems (WCS) market will be worth more than US$93 billion in 2022, with an increasing growth caused by the COVID-19. The industry and public sector are then pushing for innovative WCS solutions with high levels of dependability and trustworthiness that can efficiently operate in increasingly complex scenarios. Great strives have been made to realize WCS for the 24/7 monitoring of single users based on 3- tier architectures involving wearables, edge, and cloud systems. However, new requirements, specifically targeting cooperative multiple users, demands for radically new approaches, as promoted by the community-oriented WCS (CO-WCS). In this keynote, we first provide an overview of WCS based on the SPINE Body of Knowledge research and development (https://projects.dimes.unical.it/spine-bok/). Then, we focus on the requirements of the nextgeneration CO-WCS based on a use case driven approach. Finally, we will discuss models, architectures and tools that would be needed to implement CO-WCS.

The Internet of Things (IoT) enables the interconnection of new cyber–physical devices that generate significant traffic of distributed, heterogeneous, and dynamic data at the network edge. Since several IoT applications demand for short response times (e.g., industrial applications, emergency management, real-time systems, and healthcare systems) and, at the same time, rely on resource-constrained devices, the adoption of traditional data mining techniques is neither effective nor efficient. Therefore, conventional data mining techniques need to be adjusted for optimizing response times, energy consumption, and data traffic while still providing adequate accuracy as required by the IoT applications.

In this article, we explore the remote monitoring of patients in an on-demand service that can be implemented using the Internet of Medical Things (IoMT). Network congestion and delay are important issues that must be resolved to handle emergencies in IoMT. Backpressure scheduling is a well-known scheme to resolve network congestion and improve network throughput for emergency data packets. Traditional backpressure scheduling is known to categorize each packet and schedules emergency packets to be forwarded faster than regular packets. However, in large-scale networks, these scheduling algorithms find unnecessarily long paths. This leads to high end-to-end delay and decreases the performance of guaranteed delivery. To resolve these issues, this article proposes an event-aware priority scheduling algorithm for data packets. This model follows a single priority queue model to manage all packets and emergency packets have been identified by the threshold values. Also, a separate communication path has been identified as well to reduce the waiting time for each category of the packet. Meanwhile, to reduce delay in packet communication, a delay-efficient data aggregation tree is constructed, which is combined with the priority queue model. Our in-depth simulation results of the proposed model prove the novel contribution to reducing delay in emergency packet delivery and avoiding network congestion. Moreover, the proposed model is also compared with the existing state-of-the-art models to show its ability to outperform similar methodologies.

In the foreseeable future, the rapid growth of devices in the Internet of Things (IoT) will make it difficult for 5G networks to ensure sufficient network resources. 6G technology has attracted increasing attention, bringing new design concepts to the dynamic real-time resource allocation. The resource requirements of devices are usually variable, so a dynamic resource allocation method is needed to ensure the smooth execution of tasks. Therefore, this article first designs a 6G-enabled massive IoT architecture that supports dynamic resource allocation. Then, a dynamic nested neural network is constructed, which adjusts the nested learning model structure online to meet training requirements of dynamic resource allocation. An AI-driven collaborative dynamic resource allocation (ACDRA) algorithm is proposed based on the nested neural network combined with Markov decision process training for 6G-enabled massive IoT. Extensive simulations have been carried out to evaluate ACDRA in terms of several performance criteria, including resource hit rate and decision delay time. The results validated that ACDRA improves the average resource hit rate by about 8% and reduces the average decision delay time by about 7% compared with three reference existing algorithms.

The adoption of IoT for smart health applications is a relevant tool for distributed and intelligent automatic diagnostic systems. This work proposes the development of an integrated solution to monitor maternal and fetal signals for high-risk pregnancies based on IoT sensors, feature extraction based on data analytics, and an intelligent diagnostic aid system based on a 1-D convolutional neural network (CNN) classifier. The fetal heart rate and a group of maternal clinical indicators, such as the uterine tonus activity, blood pressure, heart rate, temperature, and oxygen saturation are monitored. Multiple data sources generate a significant amount of data in different formats and rates. An emergency diagnostic subsystem is proposed based on a fog computing layer and the best accuracy was 92.59% for both maternal and fetal emergency. A smart health analytics system is proposed for multiple feature extraction and the calculation of linear and nonlinear measures. Finally, a classification technique is proposed as a prediction system for maternal, fetal, and simultaneous health status classification, considering six possible outputs. Different classifiers are evaluated and a proposed CNN presented the best results, with the F1-score ranging from 0.74 to 0.91. The results are validated based on the diagnosis provided by two specialists. The results show that the proposed system is a viable solution for maternal and fetal ambulatory monitoring based on IoT.

The ever-growing aging of the population has emphasized the importance of in-home AAL (Ambient Assisted Living) services for monitoring and improving its well-being and health, especially in the context of care facilities (retirement villages, clinics, senior neighborhood, etc). The paper proposes a novel simulation-driven platform named E-ALPHA (Edge-based Assisted Living Platform for Home cAre) which supports both Edge and Cloud Computing paradigm to develop innovative AAL services in scenarios of different scales. E-ALPHA flexibly combines Edge, Cloud or Edge/Cloud deployments, supports different communication protocols, and fosters the interoperability with other IoT platforms. Moreover, the simulation-based design helps in preliminary assessing (i) the expected performance of the service to be deployed according to the infrastructural characteristics of each specific small, medium and large scenario; and (ii) the most appropriate applications/platform configuration for a real deployment (kind and number of involved devices, Edge- or Cloud-based deployment, required connectivity type, etc). In this direction, two different use cases modeled according to realistic input (coming from past experience involving real testbed) are shown in order to demonstrate the potentials of the proposed simulation-driven AAL platform.

The rapid development of Internet of Things (IoT) platforms provides the industrial domain with many critical solutions, such as joint venture virtual production systems. However, the extensive interconnection of industrial systems with corporate systems in industrial Internet of Things (IIoT) networks exposes the industrial domain to severe cyber risks. Because of many proprietary multilevel protocols, limited upgrade opportunities, heterogeneous communication infrastructures, and a very large trust boundary, conventional IT security fails to prevent cyberattacks against IIoT networks. Recent secure protocols, such as secure distributed network protocol (DNP 3.0), are limited to weak hash functions for critical response time requirements. As a complementary, we propose an adaptive trust boundary protection for IIoT networks using a deep-learning, feature-extraction-based semisupervised model. Our proposed approach is novel in that it is compatible with multilevel protocols of IIoT. The proposed approach does not require any manual effort to update the attack databases and can learn the rapidly changing natures of unknown attack models using unsupervised learnings and unlabeled data from the wild. Therefore, the proposed approach is resilient to emerging cyberattacks and their dynamic nature. The proposed approach has been verified using a real IIoT testbed. Extensive experimental analysis of the attack models and results shows that the proposed approach significantly improves the identification of attacks over conventional security control techniques.

In the context of new retail and personalized, small-batch, distributed collaborative production, orders arrive in real time, and each workshop needs to organize production lines based on suborders in real time under the constraints of smart contracts. However, the existing cloud centralized scheduling method has very high calculation and communication costs when inserting orders in real time, and the fully reactive edge scheduling method is difficult to meet the various order-level requirements of customers. Therefore, this article proposes a real-time edge scheduling model that considers real-time trial insertion of orders based on order-level requirements. Further, a real-time edge adjustment method to eliminate fluctuations is proposed. The proposed strategies were implemented using the lightest methods, avoiding cascading effects, to apply to the edge. Experimental results show that proposed strategies have significant advantages in order-level indicators such as customer satisfaction and have slightly better performance in workshop-level indicators such as resource utilization, energy consumption, and makespan.

Depression is the result of a complex interaction of social, psychological and physiological elements. Research into the brain disorders of patients suffering from depression can help doctors to understand the pathogenesis of depression and facilitate its diagnosis and treatment. Functional near-infrared spectroscopy (fNIRS) is a non-invasive approach to the detection of brain functions and activities. In this paper, a comprehensive fNIRS-based depression-processing architecture, including the layers of source, feature and model, is first established to guide the deep modeling for fNIRS. In view of the complexity of depression, we propose a methodology in the time and frequency domains for feature extraction and deep neural networks for depression recognition combined with current research. It is found that compared to non-depression people, patients with depression have a weaker encephalic area connectivity and lower level of activation in the prefrontal lobe during brain activity. Finally, based on raw data, manual features and channel correlations, the AlexNet model shows the best performance, especially in terms of the correlation features and presents an accuracy rate of 0.90 and a precision rate of 0.91, which is higher than ResNet18 and machine-learning algorithms on other data. Therefore, the correlation of brain regions can effectively recognize depression (from cases of non-depression), making it significant for the recognition of brain functions in the clinical diagnosis and treatment of depression.

Localization services play an important role in Internet of Connected Vehicles (IoCV) and vehicle predictive localization information can greatly improve traffic efficiency and reduce accidents. However, a huge amount of computing and communication overhead is required to obtain such information by traditional methods. In this work, we propose a Behavior-based Clustering Method (BCM) to analyze the behavioral correlation between vehicles and classify them into different clusters. Based on BCM results coupled with a deep learning model, we further propose a Clustering-learning-based Long-term Predictive Localization (CLPL) algorithm to predict vehicles' future location distribution. In the proposed CLPL algorithm, all the traffic roads are divided into consecutive small segments in order to pinpoint vehicles' precise current locations and to obtain long-term predictions. Extensive simulations, notably involving real dataset, have been carried out to evaluate BCM and CLPL in terms of several performance criteria including matching rates. The analysis of the results validated how the designed methods can predict vehicle location much more accurately than existing algorithms.

With the increasing number of vehicles, the traffic congestion is becoming more and more serious. In order to alleviate such a problem, this article considers transmission and inference delay of cloud centralized computing in the software defined Internet of Vehicles (SDIoV), and builds a new SDIoV architecture based on edge intelligence, for supporting real-time vehicle routing decision through distributed multi-agent reinforcement learning model. Then, a software defined device collaboration optimization method is designed to improve the efficiency of distributed training. Combined with multi-agent reinforcement learning, a distributed-learning-based vehicle routing decision algorithm (DLRD) is proposed to adaptively adjust vehicle routing online. The performed simulations show that the DLRD can successfully realize real-time routing decision for vehicles and alleviate traffic congestion with the dynamic changes of the road environment.

introduction Share on Introduction to the Special Section on Cognitive Robotics on 5G/6G Networks Authors: Huimin Lu Kyushu Institute of Technology, Japan Kyushu Institute of Technology, JapanView Profile , Liao Wu University of New South Wales, Australia University of New South Wales, AustraliaView Profile , Giancarlo Fortino University of Calabria (Unical), Italy University of Calabria (Unical), ItalyView Profile , Schahram Dustdar Vienna University of Technology, Austria Vienna University of Technology, AustriaView Profile Authors Info & Claims ACM Transactions on Internet TechnologyVolume 21Issue 4November 2021 Article No.: 91epp 1–3https://doi.org/10.1145/3476466Published:28 September 2021Publication History 1citation36DownloadsMetricsTotal Citations1Total Downloads36Last 12 Months26Last 6 weeks2 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my Alerts New Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

With the ever-increasing diffusion of smart devices and Internet of Things (IoT) applications, a completely new set of challenges have been added to the Data Mining domain. Edge Mining and Cloud Mining refer to Data Mining tasks aimed at IoT scenarios and performed according to, respectively, Cloud or Edge computing principles. Given the orthogonality and interdependence among the Data Mining task goals (e.g., accuracy, support, precision), the requirements of IoT applications (mainly bandwidth, energy saving, responsiveness, privacy preserving, and security) and the features of Edge/Cloud deployments (de-centralization, reliability, and ease of management), we propose EdgeMiningSim, a simulation-driven methodology inspired by software engineering principles for enabling IoT Data Mining. Such a methodology drives the domain experts in disclosing actionable knowledge, namely descriptive or predictive models for taking effective actions in the constrained and dynamic IoT scenario. A Smart Monitoring application is instantiated as a case study, aiming to exemplify the EdgeMiningSim approach and to show its benefits in effectively facing all those multifaceted aspects that simultaneously impact on IoT Data Mining.

The Internet of Things (IoT) is the latest example of the System of Systems (SoS), demanding for both innovative and evolutionary approaches to tame its multifaceted aspects. Over the years, different IoT methodologies, frameworks, platforms, and tools have been proposed by industry and academia, but the jumbled abundance of such development products have resulted into a high (and disheartening) entry-barrier to IoT system engineering. In this survey, we steer IoT developers by: 1) providing baseline definitions to identify the most suitable class of development products-methodologies, frameworks, platforms, and tools-for their purposes and 2) reviewing seventy relevant products through a comparative and practical approach, based on general SoS engineering features revised in the light of main IoT systems desiderata (i.e., interoperability, scalability, smartness, and autonomy). Indeed, we aim to lessen the confusion related to IoT methodologies, frameworks, platforms, and tools as well as to freeze their current state, for eventually easing the approach towards IoT system engineering.

With the development of Internet of Things (IoT) and communication technology, the number of next-generation IoT devices has increased explosively, and the delay requirement for content requests is becoming progressively higher. Fortunately, the edge-caching scheme can satisfy users' demands for low latency of content. However, the existing caching schemes are not smart enough. To address these challenges, we propose a human-like hybrid caching architecture based on the software-defined edge cloud, which simultaneously considers the content popularity and the fine-grained user characteristics. Then, an optimization problem with a caching hit ratio as an optimization objective is formulated. To solve this problem, using reinforcement learning, we design a human-like hybrid caching algorithm. The extensive experiments show that compared with popular caching schemes, human-like hybrid caching schemes can improve the cache hit ratio by 20%.

In Smart Factories, automated guided vehicles (AGVs) accomplish heterogeneous tasks as moving objects, restoring connectivity, or performing different manufacturing activities into production lines. These kinds of devices combine several capabilities, as artificial intelligence (visual and speech recognition, automatic fault detecting, proactive behavior) and mobility, into the so-called “mobile intelligence.” A typical scenario is represented by a workshop with a large number of mobile intelligent devices with associated agents, mutually interacting on their behalf. Here, to reach a given target by contemporary satisfying some basic requirements like effectiveness and efficiency, it is often necessary to organize ad hoc teams of free-moving vehicles, sensors, and smart devices. Therefore, a specific issue is the adequate representation of the reciprocal agent/device trustworthiness for advantaging such team formation processes within a smart factory environment. To this end, in this article, first, we define a trust measure based on reliability and reputation of AGVs, which are computed based on the feedbacks released for the AGVs activities in the factory; second, we design a trust framework exploiting the defined measures to support the formation of virtual, temporary, and trust-based teams of mobile intelligent devices; and third, we present a set of experimental results highlighting that the proposed trust framework can improve the workshop performance in terms of effectiveness and efficiency.

The papers in this special section focuses on artificial intelligent-driven informatics, sensing, imaging and big data analytics in dealing with the COVID-19 pandemic.

A conventional power grid is criticized by its poor capability of power usage management, especially in handling dynamically varying power demands over time. The concept of smart grid has been introduced to mitigate this problem by satisfying not only real-time power demands, but also by restricting power usage within the capacity. Its consistent outperformance and new perspective in computer intelligence to control the grid for autonomous power consumption has been gradually replacing the conventional power grid. However, even in smart grid, providing high satisfaction to users often leads smart grid operator (SGO) to loss and vice versa. In this paper, we develop an optimal dynamic pricing mechanism for trading-off (ODPT), for SGOs that tradeoff between user utility and operator profit in smart grid systems. It allows the operator to purchase power from multiple energy producers and to set selling price to users dynamically following the demand-supply theory of economics. It also exploits an artificial neural network model to more accurately predict the power usage. The simulation results, carried out on a commercially available optimization modeling tool using practical power usage data, prove the effectiveness of the proposed ODPT in increasing the operator profit while satisfying user demands.

The “Internet of Things” (IoT) is requiring increasing computational capabilities which can be obtained from Machine-to-Machine cooperation among smart objects. The convergence of IoT and Multi-Agent Systems gives the opportunity to benefit of their social attitude to interact and cooperate for services. In a mobile and federated context, selecting reliable partners for cooperation is a complex task because the most part of devices are not referenced with respect to their trustworthiness. The aggregation of agents in groups, in each IoT environment, on the basis of reputation information can provide a solution to this problem. In order to spread device/agent reputation information in our distributed scenario without using a central repository, we propose to rely on the blockchain technology to support a reputation approach to form groups. In particular, information about the reputation capital, certified by the blockchain, is used by an algorithm to drive group formation processes of agents in IoT environments.

In this paper, we describe how the microservices paradigm can be used to design and implement distributed edge services for Internet of Things applications. As a case study, traditionally monolithic user mobility analysis service is developed, with distributed and extendable microservices, for the standardized ETSI MEC system reference architecture. In each of the edge system three tiers, microservices implement the service logic with components for movement trace analysis, movement prediction and visualization of the results. The distributed service is implemented with Docker containers and evaluated on real-world settings with low capacity edge servers and real user mobility data. The results show that the edge promise of low latency can be met in such as implementation. The integration of a software development technology with a standardized edge system provides solid background for further development.

In the last few years, the use of Cloud and Internet of Things (IoT) technologies is dramatically growing, opening new application possibilities spanning from smart cities to the prevention of natural disasters. A flexible integration of the Cloud and IoT environments allows users and applications to combine the benefits of the two worlds. Specifically, robustness, versatility and high computation power are better supported by the Cloud, while better real-time adaptation and local control are peculiar characteristics of the IoT infrastructure. Even higher flexibility can emerge when the cloud layer combines and integrates a public cloud component, offered by an external provider, and a private cloud, managed internally to the organization, thus giving rise to the so-called hybrid cloud. The integration of a hybrid cloud with an IoT layer opens a wide range of new possibilities and applications but also presents heterogeneity and complexity issues that must be carefully addressed. This paper offers a contribution in this field by presenting a novel threelayered IoT architecture for the Hybrid Cloud, namely IoT-HC, which is devoted to enabling distributed sensing and actuation, controlling IoT networks, elaborating the collected data both at edge and cloud level, and deciding which data has to be sent to private or public cloud components for storage and high-level elaboration.

The Internet of Things (IoT) enables the interconnection of new cyber-physical devices which generate significant traffic of distributed, heterogeneous and dynamic data at the network edge. Since several IoT applications demand for short response times (e.g., industrial applications, emergency management, real-time systems) and, at the same time, rely on resource-constrained devices, the adoption of traditional Data Mining techniques is neither effective nor efficient. Therefore, conventional Data Mining techniques need to be adjusted for optimizing response times, energy consumption and data traffic while still providing adequate accuracy as required by the IoT application. In this paper, new Data Mining approaches particularly tailored for the IoT scenario have been investigated, in particular with respect to the promising, emerging novel distributed computing paradigm of Edge Computing. In detail, two approximated versions of K-Means clustering algorithm, centralized and distributed, have been implemented in the EdgeCloudSim simulation framework and validated on a real system. As highlighted by the algorithm performance analysis, choosing an approximated and distributed clustering solution can provide benefits in terms of computation, communication and energy consumption, while maintaining high levels of accuracy. The management of such trade-off, obviously, has to be done in the light of the specific IoT application requirements.

In a mobile crowdsourcing system (MCS), a platform outsources sensing tasks to numerous mobile worker devices. The collected data are analyzed and the processed information is shared among many other interested users. The platform pays the workers for the sensing data and earns money from the users receiving processed information services. Distributing the sensing workloads among the potential workers so as to maintain the required data quality and to make a reasonable amount of profit is a challenging problem for such a platform. In this paper, we develop a workload allocation policy that makes a reasonable tradeoff between worker utilities and platform profit. It quantifies the utility (i.e., the quality of sensed data) of a worker as a function of worker mobility, current location, and past sensing records. The workload allocation problem is formulated as a multiobjective nonlinear programming (MONLP) problem which aims to make the desired tradeoff between worker utilities and platform profit. The allocation problem is shown to be NP-hard and thus we develop two greedy algorithms with relaxed constraints to achieve polynomial time solutions. Performance of the proposed workload allocation policy is evaluated in a distributed computation environment using MATLAB. The results show its effectiveness compared to state-of-the-art methods in terms of platform profit, quality of sensing data, and request service satisfaction.

Edge computing paradigm has attracted many interests in the last few years as a valid alternative to the standard cloud-based approaches to reduce the interaction timing and the huge amount of data coming from Internet of Things (IoT) devices toward the Internet. In the next future, Edge-based approaches will be essential to support time-dependent applications in the Industry 4.0 context; thus, the paper proposes BodyEdge, a novel architecture well suited for human-centric applications, in the context of the emerging healthcare industry. It consists of a tiny mobile client module and a performing edge gateway supporting multiradio and multitechnology communication to collect and locally process data coming from different scenarios; moreover, it also exploits the facilities made available from both private and public cloud platforms to guarantee a high flexibility, robustness, and adaptive service level. The advantages of the designed software platform have been evaluated in terms of reduced transmitted data and processing time through a real implementation on different hardware platforms. The conducted study also highlighted the network conditions (data load and processing delay) in which BodyEdge is a valid and inexpensive solution for healthcare application scenarios.

Driven by the vision of edge computing and the success of rich cognitive services based on artificial intelligence, a new computing paradigm, edge cognitive computing (ECC), is a promising approach that applies cognitive computing at the edge of the network. ECC has the potential to provide the cognition of users and network environmental information, and further to provide elastic cognitive computing services to achieve a higher energy efficiency and a higher Quality of Experience (QoE) compared to edge computing. This article first introduces our architecture of the ECC and then describes its design issues in detail. Moreover, we propose an ECC-based dynamic service migration mechanism to provide insight into how cognitive computing is combined with edge computing. In order to evaluate the proposed mechanism, a practical platform for dynamic service migration is built up, where the services are migrated based on the behavioral cognition of a mobile user. The experimental results show that the proposed ECC architecture has ultra-low latency and a high user experience, while providing better service to the user, saving computing resources, and achieving a high energy efficiency.

Wireless sensor networks (WSNs) have well known limitations such as battery energy, computing power and bandwidth resources that sometimes limit their widespread use. Current researches are mainly concentrated to propose solutions for nodes energy optimization, network load balancing and the improvement of WSN robustness; however, the software defined network (SDN) paradigm uses the theory of forwarding phase separating from control, simplifying management and configuration of the network to improve network extension and flexibility. It could further optimize WSNs deployment and improve their transmission performance. In this paper, we firstly describe the general architecture and the main features of software defined networks; then, we analyze the current integrated SD-WSN scheme and summarize these results in detail.

Smart Environments (SEs) are pervasive systems usually built on top of IoT-based sensing and actuation devices which are spread in an environment. The increase of the on-board computational capacity of the used devices opens to the possibility of naturally exploiting the edge computing paradigm in which the computation is pushed at the edge of the network. Anyway, despite the huge interest towards SEs, there is a lack of approaches for their design. This paper proposes an enhancement of the existing Smart Environment Metamodel (SEM) framework suited for designing SEs. The provided extension aims at taking into account issues related to edge computing, management of timing information and definition of the data types involved in data sources. The effectiveness of the whole proposal is assessed through a case study describing the development of a Smart Office.

The integration of existing and future smart cyberphysical systems within a fully realized Internet of Things (IoT) cannot dismiss the requirement of interoperability. The absence of standards for the IoT along with its intrinsic complexity demand for proper methodologies in order to fully support the development of heterogeneous, but interoperable, IoT systems as well as their integration. However, at the state-of-the-art, no methodologies for IoT systems integration are available. To fill this gap, in this paper the INTER-METH engineering methodology is presented. Developed in the context of the European H2020 project named INTER-IoT, INTER-METH supports the integration process of heterogeneous IoT platforms from the analysis to the maintenance phase. Its main features as well as its abstract and instantiated process schema are described; in particular, in this paper the focus is on the analysis phase that is fundamental for driving the integration process.

Existing routing protocols of wireless sensor networks (WSNs) attempted to optimize the energy efficiency and the routing reliability from the perspective of the network itself and failed to take into consideration the environmental impact from outside, causing them cannot make prompt reaction to the dynamic changes of the environments (e.g., wildfire). Thus, in these routing protocols the routing survivability under harsh environments is questionable. To tackle this issue, in this paper by referencing the concept of potential field, we design an environment-cognitive multipath routing protocol (ECMRP) in order to provide sustainable message forwarding service under harsh environments. In ECMRP, routing decisions are made according to a mixed potential field in terms of depth, residual energy and environment. The basic idea of this approach is to instruct data packets to select routes with the tradeoff among latency, energy conservation and routing survivability. As the environmental field is constructed and updated using the sensing capability of WSN itself, constructed routes can avoid crossing through the danger zones to keep the paths safe. The experimental results show that ECMRP can obtain significant improvements in packet delivery ratio and network lifetime under harsh conditions.

This paper proposes a method for the detection of emotion-relevant activities performed when seated and its corresponding system based on wrist-worn inertial sensors combined with a pressure detection smart cushion. In particular, aiming at providing an additional source of information in traditional emotion recognition systems, we focus on shame-, fear-, and joy-related activities. Experiments are conducted and the results of performance evaluation show the proposed method achieves high recognition accuracy with a set obtained by fusing time-and frequency-domain features extracted from the different available sensors.

The concept of Social Internet of Things (SIoT) has emerged from the integration of social networking into the core of the Internet of Things (IoT). It envisions IoT objects and devices to have social interactions with each other autonomously, cooperate with other agents, and exchange information with human users and surrounding computing devices. These objects are able to sense/actuate, store, and interpret information in an opportunistic and loosely coupled fashion. The objects in the SIoT paradigm can exhibit multiple forms of social relationships derived from their collaborative activities or functional, temporal and spatial dependencies to meet a particular need of human users, which signify the difference between the SIoT domain to that of social-based mobile networks or sensor networks. The social interaction among the SIoT objects contribute a huge volume of data to be processed and used by various applications such as social VANET, social connected health, SIoT-based recommendation service, traffic service, policing, energy management etc, in the area of Smart Cities, Smart Homes, Smart Grid, and Smart Factories to satisfy human needs, interests, and objectives. Such a dynamic landscape with billions of social communities of objects and devices requires new models, theories, and approaches of interaction and collaboration, which could be established by referring to the experience that people have already gained in social networking domain over the past few years.

Social Internet of Things (SIoT) is a new paradigm where Internet of Things (IoT) merges with social networks, allowing people and devices to interact, and facilitating information sharing. However, security and privacy issues are a great challenge for IoT but they are also enabling factors to create a “trust ecosystem.” In fact, the intrinsic vulnerabilities of IoT devices, with limited resources and heterogeneous technologies, together with the lack of specifically designed IoT standards, represent a fertile ground for the expansion of specific cyber threats. In this paper, we try to bring order on the IoT security panorama providing a taxonomic analysis from the perspective of the three main key layers of the IoT system model: 1) perception; 2) transportation; and 3) application levels. As a result of the analysis, we will highlight the most critical issues with the aim of guiding future research directions.

Social development and technological advances have enabled the Internet of Vehicles (IoV) to combine the social factors to form a new intelligent transportation system: Social IoV (SIoV). The emergence of SIoV helps to find new traffic management solutions of the serious problems caused by the ever-increasing traffic flow. In this paper, we propose an algorithm called social vehicle route selection (SVRS) to reduce traffic congestion and achieve the purpose of traffic flow control. First, a social clustering method for SIoV is designed by utilizing both historical and current driving information. Then we use game evolution to calculate the optimal route for vehicles, and prove the vehicle route selection game is a potential game and its strategy selection converges to Nash equilibrium. Extensive simulations are carried out to evaluate the SVRS with several performance criteria. Our analysis and simulation results demonstrate that SVRS algorithm can achieve high performance in clustering the vehicles and reducing traffic congestion.

Exposure to air pollution poses a significant risk to human health, particularly to urban dwellers. When correlated with individual health outcomes, high resolution information on human mobility, and the spatial and temporal distribution of the pollutants can lead to a better understanding of the effects of pollution exposure. People-centric sensing is normally carried out by data sharing through a central cloud server. This system architecture is not designed to serve the ever-growing number of high fidelity connected devices, particularly when crowdsourcing urban data on location and environmental conditions. Here, we outline an architecture for a people-centric and cognitive Internet of Things (PIoT) environmental sensing platform, which involves closed loops of interactions among people nodes and physical devices as well as servers and recommendations on device connections by cognitive computing. Taking advantage of smart objects and virtual node technology in PIoT, an algorithm to aggregate on-demand user data from smart devices is proposed. A PIoT prototype sensing system is designed and deployed to measure the space-time distribution of particulate matter in air (PM2.5), and mobility counts, for quantifying personal exposure to air pollution. A case study of particulate matter PM2.5exposure in New York City is presented to illustrate the potential application of people-centric measurement system and data analysis.

Cognitive Internet of Things (IoT) is the use of cognitive computing technologies, which is derived from cognitive science and artificial intelligence, in combination with data generated by connected devices and the actions those devices can perform. Cognitive IoT provides high performance of communicating, computing, controlling, and even high degree of machine intelligence. Cognitive IoT redefines the relationship between human and their pervasive digital environment. They may play the role of assistant or coach for the user. Specifically, the IoT generated big data, when used to power predictive analytics algorithms or to develop a corps for a cognitive computing solution, can provide insights that would never be discovered in time to be useful if the departmental silos do not collaboration in data sensing and analysis. It is the integration of this data that enables cognitive computing applications for IoT of the next decade. Therefore, the services of a cognitive IoT could be constructive, prescriptive, or instructive in nature.

In a smart factory environment, a much larger amount of data are transmitted in the workshop networks bringing big challenges to data transfer capability and energy usage efficiency. In the workshop, two main networks, i.e., wired/wireless fieldbus networks and wireless sensor networks, are usually used to collect and transmit data separately; thus, this paper proposes a mobile and hierarchical data transmission architecture to integrate these two networks also taking advantages from the existing mobile intelligence in smart factories, such as automatic guided vehicles (AGVs), to implement a novel data and materials delivery scheme well suited for modern industrial wireless sensor networks (IWSNs). Simulation experiments demonstrated how the proposed approach, running within the IWSN, significantly increases data delivery efficiency along with achieving better energy usage, by 4 times, with respect to the separated networks without any mobile intelligence support.

The seven papers in this special section focus on the research and development of service-oriented collaborative computing technologies and their applications to the design of products, processes, systems and services in an industrial and social viewpoint.

The future Internet of Things (IoT) is expected to enable a new and wide range of decentralized systems (from small-scale smart homes to large-scale smart cities) in which “things” are able to sense/actuate, compute, and communicate, and thus play a central and crucial role. The growing importance of such novel networked cyber-physical context demands suitable and effective computing paradigms to fulfill the various requirements of IoT systems engineering. In this paper, we propose to explore an agent-based computing paradigm to support IoT systems analysis, design, and implementation. The synergic meeting of agents with IoT makes it possible to develop smart and dynamic IoT systems of diverse scales. Our agent-oriented approach is specifically based on the agent-based cooperating smart object (ACOSO) methodology and on the related ACOSO middleware: they provide effective agent design and programming models along with efficient tools for the actual construction of an IoT system in terms of a multiagent system. A case study concerning the development of a complex IoT system, namely a Smart University Campus, is described to show the effectiveness and efficiency of the proposed approach.

Originated from Wireless Sensor Networks (WSNs), Body Sensor Networks (BSNs) have been applied to numerous domains. However, after an in-depth analysis of the state-of-the-art, several factors have been found to limit the development of applications based on BSNs. In this paper we introduce the concept of Open BSNs for improving the development of BSNs. Specifically, open BSNs can improve key aspects such as energy efficiency, system interoperability, system usability and scalability, and privacy support. Application scenarios and future research challenges of Open BSNs are also presented.

Mobility impaired individuals need the wheelchair to support their independent life, so monitor activities performed on the wheelchair can provide significant insights on their general health status. Activity recognition related to healthy people is a well established research area; however, only few works addressed this problem for wheelchair users. This paper proposes a novel approach based on dynamic Bayesian networks to recognize physical activities performed on a wheelchair. We equipped the wheelchair seat with a pressure detection unit and attached two inertial measurement units on the user's wrists. We focus on common basic activities and specifically, to experimentally evaluate our method, we defined four dynamic activities (moving forward, moving backward, moving left-circle, moving right-circle) and two static activities (left-right swing, forward-backward swing). Data is collected using a smart wheelchair system we developed in previous research. Firstly, we generate the posture sequence from the pressure signals and detect the raw acceleration data from inertial measurement units; then, we fuse the posture sequence and inertial features to detect the postural-based activities. Results shows that our proposed method can achieve an overall classification accuracy of 91.88%.

A smart environment is a physical environment enriched with sensing, actuation, communication and computation capabilities aiming at acquiring and exploiting knowledge about the environment so as to adapt it to inhabitants' preferences and requirements. In this domain, there is the need of tools supporting the design and analysis of applications. In this paper, a meta-model framework for smart environments is proposed. This framework allows to model applications by exploiting concepts closer to the smart environment domain. The proposed meta-model framework approaches the modelling from two different points of view, namely the functional and data perspectives. The functional perspective focuses on the services provided by the environment whereas the data perspective is used to characterize data sources of the environment. The effectiveness of the proposal is shown by applying it to the modelling of a smart environment scenario well known in literature.

In recent years, the elderly population is increasing enormously, from 9% in 1994 to 12% in 2014, and is expected to reach 21% by 2050. Elderly live often alone today and even conducting an independent daily life, some of them move with the aid of walkers or using wheelchairs. Monitoring elderly activity in mobility has become a major priority to provide them an effective care service. This paper focuses on an enhancement of a smart wheelchair based on pressure sensors to monitor users sitting on the wheelchair. If the wheelchair user assumes a dangerous posture, the system triggers audio/visual alarms to avoid critical consequences such as wheelchair overturn. The paper discusses the hardware design of the system, then analyzes and compares posture recognition methods that have been applied on pressure data we collected. The experiments demonstrate the effectiveness of the proposed method and 99.5% posture recognition accuracy has been observed.

Internet of Things (IoT) networks are being continually developed in several domains, however no systematic processes for their modeling and simulation exist so far.In this paper, an agent-oriented approach to IoT networks modeling is proposed by exploiting the ACOSO model.Then, agent-modelled IoT networks of different scales are simulated through the Omnet++ simulation platform, with the goal of analyzing issues and bottlenecks at communication level.

Smart Cyber-Physical Environments are augmented physical environments whose behaviours are enhanced through the use of ICT technologies. The goal is to offer new services and functionalities devoted to meet people's needs and preferences, and to better exploit existing services and infrastructures. The use of IoT technologies, paired with the edge computing, fosters the development of Smart Environment applications having the important features of reliability, scalability and extensibility. This paper proposes an approach for the design and the implementation of Smart Cyber Physical Environment applications having the aforementioned features. The approach relies on the use of isapiens which is an IoT platform enabling edge computing through the exploitation of the agent metaphor. Such platform provides effective abstractions which are able to hide heterogeneity of both the adopted hardware devices and communication protocols. The approach is validated through a case study involving the realization of a Smart Office prototype for profiling and monitoring daily working activities and performing actuations in the environment on the basis of the obtained information.

This paper presents a monitoring system (Canoe-Sense) for canoe motion based on wearable Body Sensor Networks (BSNs). An effective motion segmentation method was applied to competitive sport, which can segment human motion phases automatically based on raw time series data that was acquired through wearable Inertial Measurement Units (IMUs). Orientation estimation algorithm was adopted to measure the attitude information of athletes' stroke motion of the canoe. By fusing the data of motion phases and attitude changes, the monitoring data may provide coach with a new performance monitoring method for improving coordination motions of two partners or adjusting the training plan in time. The experimental results showed that our system is able to simultaneously monitor motion phases and attitude changes of two athletes during training on the water.

Presents the introductory editorial for this issue of the publication.

Cardiac Defense Response (CDR) is a basic psycho-physiological response related to startle reflex and preceding negative emotional states including fear. In the health-care context, the definition of methods to automatically identify the CDR is a relevant issue, because frequent CDR activations (not associated to proper danger stimuli) can pose the subject to health risk and eventually develop into severe psychophysical disorders. Therefore, providing tools for automatic identification of this defense mechanism can significantly help psychologists and caregivers in understanding the patient's mental and health status as well as patients themselves to self-regulate and self-control against excessive defense and stress responses. This work discusses and compares different methods and specifically proposes a novel algorithm designed to detect the CDR by analyzing the electrocardiogram (ECG) signal. It is based on the extraction of specific features from a signal, directly generated from the ECG, which are compared against an ad-hoc computed reference CDR template. The proposed method has been tested on real ECG traces, a number of them containing full activations of the CDR pattern, and compared against other techniques, discussed in the paper, reaching an improvement of 10 percent in sensitivity, 18 percent in specificity, and 24 percent in precision with respect to the best performance of the other related methods.

Body Sensor Networks (BSNs) have become prominent in research and industry alike as a powerful enabler of novel applications in human-centered domains. However, developing applications on such systems is still a cumbersome process, due to the lack of suitable software abstractions and the difficulties in managing wearable computing application within the stringent constraints of embedded systems. In this paper, we introduce a novel framework, SPINE2 (Signal Processing In Node Environment), which allows task-oriented programming on a platform-independent architecture. We demonstrate how fairly sophisticated signal-processing applications can be realized in the form of easy-to-implement embedded processes. The proposed architecture is tested experimentally and its features are illustrated through a nontrivial case study. In the last years, several frameworks and middlewares have been conceived and made available to support high-level programming in WSNs. These provide a generic set of features that can only be used for the most common application domains. However, it is hard to efficiently support the more specific domain of BSNs, which requires specific capabilities. In order to fully satisfy the BSN-based requirements, SPINE2 has been conceived as an effective and efficient tool for developing distributed signal-processing applications. Its task-oriented paradigm allows developers to specify the applications' behavior by abstracting away any low-level details concerning the platform hardware and the communication protocol. Moreover, its platform-independent architecture enables code reusability and portability, as well as application interoperability and platform heterogeneity. To demonstrate the effectiveness of the proposed framework and the efficiency of the runtime environment, a BSN-based activity recognition system has been developed through SPINE2. The easiness in implementing such a complex system thanks to both the provided programming abstractions and the framework components reusability is shown, as well as the efficiency of the whole system whose performance has been evaluated under a range of metrics.

Location service is one of the primary services in smart automated systems of Internet of Things (IoT). For various location-based services, accurate localization has become a key issue. Recently, research on IoT localization systems for smart buildings has been attracting increasing attention. In this paper, we propose a novel localization approach that utilizes the neighbor relative received signal strength to build the fingerprint database and adopts a Markov-chain prediction model to assist positioning. The approach is called the novel localization method (LNM) in short. In the proposed LNM scheme, the history data of the pedestrian's locations are analyzed to further lower the unpredictable signal fluctuations in a smart building environment, meanwhile enabling calibration-free positioning for various devices. The performance evaluation conducted in a realistic environment shows that the presented method demonstrates superior localization performance compared with well-known existing schemes, especially when the problems of device heterogeneity and WiFi signals fluctuation exist.

In 1999, Kevin Ashton envisioned a novel paradigm named Internet of Things (IoT), in which all things could see, hear, and smell the world for themselves, and interact with each other and cooperate with their neighbors to reach some common desired goals. In the following years, the IoT ideas started to spread rapidly due to the technology advancements in the fields of microelectromechanical systems and most recently, nanoelectromechanical systems, computers, and wireless communications, resulting in autonomous everyday thing augmented with sensing/actuation, storage, processing, and network capabilities. Their new applications emerged daily from smart homes to smart cities, from automobiles to high-speed trains, from new-born care devices to patient operating rooms and entire hospitals, and from manufacturing factories to agricultural food plants. IoT is one of the fastest growing technical areas across almost all engineering disciplines and touches almost all verticals of the World Economy. It represents major investments in commercial and government initiatives. We expect to have over 40% Compound Annual Growth Rate year over year in the commercial marketplace and to dominate “traffic” on the Internet within the next decade.

Welcome to the 2015 19th IEEE International Conference on Computer Supported Cooperative Work in Design (CSCWD 2015), May 6–8. With great pleasure, we welcome you to the wonderful Tyrrenian coast of the Calabria region, Italy.

This paper describes a novel integrated system for detecting, monitoring, and securely recording human physical activities using wearable sensors, a personal mobile device, and a Cloud-computing infrastructure supported by the BodyCloud [1] platform. Body Sensor Networks (BSNs), empowered by wireless non-invasive wearable physiological sensors, have been widely accepted as one of the key enabling technologies for the revolution of personal-health services. In addition, the integration of BSN applications with Cloud-computing technologies can effectively supports the diffusion of such services in our daily life. Many of these personal-health systems - regardless of their final aim - are based, use or are supported by contextual information on user's physical activity (body posture, movement or action) being performed. This work, hence, aims at providing a basic physical activity service that is capable of supporting personal, mobile-Health applications with real-time activity recognition and labeling both on the personal mobile device and on the Cloud.

How to shorten time delay and enhance delivery ratio is still an open problem in the study of opportunistic networks. Most proposals are trying to deal with this issue by introducing infrastructures. Although related research has been proven to be useful in improving the routing performance of the network, there is still room for further improvement. In this article, inspired by the powerful message synchronization capability of wireless sensor networks (WSNs), we propose a new opportunistic network framework called WON that introduces WSNs into opportunistic network. With the support of WSNs, fast message delivery and high success ratio can be achieved. We specifically present the layered architecture of WON and compare WON to existing architectures in opportunistic networking. The simulation results concerning delivery delay and success ratio are highly encouraging. Finally, open issues are outlined.

A federation of cloud providers (CPs) consists of a set of self-interested CPs that cooperate in order to provide virtual machine (VM) resources requested by users. The CPs by virtue of being part of a federation can make some profit by selling their unused VM capacity. This paper presents an efficient mechanism of resource and revenue sharing in a cloud federation that motivates the CPs to cooperate. The proposed mechanism models the interactions among the CPs in a federation as a coalition game. In contrast to existing approaches, the game model aims at maximizing the social welfare or total profit of the CPs in a federation to promote their long-term individual profit. In addition, we present a comprehensive analysis of the related costs and revenue associated with the various decisions of the CPs related to their joining in a federation. Various simulations were carried out to validate and verify the effectiveness of the proposed cooperative capacity sharing mechanism. Simulation results demonstrated that the proposed mechanism can satisfy the fairness and stability properties, maximize the social welfare the CPs in a federation and achieve cost effective resource sharing.

The paper proposes AirGround, a flexible and expandable framework to support the collaboration and the coordination between aerial and terrestrial drones in order to accomplish a common mission in a more effective and fast way. In particular the designed application-level framework allows the dynamic tasks assignment to the different devices involved into the communication process and the distributed leader election according to specific executive parameters and system conditions (i.e., residual energy, computational power, abilities offered by specific on board sensors). The AirGround effectiveness has been evaluated throughout a real testbed to measure the overall system performance in terms of both neighbour discovery and leader election speed by increasing the number of drones in different channel and environmental conditions.

Swarm robotics is focused on implementation of systems which are composed of multiple simple robots rather than one single complex robot The challenge is to develop a group of robots with simple perception and communication capabilities in order to complete a task in a collective and distributed manner without central leader. In this paper, we present cooperative and non cooperative models for foraging that exploit stigmergy in the context of the classical Army Ant Raid model. Such models use the Stigmergic Multi-Ant Search Area (S-MASA) algorithm that produces a gradual search around the nest which provides less time for locating closest food and provides close-to-optimal paths using only the pheromone concentration. The proposed models have been evaluated under simulation with respect to models based on the reference c-marking algorithm. The obtained results show that the proposed models, specifically the cooperative one outperforms the c-marking based models both in obstacle-free and obstacle environments.

A major challenge in swarm robotics is to minimize energy and time costs.We focus in this paper on multi-agent foraging algorithms that uses ant-like agents with limited energy.By considering energy consumption, we propose a new Energy aware Cooperative Switching Algorithm for Foraging (EC-SAF) that optimizes the whole system search and transport operations needed to collect resources over time.Unnecessary moves are avoided according to the following two premises: (1) Quick search and optimal paths provided by Stigmergic Multi-Ant Search Area (S-MASA) algorithm; (2) Quick homing provided by using the optimal paths created while searching.Results indicate that EC-SAF is promising in reducing swarm energy consumption compared to an energy-aware version of the c-marking algorithm (Ec-marking).

As the number of mobile devices increases, so do the complexity of wireless networks and the user's requirements. This tendency makes necessary for Multimedia Services to take the needed actions to adapt to the upcoming technology. A prominent example of this type of services is HTTP Adaptive Video Streaming Applications. In this research, we have studied how the latest HTTP Adaptive Streaming techniques, mainly developed for standard computers, could be adapted and used in mobile wireless devices. Furthermore, inspired by these solutions, which usually make use of Reinforcement Learning (RL) algorithms to find the suitable streaming rate, we have conceived a novel smart video player client in Java for Android platform using the Dynamic Adaptive Streaming over HTTP (DASH) protocol. We have assessed the performance of our proposed solution in a self-developed wireless test-bed under different network conditions. Thus, we have seen that by including in the reward function contributions regarding the download speed of the video segments, especially needed due to the fluctuating nature of the wireless networks, and the segments already buffered, improves drastically the overall performance of the video client. Besides that, we have discovered that, in a cognitive adaptive approach, bandwidth constraints affect the user's experience more substantially, while impairments such as packet loss can be prevented.

The Internet of Things (IoT) is a large-scale complex networked cyber physical system in which the Smart Objects (SOs) will be the fundamental building blocks. Although, many research efforts in the IoT realm have been to date devoted to device, networking and application service perspectives, software engineering approaches for the development of IoT systems are still in their infancy. This paper introduces a novel software engineering approach aiming to support a systematic development of SOs-based systems. The proposed approach is based on metamodels that are defined at different levels of abstraction to support the development phases of analysis, design and implementation. The effectiveness of the proposed approach is demonstrated through a simple yet effective case study, showing the development of a smart office SO from the high-level design to its agent-based implementation.

Summary Personal security is an open problem in large cities. After several attempts to reduce violence and crime, there seems to be an agreement that preventive actions are the best way to address this problem. Trying to help deal with that challenge, this paper proposes a mobile collaborative application, named Personal Guardian, which is used by civilians while walking in urban areas. The application is focused on crime prevention and it implements participatory sensing to help people be aware of the risks that appear to exist in a certain place at a certain time. Based on that information, citizens can take appropriate and on‐time preventive actions. The system is supported by a human‐centric wireless sensor network, and it is complementary to the security solutions already used by public and private organizations. The system architecture and its main components are described, and the main requirements and design decisions are also discussed. A preliminary evaluation of the solution was conducted to determine its strengths and weaknesses in terms of quality of service. The obtained results indicate that the information feeding process is more relevant for end‐users than the unattended delivery of awareness information about their personal security. In addition, this former capability does not require to be adjusted to the end‐users' context. Copyright © 2014 John Wiley & Sons, Ltd.

Availability, resilience, and fault tolerance of internet and distributed computing systemsThe emergence of Web as a ubiquitous platform for innovations has laid the foundation for the rapid growth of the Internet.Side-by-side, the use of mobile and wireless devices such as PDAs, laptops, and cell phones for accessing the Internet has paved the ways for related technologies to flourish through recent developments.However, development of these technologies and systems adds complexity due to the performance, fault tolerance, and availability requirements.These issues are required to be explored for the effective design and implementation of high-performance modern computing systems.In this special issue, we are delighted to present ten technical papers on resilient, highly available, and fault-tolerant Internet and distributed systems, with a particular focus on the practical experiences with the design and implementation of related technologies as well as their theoretical perspectives.These papers were selected out of 80 submissions from 17 countries in the 6th International Conference on Internet and Distributed Computing Systems (IDCS 2013).The selection has been very rigorous, and only the best papers in the conference were selected.Performance and dependability evaluation plays a key role in the design of a broad range of systems, especially when strict requirements need to be met.In the first paper, 'Variable Operating Conditions in Distributed Systems: Modeling and Evaluation' [1], Longo et al. present an analytical method that allows the study of a class of systems where different operating conditions alternate changing the stochastic behavior of the system components but still preserving the continuity of the performance and dependability quantities to investigate.In the second paper, 'Supporting Personal Security Using Participatory Sensing' [2], Carreno et al. propose a mobile collaborative application, named Personal Guardian, which used by civilians while walking in urban areas.The application is focused on crime prevention, and it implements participatory sensing to help people be aware of the risks that appear to exist in a certain place.This is an interesting application on information feeding process and the unattended delivery of awareness information about personal security.In the third paper, 'Self-healing Wireless Sensor Networks' [3], A. Miyaji and K. Omote propose three self-healing schemes to provide availability in wireless sensor networks.This is very important for long-term use of wireless sensor networks.These self-healing schemes are effective and efficient according to the evaluations done by the authors.Smooth data streaming in wireless sensor networks is a challenging problem.In the fourth paper, 'A Slot Demand based Path Reservation Approach for the Timely and Reliable Delivery of Bursty Traffic in WSNs' [4], P. V. Vinh and H. Oh propose an efficient approach to deliver bursty data reliably by reserving time slots to transmit the required packets on all the paths from the selected multimedia nodes to the server.The nodes that are not on the selected paths are put into sleep mode to conserve energy during the transmission.In the fifth paper, 'SCRRM: A Stability-Aware Cooperative Routing Scheme for Reliable High-Speed Data Transmission in Multi-Rate Mobile Ad-hoc Wireless Networks' [5], L. T. Dung and B. An present a stability-aware cooperative routing scheme for reliable high-speed data transmission in multi-rate mobile ad-hoc wireless networks, called SCRRM, to provide high data transmission with stable and reliable routes.The performance evaluation shows that this scheme can adaptively select optimal data rate and outperforms single rate routing protocol in terms of packet delivery ratio, network throughput, and average end-to-end delay in all settings of node density and node mobility.

Wireless Sensor Networks (WSNs) introduce significant flexibility in the management of smart buildings. The elimination of wires allows to deploy battery-powered sensor and actuator nodes everywhere in both new, old and historical buildings. Thus, WSNs can allow to monitor the building structural health, to manage electric loads for energy saving, to control the comfort environmental conditions, to help building inhabitants with safety procedures. The use of management frameworks (both general-purpose and domain-specific) is very important for this kind of networks as they transparently introduce characteristics such as efficient organization of the nodes, fast reconfiguration of the network, battery consumption optimization, and low-level and high-level data and event processing. The aim of this paper is to introduce an evaluation framework for the comparative analysis of systems that allow the management of Building WSNs. Specifically, a set of both qualitative and quantitative (performance evaluation) indicators have been defined on the basis of a study of the state of the art. Finally, the defined framework is used to evaluate the Building Management Framework system.

In the future Internet of Things (IoT), smart objects will be the fundamental building blocks for the creation of cyber-physical smart pervasive systems in a great variety of application domains ranging from health-care to transportation, from logistics to smart grid and cities. The implementation of a smart objects-oriented IoT is a complex challenge as distributed, autonomous, and heterogeneous IoT components at different levels of abstractions and granularity need to cooperate among themselves, with conventional networked IT infrastructures, and also with human users. In this paper, we propose the integration of two complementary mainstream paradigms for large-scale distributed computing: Agents and Cloud. Agent-based computing can support the development of decentralized, dynamic, cooperating and open IoT systems in terms of multi-agent systems. Cloud computing can enhance the IoT objects with high performance computing capabilities and huge storage resources. In particular, we introduce a cloud-assisted and agent-oriented IoT architecture that will be realized through ACOSO, an agent-oriented middleware for cooperating smart objects, and BodyCloud, a sensor-cloud infrastructure for large-scale sensor-based systems.

We explore the on-line problem of coverage where multiple agents have to find a target whose position is unknown, and without a prior global information about the environment.In this paper a novel algorithm for multi-target search is described, it is inspired from water vortex dynamics and based on the principle of pheromone-based communication.According to this algorithm, called S-MASA (Stigmergic Multi Ant Search Area), the agents search nearby their base incrementally using turns around their center and around each other, until the target is found, with only a group of simple distributed cooperative Ant like agents, which communicate indirectly via depositing/detecting markers.This work improves the search performance in comparison with random walk and S-random walk (stigmergic random walk) strategies, we show the obtained results using computer simulations.

Sensor Networks are increasingly playing a fundamental role in many application scenarios such as agriculture, maritime, healthcare, industrial and even military application. However, the high heterogeneity of sensor networks poses a great challenge in interoperability and cooperative work, and the building of technological bridges among wireless sensor network (WSN) islands is more and more a must for smart cities for an efficient operation. In this paper, we present a model for an Area Sensor Network (ASN) that combines and connects small networks (Body Sensor Networks), WSNs and even fixed sensor networks within a particular area of interest. Initial tests show that the combination of sensed data from multiple sources (sensor networks) produces synergetic services useful for smart cities.

Monitoring human movements using wireless wearable sensors finds applications in a variety of domains including healthcare and wellness. In these systems, sensory devices are tightly integrated with the human body and infer status of the user through signal and information processing. Typically, highly accurate observations can be made at the cost of deploying a sufficiently large number of sensors, which in turn results in increased energy consumption of the system and reduced adherence to using the system. Therefore, optimizing power consumption of the system while maintaining acceptable accuracy plays a crucial role in realizing these stringent resource constraint systems. In this paper, we present an activity monitoring approach that minimizes power consumption of the system subject to a lower bound on the classification accuracy. The system utilizes computationally simple template-matching blocks that perform classifications on individual sensor nodes. The system further employs a boosting approach to enhance accuracy of the distributed classifier by selecting a subset of sensors optimized in terms of power consumption and capable of achieving a given lower bound accuracy criterion. A proof-of-concept evaluation with three participants performing 14 transitional actions was conducted, where collected signals were segmented and labeled manually for each action. The results indicated that the proposed approach provides more than a 65% reduction in the power consumption of the signal processing, while maintaining 80% sensitivity in classifying human movements.

In this paper we focus on state-of-the-art analysis and open research issues in the context of Cloud-enabled large-scale sensor networks, which naturally marry with the emerging Big Data paradigm. We particularly address how data and processes are represented and managed in such infrastructures, by highlighting benefits and limitations. We also provide rigorous and critical discussion on actual trends and solutions available in literature, along with future research directions in this scientific field.

Network invulnerability is an important property of networks that operate under very likely physical attacks and failures due to operating environmental conditions. A notable example of such networks is wireless fire alarming networks (WFANs) that are strongly related to the safety of the public and to the efficiency of rescuing. WFANs based on wireless sensor networks (WSN) are gaining momentum as they considered a viable and effective solution. However, the current research on invulnerability in the WSN domain mainly focuses on the optimization of the sensor node layout in the initialized network and on routing protocols, whereas the importance of optimization of the deployed network is less explored. In this paper, we show that the invulnerability of WSNs can be improved by introducing two new elements: super wires and super nodes. Moreover, on the basis of the definition of a novel centrality measurement, we propose two layout schemes based on super wires and super nodes for enhancing network invulnerability. The simulation analysis indicates that the proposed schemes are able to enhance the invulnerability of the network with low network construction costs.

Wireless Sensor Networks (WSNs) are networks of cooperating tiny wireless devices having computational, communication and sensing capabilities. To effectively program WSNs, specific high-level frameworks exist that make transparent the low-level mechanisms of sensor platforms. However, deployment of applications developed by such frameworks onto real WSNs is a challenge, specifically as some design and implementation choices can be carefully tuned only after application testing, which is well-known to be non practicable atop WSNs. This paper proposes a simulation-based approach for designing WSN applications based on the Building Management Framework (BMF). A BMF-based design can be evaluated on the Castalia simulator before its deployment. Such an evaluation provides important performance measurements such as data delivery ratio and energy consumption with respect to the scale of the network, the duty cycle and the BMF operating modes. Simulation results can allow to finely tune the WSN application parameters for an optimal deployment and execution.

Opportunistic network as a representative network evolved from social networks and ad hoc networks, has been on cutting edges in recent years. Due to its inherent characteristics serving for intermittent networking setting specifically, the opportunistic network has been also widely applied in the domain of Internet of Things (IoT). Many researchers have focused on the realistic mobility model and cost-effective routing scheme. Community as one of the most inherent attributes of the opportunistic network has been proved to be much helpful in simulating mobility traces of human society and selecting suitable message forwarders. This paper proposes a community-structured mobility model with consideration of geographical location preference and time-variance in human behavior patterns. Based on this model, a novel routing algorithm is presented by jointly considering utilities generated by social degree and relation. The results show that our routing scheme is able to improve success rate while control the routing cost and transmission delay into a reasonable range.

Wireless sensor networks have been widely used in many different applications and in the future they will play an increasingly important role. Since these networks have no fixed infrastructure and are usually distributed over large areas, the use of routing protocols is indispensable. However, when the number of nodes within an area increases, the communication interferences and collisions increase significantly, thus reducing the network performance. In this paper, we first introduce a new measurable quantity, the "node concentration", in contrast to the standard network density. Then, the performance of the AODV (Ad-hoc On-demand Distance Vector) routing protocol is evaluated with respect to the variation in node concentration. Finally, we propose an enhancement of AODV, called CG-AODV, by introducing a "node concentration-driven gossiping" approach for limiting the flooding of control packets. The simulation results demonstrate that CG-AODV provides significant improvements in terms of packet delivery ratio and path discovery delay.

Wireless body sensor networks (BSNs) possess enormous potential for changing people's daily lives. They can enhance many human-centered application domains such as m-Health, sport and wellness, and human-centered applications that involve physical/virtual social interactions. However, there are still challenging issues that limit their wide diffusion in real life: primarily, the programming complexity of these systems, due to the lack of high-level software abstractions, and the hardware constraints of wearable devices. In contrast with low-level programming and general-purpose middleware, domain-specific frameworks are an emerging programming paradigm designed to fulfill the lack of suitable BSN programming support with proper abstraction layers. This paper analyzes the most important requirements for an effective BSN-specific software framework, enabling efficient signal-processing applications. Specifically, we present signal processing in node environment (SPINE), an open-source programming framework, designed to support rapid and flexible prototyping and management of BSN applications. We describe how SPINE efficiently addresses the identified requirements while providing performance analysis on the most common hardware/software sensor platforms. We also report a few high-impact BSN applications that have been entirely implemented using SPINE to demonstrate practical examples of its effectiveness and flexibility. This development experience has notably led to the definition of a SPINE-based design methodology for BSN applications. Finally, lessons learned from the development of such applications and from feedback received by the SPINE community are discussed.

Body Sensor Networks (BSNs) are conveying notable attention due to their capabilities in supporting humans in their daily life. In particular, real-time and noninvasive monitoring of assisted livings is having great potential in many application domains, such as health care, sport/fitness, e-entertainment, social interaction and e-factory. And the basic as well as crucial feature characterizing such systems is the ability of detecting human actions and behaviors. In this paper, a novel approach for human posture recognition is proposed. Our BSN system relies on an information fusion method based on the D-S Evidence Theory, which is applied on the accelerometer data coming from multiple wearable sensors. Experimental results demonstrate that the developed prototype system is able to achieve a recognition accuracy between 98.5% and 100% for basic postures (standing, sitting, lying, squatting).

Spatially distributed sensor nodes can be used to monitor systems and humans conditions in a wide range of application domains. A network of body sensors in a community of people generates large amounts of contextual data that requires a scalable approach for storage and processing. Cloud computing can provide a powerful, scalable storage and processing infrastructure to perform both online and offline analysis and mining of body sensor data streams. This paper presents BodyCloud, a system architecture based on Cloud Computing for the management and monitoring of body sensor data streams. It incorporates key concepts such as scalability and flexibility of resources, sensor heterogeneity, and the dynamic deployment and management of user and community applications.

The Internet of Things (IoT) term is recently emerging to envision a global infrastructure of networked physical objects. As different definitions of IoT do currently exist, we specifically refer to IoT as a loosely coupled, decentralized system of smart objects (SOs), which are autonomous physical/digital objects augmented with sensing/actuating, processing, and networking capabilities. SOs are able to sense, log, and interpret information generated within themselves and around the neighboring external world where they are situated, act on their own, cooperate with each other, and exchange information with humans. The development of a IoT based on SOs raises many issues involving hw/sw system architecture and application development methodology. A few approaches (e.g. FedNet, UbiComp, Smart Products) have been to date proposed to support the vision of an SO-based IoT infrastructure. In this paper we first discuss the suitability of the agent paradigm and technology to effectively support the development of such an IoT infrastructure and then propose a multi-layered agent-based architecture for the development of proactive, cooperating and context-aware smart objects. Our architecture takes into account a wide variety of smart objects, from reactive to proactive, from small to very large, from stand-alone to social. The implementation phase can be based on multiple agent languages and platforms (JADE, JADEX, LEAP, MAPS) atop heterogeneous computing systems (computers, smartphones, and sensor nodes).

Wireless Body Sensor Networks (WBSNs) have proved to be a suitable technology for supporting the monitoring of physical and physiological activities of the human body. However, avoiding erroneous behavior of WBSN-based systems is an issue of fundamental importance, especially for critical health-care applications. In this regard, proper self-healing techniques should be able to fulfill requirements such as fault tolerance and reliability by detecting, and possibly recovering, faults and errors at runtime. In this paper, we focus on data faults, by first studying the impact of corrupted data, affecting sensed data by different kind of data-fault models, on the accuracy of a human activity recognition system. Then, we describe how the SPINE-* framework is able to enhance the WBSN system by adding instrumental autonomic elements providing the necessary self-healing operations. We find that the use of autonomic elements makes the system much more efficient and reliable thanks to its improved tolerance to data faults, as demonstrated by experimental results.

In this paper we propose reference architectures and SPINE-based middleware for Collaborative Body Sensor Networks (CBSNs) that can enable new smart wearable systems in the context of physical pervasive computing environments. CBSNs are wireless BSNs that are able to cooperate to support a common goal. Cooperation is therefore based on interaction among the CBSNs and distributed computation across the interacting CBSNs. In particular, interaction can be activated when CBSNs are in proximity and based on service-specific protocols that allow for service management between the involved CBSNs. Specifically, the paper presents C-SPINE, an enhancement of the SPINE middleware for CBSN applications. Finally, a collaborative emotion detection system, integrating heart rate sensing with handshake detection, is developed through C-SPINE and experimentally analyzed.

Wireless sensor networks (WSNs) are emerging as powerful platforms for distributed embedded computing supporting a variety of high-impact applications. However, programming WSN applications is a complex task that requires suitable paradigms and technologies capable of supporting the specific characteristics of such networks which uniquely integrate distributed sensing, computation and communication. Mobile agents are a distributed computing paradigm based on code mobility that has already demonstrated high effectiveness and efficiency in IP-based highly dynamic distributed environments. Due to their intrinsic characteristics, mobile agents may provide more benefits in the context of WSNs than in conventional distributed environments. In this paper we present the design, implementation and experimentation of MAPS (Mobile Agent Platform for Sun SPOT), an innovative Java-based framework for wireless sensor networks based on Sun SPOT technology which enables agent-oriented programming of WSN applications. The MAPS architecture is based on components that interact through events. Each component offers a minimal set of services to mobile agents that are modeled as multi-plane state machines driven by ECA rules. In particular, the offered services include message transmission, agent creation, agent cloning, agent migration, timer handling and easy access to the sensor node resources (sensors, actuators, input switches, flash memory and battery). Agent programming with MAPS is presented through both a simple example related to mobile agent-based monitoring of a sensor node and a more complex case study for real-time human activity monitoring based on wireless body sensor networks. Moreover, a performance evaluation of MAPS carried out by computing micro-benchmarks, related to agent communication, creation and migration, is illustrated.

Abstract Wireless body sensor networks (WBSNs) enable a broad range of applications for continuous and real‐time health monitoring and medical assistance. Programming WBSN applications is a complex task especially due to the limitation of resources of typical hardware platforms and to the lack of suitable software abstractions. In this paper, SPINE (signal processing in‐node environment), a domain‐specific framework for rapid prototyping of WBSN applications, which is lightweight and flexible enough to be easily customized to fit particular application‐specific needs, is presented. The architecture of SPINE has two main components: one implemented on the node coordinating the WBSN and one on the nodes with sensors. The former is based on a Java application, which allows to configure and manage the network and implements the classification functions that are too heavy to be implemented on the sensor nodes. The latter supports sensing, computing and data transmission operations through a set of libraries, protocols and utility functions that are currently implemented for TinyOS platforms. SPINE allows evaluating different architectural choices and deciding how to distribute signal processing and classification functions over the nodes of the network. Finally, this paper describes an activity monitoring application and presents the benefits of using the SPINE framework. Copyright © 2010 John Wiley & Sons, Ltd.

Employment of BSN-based technologies in real world scenarios requires a flexible infrastructure at both hardware and software level. In this paper, we emphasize how the use of SPINE (Signal Processing In-Node Environment), a software framework for BSN, supports the development of heterogeneous health-care applications based on reusable subsystems. One of the main goal of SPINE is to provide a flexible architecture that can support variety of practical applications without the need for costly redeployment of the code running on sensor nodes. We also present a SPINE sensor node emulator that supports the first phase of the algorithm design, when the actual hardware devices may not be available. This approach can guide the choice of the required hardware (e.g. the sensors) to meet the application requirements based on the results obtained in the emulated environment. Such tool can simplify the research collaboration during the specification stage of a project, due to availability of a common (virtual) architecture.

The handshake gesture is an important part of the social etiquette in many cultures. It lies at the core of many human interactions, either in formal or informal settings: exchanging greetings, offering congratulations, and finalizing a deal are all activities that typically either start or finish with a handshake. The automated detection of a handshake can enable wide range of pervasive computing scanarios; in particular, different types of information can be exchanged and processed among the handshaking persons, depending on the physical/logical contexts where they are located and on their mutual acquaintance. This paper proposes a novel handshake detection system based on body sensor networks consisting of a resource-constrained wrist-wearable sensor node and a more capable base station. The system uses an effective collaboration technique among body sensor networks of the handshaking persons which minimizes errors associated with the application of classification algorithms and improves the overall accuracy in terms of the number of false positives and false negatives.

Rapid development of wireless body sensor network (WBSN) applications can be enabled by suitable domain-specific frameworks which are usually organized in two parts: a base-station-side (or coordinator) and a sensor-node-side. While the former can be based on the Java language so being highly portable, the latter is usually highly dependent on the exploited sensor platform. Available state of the art frameworks follow such an organization and, in particular, the current version of SPINE is based on TinyOS and can be only used to effectively develop collaborative WBSN applications for TinyOS-based sensor platforms. To develop SPINE-based applications for new sensor platforms, the SPINE framework should be re-implemented for each new sensor platform to be exploited. This not only increases development efforts but also enforces SPINE-oriented developers to become skilled on the low-level programming abstractions provided by a new employed sensor platform. In this paper we discuss issues related to platform-independent development of collaborative WBSN applications and, specifically, describe the requirements, architecture and first implementation experiences of SPINE2 which aims at reaching a very high platform independency and raising the level of the used programming abstractions by providing a task-oriented programming model. The paper also discusses how such a task-oriented model enables dynamic task assignment and holistic collaborative task execution also for resource-constrained environments such as tiny sensor nodes.

SPINE (signal processing in node environment) is a framework for the development of body sensor network (BSN) applications. It provides developers of signal processing algorithms with APIs and libraries of protocols, utilities and data processing functions. Hence, it offers application developers new abstractions that improve interoperability and allow to reduce development time. This paper presents the architecture and the capabilities of the SPINE framework, and shows its use in the development of a real-time activity monitoring system prototype.

This work proposes the modeling and the analysis through simulation of a multi-policy, application-level multicast protocol for the cooperative control of media streams transmitted by a multimedia server. The proposed protocol basically relies on a lower-level protocol incorporating a multicast-based coordination mechanism that reduces competition among clients for accessing a shared resource. Performance evaluation of the lower-level protocol was carried out on multicast control trees using a discrete-event simulator. Simulation results showed that the lower-level protocol provides higher performances than random floor control protocols which do not exploit coordination among clients.

Distributed platforms for live and on-demand media streaming delivery such as content distribution networks and media on-demand systems, are being diffused mainly due to the widespread availability of IP-based, bandwidth-capable digital networks. Provision of multimedia group services is usually supported by transmitting media streams to subscribers organized in a multicast group. Although multicast streaming saves bandwidth and improves scalability, it is prone to be hacked. This paper proposes an efficient technique centered on the Blowfish symmetric encryption algorithm for securing media streams based on the real-time transport protocol (RTP). The developed technique along with an ad-hoc key distribution mechanism is seamlessly embedded into our Java-based cooperative playback system - ViCRC/sup C/, which allows multicast transmission on-demand of archived multimedia sessions to a cooperative group of clients.

This paper proposes the modeling and the performance analysis of a high-level control protocol - COCOP, which enables cooperative groups of clients to control a shared server delivering time-dependant data services. Several synchronous multimedia systems such as media on-demand, web casting, and networks of real/virtual sensors, can beneficially exploit COCOP to furnish cooperative control sessions. The protocol can be mapped onto a multicast transport support based either on IP-multicast or on an application level multicast infrastructure. An event-driven simulation framework is purposely customized and exploited to analyze the protocol performance and thus the dynamics of a cooperative control session over simple, yet representative topologies.

This paper presents the design and the implementation of the Virtual Video Gallery (VVG), an advanced distributed video on-demand system accessible through a dynamic virtual world, which mimes an art gallery where movie posters are exhibited The design phase is driven by object-oriented modeling techniques which depend on UML and its extensions purposely suited to model multimedia information systems. The implementation centers on an approach blending (i) Java, which provides powerful computing, multimedia and networking capabilities, (ii) VRML, which supports an easy construction of complex virtual worlds, and (iii) WWW facilities.

The paper describes a methodology for the development of real time systems and shows its application to the modeling, analysis and implementation of distributed multimedia systems. The methodology is centered on Communicating Real-Time State Machines as the modeling language and is supported by jCRSM, a Java2 based toolset. The latter provides a graphical environment for editing, testing, debugging and Java code generation of a prototyped system. Multimedia systems are particular real time systems which normally do not have hard deadlines to fulfill but only soft deadlines concerning the achievement of a user-defined level of quality of service. For instance, timing QoS parameters refer to jitter, skew and end-to-end delay, which are to be kept bounded throughout a multimedia session. QoS constraints are monitored by assertions on the recorded timestamped event histories. The paper reports some experimental results of a modeled remote multimedia presentation system.