Low-Power Wide-Area Network (LPWAN) technologies such as LoRa are key enablers of large-scale IoT monitoring systems, where long communication range and low energy consumption are essential. These advantages, however, come at the cost of strict throughput limitations, which significantly shape IoT system design. Compressive Sensing (CS) can mitigate this constraint by reducing transmitted data volumes, effectively trading communication load for additional processing at the sensing node and receiver. From an IoT perspective, this shift impacts node lifetime, hardware requirements, and overall network scalability. In this paper, we propose a data-driven optimization framework for LoRa-based IoT sensing systems employing CS. The approach jointly analyzes reconstruction quality and energy consumption through surrogate regression models that capture the interaction between physical-layer parameters and compression levels. This enables efficient multi-objective optimization via Pareto-front analysis and utopia-based selection. Results show that CS does not always dominate the quality-energy trade-off and that unified and stratified surrogate strategies identify closely aligned optimal operating points. Overall, the framework provides a practical and interpretable tool for the design of energy-efficient IoT sensing deployments.

The rapid advancement and proliferation of modern communication technologies have introduced a wide range of protocols, offering a compelling framework for investigating their compatibility and performance. Within this context, this paper presents a comprehensive study on decision-making approaches for Vertical Handover (VHO) in heterogeneous wireless networks. Specifically, we analyze several Multi-Attribute Decision-Making (MADM) methods, namely SAW, WPM, TOPSIS, and FUZZY, highlighting their strengths, limitations, and the criteria required to achieve the Always Best Connected (ABC) paradigm. Furthermore, we examine the integration of optimization techniques, such as hysteresis, threshold margin, and time-to-trigger, with the aforementioned methods, in order to enhance the efficiency and robustness of the VHO procedure in next-generation wireless environments.

Recently, energy efficient signal processing has emerged as a critical area of research, driven by the increasing demand for cost-effective solutions in modern communication systems, particularly in IoT sensor networks applications. At the same time, AI-based approaches have gained significant popularity, extending their potential far beyond user-driven content creation. In this work, we explore the intersection of these emerging technologies, aiming to enable faster processing, lower power consumption, and reduced hardware complexity. We apply the combined potential of both frameworks to Compressive Sensing (CS) for Structural Health Monitoring (SHM) applications. Specifically, we propose a conceptual sensing and monitoring framework incorporating an adaptive data compression strategy that jointly leverages traditional CS and deep learning-based signal compression. This work represents a feasibility study toward the development of a low-power, low-rate, and cost-effective IoT-based sensor network for SHM capable of adapting to changing channel conditions. To support this framework, the study provides a comprehensive comparison between traditional CS methods and AI-based compression, focusing on compression efficiency and reconstruction capabilities when processing data streams from SHM systems.

Addressing class imbalance is a significant issue in vision-based internet-of-things, especially when datasets have long-tail distributions, and minority classes might represent rare yet critical events, making this problem crucial. This work presents a comprehensive investigation of a weighting and reweighting approach aimed at addressing class imbalance in long-tail object detection tasks. Specifically, this study examines the effect of the weight shift factor employed in the reweighting phase, analyzes the influence of the weighting strategies on the individual loss components of the object detection model, and proposes a dynamically integrated strategy to shift the weights during model training. In addition, this work assesses the robustness and versatility of the proposed weighting and reweighting strategy by evaluating it across various datasets including aerial object detection (VisDrone-DET dataset) and ground-view long-tail detection (COCO-zipf dataset). Experimental findings utilizing the above datasets indicate that modifying the weight shift parameter proficiently regulates the intensity of weight recalculations. The analysis further indicates that the weighting strategy continues to be robust and effective, even with the dataset having a lower number of classes. Experimental results demonstrate consistent improvements across both datasets, including AP50 gains of up to 2.3 percentage points on VisDrone-DET and 9.9 percentage points on COCO-zipf compared with the corresponding baselines. Finally, deployment measurements on the NVIDIA platform indicate the viability of edge inference a potential candidate for deployment in edge analytics frameworks, including NTN-assisted IoT scenarios.

Accurate monitoring of postural stability is essential for preventing falls and supporting rehabilitation, but traditional force-platform-based assessments are costly and impractical for continuous, real-world use. Wearable Inertial Measurement Units (IMUs) offer a low-cost, portable alternative for capturing balance-related motion data in daily life. While deep learning models have shown promise in predicting Center of Pressure (CoP) trajectories from IMU signals, their black-box nature limits clinical trust and adoption due to a lack of interpretability. This paper presents a CNN-BiLSTM-Attention framework for IMU-based CoP trajectory prediction, combined with Integrated Gradients (IG) to interpret the model’s decisions. By quantifying and visualizing the contribution of input features across both time and sensor channels, IG highlights the physiological relevance of the learned patterns. In this way, the proposed approach bridges the gap between predictive performance and clinical trustworthiness, supporting safer and more transparent deployment of wearable-sensor-based balance monitoring in healthcare.

Human Activity Recognition (HAR) in Internet of Medical Things (IoMT) environments is critical for monitoring patient behaviors in hospitals, rehabilitation centers, and eldercare facilities. Wi-Fi Channel State Information (CSI) sensing has emerged as a non-intrusive and infrastructure-friendly alternative to camera-based systems for continuous activity monitoring. While data stream mining algorithms offer the adaptability required for real-time HAR, they remain constrained by a closed-world assumption, where the set of activity classes is fixed in advance. This poses a significant limitation in medical scenarios where patient behaviors naturally evolve from coarse to refined states — for example, a fall incident involves sequential transitions from standing to falling to lying down, potentially forming new, unforeseen activity patterns. Existing concept drift detectors are designed to handle changes in data distribution within known classes but lack the capacity to identify emergent, unseen classes in data streams. To address this gap, we propose LAND (Lightweight Auxiliary Novelty Detection), a distributed federated learning framework integrating auxiliary lightweight novelty detection modules alongside incremental data stream classifiers on Wi-Fi CSI edge devices. LAND enables on-device detection of novel activity patterns, triggers distributed similarity checking across federated nodes, and supports adaptive label evolution without compromising computational efficiency. Experimental evaluations in realistic, hospital‑like simulated environments demonstrate the feasibility and strong prototype‑level performance of LAND in mitigating accuracy degradation during emergent activity episodes, outperforming conventional stream learners in both responsiveness and adaptability. This work advances the development of open-set, lightweight, and adaptive HAR frameworks for sensitive IoMT deployments. The results establish LAND as a robust proof‑of‑concept system, forming a foundation for future clinical deployment studies.

The integration of the Social Internet of Things (SIoT) with wireless sensor networks (WSNs) significantly enhances the efficiency, scalability, and intelligence of distributed sensing systems. However, these networks often encounter severe resource constraints, limiting the control and management traffic that can be introduced. WSNs, typically composed of battery-powered edge sensor nodes with limited computational capabilities, memory, and communication bandwidth, face challenges in optimizing performance. In this work, we address the multiobjective optimization problem within the context of resource-constrained SIoT, aiming to reduce the energy consumption of edge nodes while simultaneously enhancing the quality of the received data based on channel conditions. We propose a Pareto optimization framework to jointly optimize the compression factor and coding rate (CR) in a WSN scenario utilizing Long-Range (LoRa) technology for communication. This framework explores the tradeoffs between energy consumption and data reconstruction quality, leveraging compressive sensing (CS) for efficient data compression to alleviate the transmission load on edge nodes. Furthermore, we present a distributed optimization solution to minimize energy consumption while maximizing data quality, thereby reducing signaling and control overhead. This study contributes to the development of energy-efficient, scalable, and sustainable SIoT systems by providing a foundation for optimizing data transmission in LoRa networks.

Compressive Sensing (CS) has emerged as a modern and alternative signal processing paradigm that challenges the traditional Nyquist-Shannon sampling theorem by enabling accurate signal reconstruction from significantly fewer measurements. This paradigm proves particularly valuable in edge-enabled IoT systems, where devices operate under strict constraints in terms of energy, bandwidth, and computational capacity. In this article, we explore the application of CS in edge-enabled IoT scenarios, with a focus on smart infrastructure monitoring. We examine various system architectures and frameworks that leverage AI-based solutions to optimize resource utilization while maintaining high levels of accuracy. However, the mere integration of CS or AI models does not inherently ensure optimal performance. Achieving tangible benefits requires careful co-optimization of communication and computation strategies at the edge, in order to fully harness the energy-saving potential of these technologies in real-world IoT deployments. To this end, the article presents a range of strategies designed to address this challenge, highlighting the effectiveness of combining CS and AI in edge IoT systems and paving the way for more efficient and intelligent resource management in AI-empowered IoT environments.

The convergence of Symbiotic Internet of Things (IoT), artificial intelligence (AI) foundational models, and 6G is ushering in a new era of intelligent connectivity, where networks, devices, and algorithms operate in close coordination to enable real-time, adaptive, and efficient systems. In the context of structural health monitoring (SHM), this integrated vision provides a powerful framework to tackle challenges, such as limited resources, harsh environments, and the need for timely, high-fidelity data. By enabling intelligent, collaborative processing across edge and network layers, it supports efficient data compression, transmission, and decision-making-ensuring robust and adaptive monitoring even in complex structural settings. In this work, we investigate the use of AI for data compression in an IoT-based SHM scenario. Specifically, we evaluate and compare the performance of four different convolutional autoencoder (convolutional AEs (CAEs)) in compressing and reconstructing inertial signals collected from various structural systems, aiming to enable adaptive and context-aware processing directly at the edge. By testing across heterogeneous sources, we assess the generalizability and robustness of each CAE model, providing insights into the potential of deep learning-based compression techniques for SHM applications.

This paper explores the complex dynamics of Wi-Fi probe request frames, focusing on their transmission behaviors and the implications for user privacy. As mobile devices become increasingly sophisticated, their capacity for wireless communication presents both opportunities for enhanced connectivity and privacy-related challenges. This study examines critical features of probe request frames and their role in device fingerprinting. We highlight the challenges and advancements in communicating between mobile devices and nearby networks by analyzing the randomization policies employed by various operating systems and their effectiveness in masking device identities. Through comprehensive experiments conducted across multiple mobile devices and operating system versions, we capture and analyze probe request frames under different user-device interaction phases. Our findings indicate variations in transmission frequency and the impact of connection states on broadcasting characteristics. We also investigate the deployed randomization policies in different brands of mobile devices, uncovering their influence on privacy preservation. This research provides notable insights for developing more effective wireless communication approaches, contributing to the broader goal of enhancing user privacy in this field.

The Widespread adoption of randomization in modern operating systems has introduced significant challenges for passive monitoring and user tracking in wireless environments. These challenges are further increased in large-scale environments covered by multiple access points, where associating transmission across zones becomes more complicated. This paper presents a frame association-based approach that enables cross-zone tracking and trajectory reconstruction using Wi-Fi probe request frames. The proposed method correlates transmissions across multiple access points by analyzing a combination of fingerprints. Results show that our approach effectively associates frames transmitted from the same origin, tracks devices across multiple zones, and provides insights into user movement and behavior, such as the type of transitions between zones and the reconstructed trajectory, all while preserving user privacy.

The healthcare landscape is undergoing a transformative shift driven by the rapid advancements of sixth-generation (6G) cellular networks and Internet of Things (IoT) technologies. This convergence has led to the emergence of medical digital twins (MDTs) — innovative virtual replicas of physical entities or processes. MDTs enable safe monitoring and testing of therapeutic decisions, resulting in substantial cost savings and enhanced decision-making capabilities for physicians. Furthermore, they provide real-time assistance and facilitate comprehensive data analysis, revolutionizing patient care. However, their full potential is constrained by the intricate nature of human life cycles. Key factors include the choice of data types and the communication protocols used to connect physical and digital twins. For individuals with disabilities, integrating wearable devices with edge computing and WiFi-based Channel State Information (CSI) can effectively monitor health metrics and recognize activities. Although the integration of 6G and IoT technologies holds the promise of enabling seamless communication between physical and digital twins, this paper, which is developed in the framework of the HIPPOCRATES project, focuses on a currently implementable architecture for MDTs based on the Message Queuing Telemetry Transport (MQTT) protocol. The proposed system integrates CSI to augment health monitoring and intervention, particularly for individuals with disabilities.

The Internet of Things (IoT) has garnered significant attention in recent years, with the integration of AI solutions and wireless sensing technologies enabling innovative approaches to context awareness and user location. Additionally, Channel State Information (CSI) from WiFi channels is emerging as a key component in next-generation wireless systems. In this work, we conduct a comprehensive analysis of the sensing capabilities of state-of-the-art CSI tools, namely the Intel 5300 and the ESP32 CSI tools, through extensive experimental tests in a dedicated testbed. The results offer valuable insights into CSI-based techniques, demonstrating their strong potential for activity detection and context aware applications.

In recent years, the increase in the elderly population has placed significant burdens on post-rehabilitation schemes, resulting in high logistical costs and considerable social impacts due to hospitalization or frequent visits. These challenges call for a transformation in the traditional approach to physical patient care, which can be achieved by leveraging the Internet of Medical Things (IoMT), particularly through the use of pervasive wearable sensors. When attached to patients during treatment or therapy, these sensors can provide valuable supplementary information to healthcare professionals. When it comes to adopting IoMT technologies, cost efficiency, portability, and generalization are key factors. Specifically, this study aims to enhance the cost-effectiveness and versatility of wearable eHealth monitoring architectures that utilize foot pressure-sensing hardware for the motor assessment of post-stroke and neurologically impaired patients. It leverages lower limb inertial measurement unit sensory information and machine learning to mitigate the reliance on foot pressure-sensing hardware. We demonstrate the potential of artificial intelligence (AI) in predicting fine-scale foot pressure using only inexpensive, off-the-shelf motion sensors. We propose a self-supervised, exercise-agnostic asynchronous foot pressure decoding model that does not require human annotation. The algorithm is thoroughly evaluated using appropriate performance metrics, and our experimental tests show promising results.

The healthcare sector is experiencing a profound transformation, fueled by the rapid evolution of sixth-generation (6G) cellular networks and Internet of Things (IoT) technologies. At the heart of this shift lies the concept of medical digital twins (MDTs), which serve as dynamic virtual representations of physical systems or biological processes. MDTs offer a secure environment to simulate and evaluate therapeutic strategies, leading to reduced costs and more informed clinical decision-making. They also enable real-time support and in-depth data analysis, setting new standards for patient care. Nonetheless, realizing the full capabilities of MDTs remains challenging due to the inherent complexity of human life cycles. Crucial aspects include selecting appropriate data sources and defining robust communication protocols between the physical and digital realms. In particular, integrating wearable technologies with edge computing and WiFi-based Channel State Information (CSI) can significantly enhance health monitoring and activity recognition for elderly individuals within indoor settings. The synergy of IoT advancements and 6G networks paves the way for improved data exchange and continuous synchronization between digital and physical counterparts. This paper, part of the HIPPOCRATES project, presents an IoT-driven architecture for MDTs that incorporates wearable sensors and CSI data to strengthen health monitoring and early intervention strategies, with a focus on elderly care.

Drones are integral to various applications, out of which traffic surveillance is an important application. However, their operational efficiency is limited by battery life, which restricts their capacity for extended critical missions. Additionally, in remote or high-interference areas, the bandwidth for drone communication is often limited, leading to a decrease in the quality of images transmitted to the base station. This paper aims to address such challenges by having drones transmit video data in real-time at lower resolutions for traffic monitoring. This approach conserves energy and optimizes transmission. However, it adversely affects object detection accuracy at the base station due to compromised data quality. To address this issue, we incorporate Generative Adversarial Networks (GANs) to improve LR images, restoring their quality for precise object detection. Results indicate that the accuracy of traffic analytics achieved with GAN-enhanced images is comparable to that obtained with high-resolution data transmission. Consequently, our approach allows a fundamental trade-off among drone energy consumption, transmission time, flight time, and object detection accuracy, enabling robust detection performance while conserving energy and enhancing operational capabilities.

The use of multiple access protocols information in Internet of Things (IoT) environments has gained significant interest over the past decade, particularly for crowd behavior monitoring. Due to its high data rates and low infrastructure requirements, WiFi is considered one of the most promising wireless technologies for leveraging the explosive growth of transmitted data from mobile devices. However, with the introduction of MAC address randomization and the application of new randomization policies on assigning randomized sequence numbers (SNs) to transmitted frames in mobile devices to enhance privacy, traditional approaches to device identification face significant challenges. To tackle this issue, we conduct a comprehensive analysis at the multiple access level and propose CrowdWatch which is an innovative framework designed to enhance the understanding and utilization of MAC randomization dynamics. Additionally, we introduce a novel approach that leverages multiple device-specific features to accurately associate frames with randomized MAC addresses to their corresponding devices. By integrating multimodal fingerprints, the framework can effectively identify mobile devices and track their behavior. The presented approach ensures reliable detection even under the latest randomization policies. We examined the introduced framework through real-world experiments, and the findings prove that it is an effective solution for smart building management and occupancy estimation in dynamic environments.

The integration of Artificial Intelligence (AI) with the Internet of Medical Things (IoMT) has revolutionized healthcare by enhancing diagnostic accuracy, treatment efficiency, and patient monitoring capabilities. This article explores the global architecture of AI-enabled IoMT platforms, emphasizing their foundational frameworks and operational dynamics. Representative case studies across various medical domains illustrate the practical applications and benefits of these platforms in real-world scenarios. Results from extensive testing and validation of these use cases underscore their efficacy and reliability in clinical settings, paving the way for more sophisticated and effective solutions. Furthermore, the article discusses prevalent challenges that impact the widespread adoption of AI-driven IoMT systems. Insights gleaned from this comprehensive analysis provide valuable guidance for researchers, healthcare practitioners, and policymakers navigating the evolving landscape of AI in medical technology.

Integrated sensing and communication (ISAC) enables several emerging applications while suffering severe security issues. In this article, the novel illegal sensing suppression (ISS) is proposed to protect the privacy of the specific legitimate user in the presence of an adversary with the sensing capacity in the ISAC system. The target detection and the direction of arrival estimation as common sensing tasks are investigated to deteriorate the sensing performance at the adversary side. To this end, we propose the missed detection probability maximization and the Cram & eacute;r-Rao bound (CRB) maximization-based ISS model. This model also ensures the achievable rate and the legitimate sensing CRB toward the adversary at the ISAC base station side. The beamforming optimization algorithms are developed based on the alternating optimization and semidefinite relaxation techniques in order to address the nonconvex issue in the formulated ISS models and explore the rate-ISS tradeoff in the ISAC system. We further propose the closed-form solution for the single user case in the target detection suppression scenario. Numerical results demonstrate the necessity of the ISS and the ISS gain of proposed algorithms compared with other baseline schemes.

Enabled by the integration of AI and the Internet of Medical Things (IoMT), smart and remote monitoring systems are poised to play a pivotal role in the future of healthcare. Specifically, monitoring respiration is a critical component of this evolution, offering a straightforward yet effective method for assessing an individual's health status. In this paper, we introduce an innovative contactless approach to respiration monitoring that leverages Channel State Information (CSI) from Wi-Fi channels. By analyzing variations in the amplitude and phase of CSI data, we infer human respiration patterns. To validate our method, we conducted experiments with multiple subjects in indoor environments, assessing the system's capability to track the respiration cycle and determine breathing frequency. We also compared the performance of various AI algorithms in identifying accurate breath rates. Our findings indicate that the CSI-based system is a promising solution for respiration monitoring, achieving an average accuracy of approximately 84 % in estimating breathing frequency, thus paving the way for future studies to enhance the robustness of the proposed approach.