As touchless interaction becomes increasingly important in wearable and ambient computing, gesture-based air handwriting offers a promising input modality, particularly for low-power embedded devices. While vision-based and radar-based systems have achieved high accuracy in gesture recognition, they are often unsuitable for deployment on microcontrollers due to their computational and energy demands. In contrast, IMU-based systems provide a lightweight and privacy-preserving alternative, yet existing research rarely addresses full alphabet recognition or deployment-ready pipelines for resource-constrained environments. This paper proposes a complete TinyML pipeline for inertial-based air handwriting recognition of English alphabets, integrating structured preprocessing of raw IMU data into 2D rasterized gesture images, followed by training and deployment of four lightweight deep learning models: SqueezeNet, EfficientNet-Lite0, ShuffleNetV2, and FastKAN. The models are evaluated under a unified training configuration and subjected to compression techniques including quantization, pruning, and knowledge distillation. Among them, FastKAN demonstrates significant superiority, achieving a test accuracy of 97.4% with a minimal model size of 120 KB and energy consumption as low as 0.0011J per inference after hybrid compression. This work explicitly targets isolated characters (A–Z, a–z); continuous handwriting and word-level recognition are out of scope and left for future work. Extensive evaluations, including confusion matrix analysis, compression benchmarking, and successful deployment on an Arduino Nano 33 BLE Sense, demonstrate the practicality, efficiency, and robustness of the proposed system for real-time TinyML-based handwriting recognition applications.

IoT network security faces dual challenges: effectively detecting intrusions while safeguarding user privacy. This paper introduces HFEL (Hybrid Federated Ensemble Learning), an innovative approach integrating tree-based algorithms with neural networks in a privacy-preserving federated architecture. By combining Random Forest and XGBoost with a Federated Multi-Layer Perceptron, our framework maintains data locality while enhancing detection capabilities through ensemble techniques. HFEL uniquely addresses the performance initialization issues common in federated learning systems through its hybrid design. Experimental validation on BoT-IoT and Edge-IIoT datasets demonstrates HFEL's exceptional performance, achieving detection accuracies of 99.94 % and 97.53 % respectively. Comparative analysis reveals substantial improvements over both traditional federated approaches and current state-of-the-art methods. This research advances the field by demonstrating how ensemble strategies can overcome key limitations in federated intrusion detection without compromising privacy constraints.

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

Air handwriting recognition has emerged as a promising solution for touchless human-computer interaction, particularly in the context of Internet of Things (IoT) systems and wearable devices, where traditional input modalities are often infeasible. However, despite extensive research on Latin, Arabic, and Chinese scripts, no prior work has explored real-time air-written recognition of Tifinagh characters a historically and culturally significant script used by Amazigh communities in North Africa. To address this gap, we present the first end-to-end air handwriting recognition framework for the Tifinagh alphabet, designed specifically for constrained IoT environments. At the core of our system is XTiny-FastKAN, a novel, interpretable TinyML model based on a fast variant of the Kolmogorov–Arnold Network (KAN), optimized for ultra-low-latency inference and minimal memory consumption. The system captures inertial motion signals using a consumer-grade IMU, applies a rasterization-based preprocessing pipeline, and uses temporal saliency mapping for explainable predictions. Our quantized model achieves a recognition accuracy of 96.6%, with a memory footprint of just 35 KB and an inference time of 0.04 ms, enabling real-time execution on microcontroller-class IoT hardware. This work not only fills a critical gap in the digitization of underrepresented languages but also contributes a deployable, energy-efficient, and explainable edge AI solution aligned with the growing demands of TinyML in IoT ecosystems. Our codes and dataset are available at https://github.com/Ism-ail11/XTiny-FastKAN.

Cloud Computing (CC) and virtualization concepts are two advanced technologies introduced to empower distance and blended learning. Besides, they play a crucial role in equipping learners with practical skills and fostering hands-on experience to defend against cyber-attacks. Many Higher Education Institutions (HEIs) in developed countries have already embraced the promise of CC to raise educational standards. However, the pace of its adoption in developing countries has stagnated. Moreover, existing solutions in the literature are not sustainable. They either rely on on-premise infrastructure or are bound by a single cloud service provider. Consequently, they are likely prone to failures and a sudden outage. To fill this gap, this paper is a first that comprehensively addresses the above issues and introduces a federated hybrid CC system based on an extension of Apache Virtual Computing Lab (VCL). The proposed system provides an independent open-source implementation, greater configuration flexibility, and methodological improvements as compared to existing studies in the literature. In addition, it promotes the sustainability of the CC services, extensible cloud architecture, and fault tolerance. VCL is primarily focused on provisioning Virtual Laboratories (VL) for remote cybersecurity and computer networks education, with potential expansion to other domains of engineering education. In addition, this paper introduces GPT-TerminalPro, a terminal-based tool driven by OpenAI’s Generative Pretrained Transformer (GPT-3.5) that provides intelligent assistance to users while performing lab tasks. To experimentally evaluate the VCL’s performance, the standard Linux tools as well as the Apache benchmark and httperf HTTP load generators are utilized. VCL has been tested with 30 users and 61 virtual user computing environments provisioning to validate the overall performance. The results are fascinating: the provisioning time including all VCL background tasks is always less than a minute and utilizes fewer computing resources while providing a better user experience. This paper will encourage the adoption of CC in low-income countries.

The rapid expansion of the Internet of Things (IoT) gives Intruders a wide attack surface from which they can conduct more damaging cyber-attacks. Scan, Spying, Denial of Service, Data Type Probing, Malicious Control, and Malicious Operation are such attacks and anomalies that can bring down an IoT system, which makes Attack and anomaly detection in IoT a rising concern and creating a powerful Intrusion Detection System (IDS) a primary need. The main goal of an intrusion detection system (IDS) is detecting attacks and any attempt to break down networks. Machine learning techniques have recently been used in intrusion detection systems since they have shown the ability to learn and adapt, besides providing a quick response. This work proposes an intrusion detection framework that can classify network activities as “Normal” or “Attack” using various machine learning methods. A common dataset called BoT-IoT evaluated the suggested model using KNIME analytics Platform.

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

Virtual laboratories are fully interactive simulations in which students conduct experiments, collect data and answer questions to assess their understanding. The primary objective of virtual cloud-based lab solutions is to provide customized IT environments for users. In today's information, networks and security training, time and location are one of the crucial elements for participants. Many universities and organizations have already taken some advancements by providing online training and labs that are accessible from anywhere in the world. In this paper, we conduct a small-scale trial of VCL solution to support higher education in networks and security courses of some hard concept such as the security or networks tests and experimentations. We describe and analyze in more detail the performance of this solution using load testing.

The main objective of the work is to enhance and optimize the performance of the Datagram Transport Layer Security (DTLS) network for Constrained Application CoAP in Internet of Things. So two mechanisms were designed which lead to the higher performance and the efficiency of the protocol DTLS. Our contribution in this paper is to reduce the cost of communication of the DTLS protocol and improve the weakness of cookies exchange in the handshake process in order to counter DoS attacks. The proposed enhanced DTLS protocol is integrated inside the Constrained Application Protocol (CoAP) to reduce the cost in terms of messages and size taken by the security layer in each message. The proposed protocol is performed on Contiki operating system for the Internet of Things, and compared with original DTLS for CoAP. The simulation results lead to better performance of the proposed protocol in terms of packet overhead, handshake time processing, and energy consumption.