Search Results (70)

This study addresses the limitations of conventional filtering methods in handling irregular outliers and missing observations, which can compromise filter robustness and accuracy. We propose the Transformer-based Outlier-Robust Kalman Filter (TORKF), a hybrid data and knowledge hybrid-driven framework for stochastic discrete-time systems. Initially, this study derives the filtering formulas applicable when outliers exist in observation vectors and, based on these formulations, proposes a novel method capable of accurately identifying observation vectors containing outliers. In addition, a transformer-based prediction compensation approach is employed to compute the prediction vector compensation value in scenarios involving outliers. This method utilizes a specially designed data structure to ensure the transformer encoder fully extracts the input features. Furthermore, to address outlier-induced inaccuracy in prediction error covariance, a compensation method aggregating all prediction outcomes is proposed, leading to enhanced filtering accuracy. Aircraft tracking presents challenges from complex motion models and outlier-prone observations, making it an ideal testbed for robust filtering algorithms. TORKF demonstrates superior performance, with a 12.7% lower RMSE than state-of-the-art methods across both propeller and jet datasets, while maintaining sub-90 ms single-frame processing to meet real-time requirements. Ablation studies confirm that all three proposed methods enhance accuracy and demonstrate synergistic improvements. Full article

The rapid expansion of Edge and Industrial Internet of Things (IIoT) systems has intensified the risk and complexity of cyberattacks. Detecting advanced intrusions in these heterogeneous and high-dimensional environments remains challenging. As the IIoT becomes integral to critical infrastructure, ensuring security is crucial to prevent disruptions and data breaches. Traditional IDS approaches often fall short against evolving threats, highlighting the need for intelligent and adaptive solutions. While deep learning (DL) offers strong capabilities for pattern recognition, single-model architectures often lack robustness. Thus, hybrid and optimized DL models are increasingly necessary to improve detection performance and address data imbalance and noise. In this study, we propose an optimized hybrid DL framework that combines a transformer, generative adversarial network (GAN), and autoencoder (AE) components, referred to as Transformer–GAN–AE, for robust intrusion detection in Edge and IIoT environments. To enhance the training and convergence of the GAN component, we integrate an improved chimp optimization algorithm (IChOA) for hyperparameter tuning and feature refinement. The proposed method is evaluated using three recent and comprehensive benchmark datasets, WUSTL-IIoT-2021, EdgeIIoTset, and TON_IoT, widely recognized as standard testbeds for IIoT intrusion detection research. Extensive experiments are conducted to assess the model’s performance compared to several state-of-the-art techniques, including standard GAN, convolutional neural network (CNN), deep belief network (DBN), time-series transformer (TST), bidirectional encoder representations from transformers (BERT), and extreme gradient boosting (XGBoost). Evaluation metrics include accuracy, recall, AUC, and run time. Results demonstrate that the proposed Transformer–GAN–AE framework outperforms all baseline methods, achieving a best accuracy of 98.92%, along with superior recall and AUC values. The integration of IChOA enhances GAN stability and accelerates training by optimizing hyperparameters. Together with the transformer for temporal feature extraction and the AE for denoising, the hybrid architecture effectively addresses complex, imbalanced intrusion data. The proposed optimized Transformer–GAN–AE model demonstrates high accuracy and robustness, offering a scalable solution for real-world Edge and IIoT intrusion detection. Full article

A secure and energy-efficient network architecture is essential due to the rapid proliferation of Internet of Things (IoT) devices in critical sectors such as healthcare, smart cities, and industrial automation. In this paper, we propose a secure and energy-efficient cross-layer IoT architecture. The security features and energy-saving techniques across various open-system interconnected protocol layers are incorporated in the proposed architecture. The improved security and energy efficiency are achieved using the lightweight Speck and Present ciphers, as well as adaptive communication strategies. The system performance is evaluated by testbeds and extensive simulation experiments using Cooja (Contiki operating systems) and NS-3. The simulation results obtained show a 95% attack mitigation effectiveness, 30% reduction in energy usage, and 95% packet delivery ratio. In a 20-node network scenario, Speck uses 5.2% less radio power than the Advanced Encryption Standard (AES), making it the best tradeoff among the investigated encryption techniques. The NS-3 simulation results confirm that lightweight encryption, such as Present and Speck, uses much less power than the traditional AES, which makes them more appropriate for IoT contexts with limited energy. The scalability across various IoT contexts is ensured through a hybrid assessment approach that combines hardware testbeds and simulation for system validation. Our research findings highlight opportunities for advancing IoT systems toward secure and energy-efficient smart ecosystems. Full article

This work presents the initial experimentation of a real-time 5G mmWave downlink positioning testbed deployed at Airbus premises. This experimentation is part of a first-of-a-kind testbed for hybrid Global Navigation Satellite Systems (GNSS), fifth-generation (5G) new radio (NR) and sensor positioning, called the Hybrid Overlay Positioning with 5G and GNSS (HOP-5G) testbed. The mmWave 5G base station (BS) exploits the 5G standard positioning reference signal (PRS) to support positioning capabilities within the 5G NR downlink transmissions. Outdoor field results are used to characterize the received power levels and beam-based angle-of-arrival (AoA) estimation accuracy of this 5G mmWave PRS platform. The goal is to assess the suitability of this platform to enhance the positioning performance thanks to the 5G downlink mmWave transmissions. To the best of the authors’ knowledge, this paper presents the first AoA results using OpenAirInterface (OAI) PRS mmWave signal transmissions at 27 GHz for positioning. These initial field results indicate a maximum coverage of 30 m and an AoA accuracy limited by the reduced array size. The limitations and potential enhancements of this platform are provided as future recommendations. Full article

Thermal processes with prolonged and variable delays pose considerable difficulties due to unpredictable system dynamics and external disturbances, often resulting in diminished control effectiveness. This work presents a hybrid control strategy that synthesizes deep reinforcement learning (DRL) strategies with nonlinear model predictive control (NMPC) to improve the robust control performance of a thermal process with a long time delay. In this approach, NMPC cost functions are formulated as learning functions to achieve control objectives in terms of thermal tracking and disturbance rejection, while an actor–critic (AC) reinforcement learning agent dynamically adjusts control actions through an adaptive policy based on the exploration and exploitation of real-time data about the thermal process. Unlike conventional NMPC approaches, the proposed framework removes the need for predefined terminal cost tuning and strict constraint formulations during the control execution at runtime, which are typically required to ensure robust stability. To assess performance, a comparative study was conducted evaluating NMPC against AC-based controllers built upon policy gradient algorithms such as the deep deterministic policy gradient (DDPG) and the twin delayed deep deterministic policy gradient (TD3). The proposed method was experimentally validated using a temperature control laboratory (TCLab) testbed featuring long and varying delays. Results demonstrate that while the NMPC–AC hybrid approach maintains tracking control performance comparable to NMPC, the proposed technique acquires adaptability while tracking and further strengthens robustness in the presence of uncertainties and disturbances under dynamic system conditions. These findings highlight the benefits of integrating DRL with NMPC to enhance reliability in thermal process control and optimize resource efficiency in thermal applications. Full article

The increasing demand for sustainable and autonomous monitoring solutions in critical infrastructure has driven interest in Green Internet of Things (G-IoT) systems. This paper presents an analytical and experimental framework for designing energy-efficient, self-sustaining pipeline monitoring systems that leverage renewable energy harvesting and low-power operation techniques. We propose a hybrid approach combining solar energy harvesting with energy-saving strategies such as adaptive sensing, duty cycling, and distributed computing to extend the lifetime of IoT nodes without human intervention. Using real-world irradiance data and energy profiling from a prototype testbed, we analyze the impact of solar panel sizing, energy storage capacity, energy-saving strategies, and energy leakage on the energy balance of IoT nodes. The simulation results show that, with optimal dimensioning, harvested solar energy can sustain pipeline monitoring operations over multi-year periods, even under variable environmental conditions. We investigated the influence of design parameters such as duty cycling, solar panel area, the capacity of the energy storage system, and the energy leakage coefficient on energy performance metrics such as the autonomy or lifetime of the node (time required to drain all the stored energy), which is an important design object. This framework provides practical design insights for the scalable deployment of G-IoT systems in energy-constrained outdoor environments. Full article

Traditional centralized water systems are facing sustainability challenges due to climate and socioeconomic changes, extreme weather events, and aging infrastructure and their uncertainties. The energy sector has addressed similar challenges using the microgrid approach, which involves decentralized energy sources and their supply, improving system resilience and sustainable energy supply. This study investigated the resilience effects of water microgrids, which feature operational interactions between centralized and local systems for sustainable water supply. A lab-scale water distribution model was tested to demonstrate centralized, decentralized, and microgrid water systems under the disruption scenarios of pump shutdown, pump rate manipulation, and pipe leaks/bursts. The water microgrids integrate centralized and local systems’ operations, while the decentralized system operates independently. Then, functionality-based resilience and its attributes were evaluated for each disruption scenario. The results reveal that, overall, the microgrid configuration, with increased water supply redundancy and flexible operational adjustment based on system conditions, showed higher resilience, robustness, and recovery rate and a lower loss rate across disruption scenarios. The resilience effect of water microgrids was more evident with longer and more severe disruptions. Considering global challenges in water security under climate and socioeconomic changes, the findings suggest insights into a hybrid water system as a strategy to enhance resilience and water use efficiency and provide adaptive operations for sustainable water supply. Full article

Long-range wireless area networks (LoRaWAN) typically use a simple star topology. However, some nodes may experience connectivity issues with the gateway due to signal degradation or limited coverage, often resulting from challenging environments in sectors such as agriculture, industry, smart cities, smart grids, and healthcare, where LoRaWAN-based IoT solutions have expanded. The main contribution of this paper is the implementation of a hybrid topology for LoRaWAN networks that remains fully transparent to current spec LoRaWAN servers and IoT applications. It enables the coexistence of mesh (multi-hop) and star (single-hop) communication schemes, dynamically adapting a node’s transmission mode based on physical link quality metrics. Additionally, the user interface allows for customizing network topology and parameters. Experimental proof-of-concept tests were conducted on a campus-wide testbed. Results showed that all devices successfully switched topology mode in 100% of the instances, enabling data transmission across all three scenarios under test. Network performance metrics were evaluated, with latencies ranging from 0.5 to 3.2 s for both single-hop and multi-hop transmissions. Additionally, improvements in RSSI and SNR were observed, validating the efficiency of the proposed solution. These results demonstrate the feasibility and effectiveness of our approach in extending network connectivity to areas beyond the gateway’s coverage. Full article

We report on the first stage of an energy systems integration project to develop hybrid renewable energy generation and storage of hydrogen for subsequent use via research-based low regret system testbeds. This study details the design and construction of a flexible plug-and-play hybrid renewable power and hydrogen system testbed with up to 50 kW capacity aimed at addressing and benchmarking the operational parameters of the system as well as key components when commissioned. The system testbed configuration includes three different solar technologies, three different battery technologies, two different electrolyser technologies, hydrogen storage, and a fuel cell for regenerative renewable power. Design constraints include the current limit of an AC microgrid, regulations for grid-connected inverters, power connection inefficiencies, and regulated hazardous area approval. We identify and show the resolution of systems integration challenges encountered during construction that may benefit planning for the emerging pilot, or testbed, configurations at other sites. These testbed systems offer the opportunity for informed decisions on economic viability for commercial-scale industry applications. Full article

Rolling Bearings are important supporting components of rotating machines in industrial processes; the faults of rolling bearings will cause the deterioration of the operation conditions of rotating machines. How to effectively extract the fault features and identify the fault modes of rolling bearings quickly and accurately has become a key issue for the safe operation of rotating machines. In this paper, a novel hybrid fault diagnosis method of an optimized random forests classifier for rolling bearings is proposed. Firstly, the original vibration signals are decomposed by recursive variational mode decomposition (RVMD), and the typical time–frequency domain features are extracted from decomposed signals at different scales. The Relief-F ranking method is utilized to assess the quality of time–frequency domain features, and the top-ranked features with high weight gain are selected for identifying the fault modes. Then, an improved bee colony algorithm is proposed based on the simulated binary crossover criterion, which is used to optimize the key parameters of random forests (RF). Finally, several experiments are conducted on the Case Western Reserve University bearing dataset and the dataset collected from our rolling bearing fault testbed. The experimental results show that the proposed method can accurately identify bearing faults and outperform other state-of-the-art methods. Full article

Accurate modelling of occupancy patterns is critical for reliable estimation of building stock energy demand, which is a key input for the design of district energy systems. Aiming to investigate the suitability of different occupancy-modelling approaches for the design of district energy systems, the present study examines a set of standard-based schedules (from the UK National Calculation Methodology), a widely used stochastic occupancy model, and a novel energy-data-driven occupancy model. To this end, a dynamic energy model of a higher education office building developed within a stock model of London’s Bloomsbury district serves as a testbed to implement the occupancy models, explore their implications for the estimation of annual and peak heating and cooling demand, and extrapolate the findings to the computationally demanding building stock stimulations. Furthermore, the simulations were conducted in two years before and after the COVID-19 pandemic to examine the implications of hybrid working patterns after the pandemic. From the results, the energy-data-driven model demonstrated superior performance in annual heating demand estimations, with errors of ±2.5% compared to 14% and 7% for the standard-based and stochastic models. For peak heating demand, the models performed rather similarly, with the data-driven model showing 28% error compared to 29.5% for both the standard-based and stochastic models in 2019. In cooling demand estimations, the data-driven model yielded noticeably higher annual cooling demand and lower peak cooling demand estimations as compared with the standard-based and stochastic occupancy models. Given the adopted building-modelling approach, these findings can be extended to district-level investigations and inform the decision on the choice of occupancy models for building stock energy simulation. Full article

Hybrid rockets have very interesting characteristics like simplicity, reliability, safety, thrust modulation, environmental friendliness and lower costs, which make them very attractive for several applications like sounding rockets, small launch vehicles, upper stages, hypersonic test-beds and planetary landers. In recent years, advancements have been made to increase hybrid motor performance, and two of the most promising solutions are vortex injection and paraffin-based fuels. Moreover, both technologies can be also used to tailor the fuel regression rate, in the first case varying the swirl intensity, and in the second case with the amount and type of additives. In this way, it is possible not only to design high-performing hybrid motors, but also to adjust their grain and chamber geometries to different mission requirements, particularly regarding thrust and burning time. In this paper, the knowledge about these two technical solutions and their coupling is extended. Three sets of experimental campaigns were performed in the frame of the Italian Space Agency-sponsored PHAEDRA program. The first one investigated a reference paraffin fuel with axial and standard vortex injection. The second campaign tested vortex injection with low values of swirl numbers down to 0.5 with a conventional plastic fuel, namely polyethylene. Finally, the last campaign tested another, lower regressing, paraffin-based fuel with the same low swirl numbers as the second campaign. Full article

The Internet architecture has recently shifted towards a framework characterized by multiple interconnected cloud sites, all linked via an L3 IP network. With this shift, managing networking controls among multiple cloud sites is becoming a significant operational challenge. In particular, ensuring effective networking control necessitates a deeper understanding of flow-level dynamics for comprehensively monitoring interconnection statuses across multiple sites. In this paper, we first propose an IO Visor-enabled tracing solution for Linux-based boxes to efficiently enable the comprehensive collection of network packet flows across interconnected sites. Next, we apply IP prefix-based flow-level analysis at a centralized location to support the intent-based networking control application. This flow-level analysis involves generating policy-based specific action (i.e., redirect) via SDN controllers for specific source IP prefixes, which are causing unknown or potentially vulnerable flows. Furthermore, we employ an open-source ONOS SDN controller to tackle the challenge of managing the hybrid SDN-IP interconnections. By leveraging intent-based networking control, we effectively apply ONOS intents based on IP routing information and generated a set of forwarding action. We evaluate our proposed solution in an experimental SDN-cloud testbed, demonstrating its effectiveness in real-world scenarios. Overall, through the seamless integration of these monitoring and control approaches, we manage to enhance the adaptability and security of the interconnected cloud sites of the testbed. Full article

Cyber-physical systems (CPS) are vital in automating complex tasks across various sectors, yet they face significant vulnerabilities due to the rising threats of cybersecurity attacks. The recent surge in cyber-attacks on critical infrastructure (CI) and industrial control systems (ICSs), with a 150% increase in 2022 affecting over 150 industrial operations, underscores the urgent need for advanced cybersecurity strategies and education. To meet this requirement, we develop a specialised cyber-physical testbed (CPT) tailored for transportation CI, featuring a simplified yet effective automated level-crossing system. This hybrid CPT serves as a cost-effective, high-fidelity, and safe platform to facilitate cybersecurity education and research. High-fidelity networking and low-cost development are achieved by emulating the essential ICS components using single-board computers (SBC) and open-source solutions. The physical implementation of an automated level-crossing visualised the tangible consequences on real-world systems while emphasising their potential impact. The meticulous selection of sensors enhances the CPT, allowing for the demonstration of analogue transduction attacks on this physical implementation. Incorporating wireless access points into the CPT facilitates multi-user engagement and an infrared remote control streamlines the reinitialization effort and time after an attack. The SBCs overwhelm as traffic surges to 12 Mbps, demonstrating the consequences of denial-of-service attacks. Overall, the design offers a cost-effective, open-source, and modular solution that is simple to maintain, provides ample challenges for users, and supports future expansion. Full article

This paper showcases the integration of the Interfacing Toolbox for Robotic Arms (ITRA) with our newly developed hybrid Visual Servoing (VS) methods to automate the disassembly of electric vehicle batteries, thereby advancing sustainability and fostering a circular economy. ITRA enhances collaboration between industrial robotic arms, server computers, sensors, and actuators, meeting the intricate demands of robotic disassembly, including the essential real-time tracking of components and robotic arms. We demonstrate the effectiveness of our hybrid VS approach, combined with ITRA, in the context of Electric Vehicle (EV) battery disassembly across two robotic testbeds. The first employs a KUKA KR10 robot for precision tasks, while the second utilizes a KUKA KR500 for operations needing higher payload capacity. Conducted in T1 (Manual Reduced Velocity) mode, our experiments underscore a swift communication protocol that links low-level and high-level control systems, thus enabling rapid object detection and tracking. This allows for the efficient completion of disassembly tasks, such as removing the EV battery’s top case in 27 s and disassembling a stack of modules in 32 s. The demonstrated success of our framework highlights its extensive applicability in robotic manufacturing sectors that demand precision and adaptability, including medical robotics, extreme environments, aerospace, and construction. Full article