IET Wireless Sensor Systems最新文献

Underwater Wireless Sensor Networks (UWSNs) are pivotal for ocean monitoring, exploration, and surveillance, comprising sensor nodes with limited battery capacity that rely on acoustic communication and are deployed at the seabed, making battery recharging impractical. Despite the challenges in underwater communication, numerous routing protocols (RPs), such as Energy Balanced Efficient and Reliable Routing (EBER²) and Shifted Energy Efficiency and Priority (SHEEP), have been developed to optimise forwarding node selection and improve communication efficiency by incorporating parameters such as depth information (DI), transmission distance (TD), and residual energy (RE). However, designing energy-efficient (EE) and energy-balanced (EB) RPs for large-scale UWSNs remains an NP-hard problem due to the network's inherent complexity. This study introduces a Depth-Distance-based Energy-Efficient and Energy-Balanced (DDE²) routing protocol, which optimises energy consumption using TD, DI, RE, and additional parameters such as transmission range (TR). The DDE² protocol extends network lifetime while meeting critical quality-of-service (QoS) requirements, including scalability and low latency, and outperforms recent state-of-the-art RPs in energy efficiency, network longevity, and overall QoS for UWSNs.

Precision agriculture (PA) plays an essential role in resource use and crop yields while minimising environmental impact through data-driven farming techniques. The combination of unmanned aerial vehicles (UAVs), the Internet of Things (IoT) and wireless sensor networks (WSNs) has significantly transformed the current state of farming, enabling decisions based on data, predicting outcomes and precise control. This review presents the current developments, challenges and complementary advantages of these technologies to improve agricultural efficiency and sustainability in a comprehensive manner. The search timeframe of this search is 2019–2025. The analysis of the WSN-based systems begins with the analysis of sensing technologies, communication protocols (LoRa, Sigfox, Wi-Fi, Bluetooth, ZigBee, NB-IoT and RFID), sensor architecture, energy consumption and path-loss models, which affect the data transmission in an agricultural setting. It highlights the weaknesses of WSN deployment, such as power consumption and coverage. Second, the use of UAVs in crop monitoring, irrigation, pest detection and resource optimisation is reviewed with references to the incorporation of sensing and data analytics algorithms and the challenges associated with UAV use, such as the short flight duration and energy consumption. Third, IoT-based frameworks are researched in the context of their roles in the PA of real-time monitoring, automated controls and smart decision-making. The findings suggest that a network of WSNs, UAVs and the IoT can be used to enhance monitoring quality, data quality and resource utilisation by multiple orders of magnitude. However, such barriers as energy consumption, connectivity differences, complexity of data and costs will be topical. Overall, the WSN-UAV-IoT combo is a potentially fruitful direction that could assist PA to take a step forward in terms of productivity, sustainability and environmental friendliness.

Radio frequency fingerprinting identification (RFFI) leverages the unique features of communication transmitter signals to classify Internet of Things (IoT) devices, enabling individual recognition through waveform analysis. Traditional RFFI methods face challenges in extracting nonlinear features, which machine learning (ML) techniques help overcome by providing advanced wave characteristic analysis. This study introduces RFFI-SCNN, a hybrid model integrating RFFI with a spiking conventional neural network (SCNN) to enhance IoT device authentication within networks. The model operates in two phases: signal processing, where wave data are collected and preprocessed, and SCNN-based classification, where features are extracted and devices are authenticated. The proposed model's performance is evaluated against three ML-based models—1SNN, 1CNN and DCNN—based on accuracy, execution time and memory usage. Experimental results, conducted using a publicly available dataset from the Institute for the Wireless Internet of Things at Northeastern University, indicate that RFFI-SCNN achieves superior accuracy in classifying communication devices compared to 1CNN and 1SNN while also requiring less memory and shorter execution time than DCNN and 1CNN. These findings highlight the effectiveness of RFFI-SCNN in secure and efficient IoT device identification.

This paper proposes a hybrid framework for wireless sensor networks (WSNs) that enhances energy efficiency, data collection and sensor charging through a standard mobile sink (SMS) and a data fusion technique. The system incorporates rechargeable sensor nodes equipped with solar panels and an SMS that collects data while simultaneously recharging the sensors in the field. The proposed path optimisation and node utilisation using joint analysis (PANUJA) algorithm directs the SMS to gather fused data from cluster heads (CHs), which locally process the data using an ANN-based fusion method. To optimise network management, spectral clustering (SC) is employed for network partitioning, and Dijkstra's algorithm is used to determine the optimal anchor points and sink trajectories. Simulation results highlight significant improvements in energy efficiency, data transmission reliability and resilience to node failures. These advancements make the proposed approach well-suited for applications in smart cities and remote sensing. The framework also addresses limitations in traditional protocols such as low-energy adaptive clustering hierarchy (LEACH) and tree-based flooding technique (TBFT).

The rapid advancements in vehicular ad hoc networks (VANETs) call for development of effective networking schemes. Managing heterogenous traffic in VANETs becomes a critical challenge, especially when dealing with critical scenarios. In this paper, we present a novel dynamic fragmentation-based MAC protocol, DFROG-MAC for Internet of Things (IoT) applications in VANET environment. This protocol is focused on facilitating prioritised heterogenous traffic in sensor networks and hence, can offer a distinguished quality of service for various application areas such as vehicular, industrial or body sensor networks. DFROG-MAC deploys fragmentation scheme for low priority data, so the high priority data may interrupt and access channel without needing to wait for the complete transmission of lower priority data. The fragment size for the lower priority data is dynamically adjusted at the runtime based on the frequency of urgent traffic arrival. This dynamic approach helps to ensure that the channel does not remain idle, and lower priority traffic could be served quickly, in the absence of urgent traffic. Two types of traffic priorities, normal and urgent have been used for performance evaluation of FROG-MAC and DFROG-MAC, over Contiki platform, with the scope of this study focused on vehicle-to-infrastructure (V2I) single-hop communication. The delay and throughput both have been found to improve for DFROG-MAC due to the possibility of dynamic fragment size selection.

In wireless sensor networks (WSNs), optimising energy efficiency while maintaining coverage and managing resource constraints remains a critical challenge. This paper introduces a novel Region-Based Multilevel Energy Efficiency Protocol (RBMEEP), which innovatively partitions the network into regions and sub-regions to enhance energy utilisation through optimised clustering and communication with the base station (BS). Unlike conventional protocols, RBMEEP significantly extends network lifetime, outperforming the Stable Election Protocol (SEP). The novelty lies in the integration of a Regression Prediction Model (RPM), which accurately predicts network lifetime based on node density and packet size. Simulation results demonstrate the model's high prediction accuracy, with up to 99.94% in smaller network areas and 99.87% in larger areas. This predictive capability allows for adaptable and efficient WSN design, tailored to specific user requirements. The proposed approach presents a significant advancement in extending the operational life of WSNs, offering a robust solution for energy and coverage optimisation. This work not only improves the theoretical understanding of WSN energy efficiency but also provides a practical framework that can be deployed in real-world scenarios.

The Internet of Things (IoT) revolutionises communication systems and enables transformative applications across diverse domains. However, existing reviews often focus on integrating IoT with only one or two computing paradigms—cloud, fog, or edge computing—overlooking the holistic synergy of these architectures. This review bridges that gap by providing a comprehensive analysis of IoT integration with all three paradigms, emphasising their collective potential to address the challenges of scalability, latency, and computational efficiency. The findings highlight that cloud computing ensures scalable storage and processing but struggles with latency-sensitive IoT applications. Fog computing reduces latency by processing data near the network edge, achieving up to a 40% improvement in response times for real-time applications. Edge computing complements this by ensuring immediate data handling, reducing transmission delays by approximately 30% compared to cloud-centric models. Despite these advances, challenges persist, including high energy consumption, security vulnerabilities, and the complexity of managing dynamic workflows across architectures. This review provides actionable recommendations for future research, including the development of energy-efficient algorithms, robust security protocols, and adaptive frameworks for seamless integration. These directions are vital for advancing IoT as an indispensable component of the future Internet, fostering smarter and more connected systems across industries.

Wireless sensor networks (WSNs) are increasingly used in critical sectors such as defence, healthcare and environmental monitoring. These networks rely on small resource-constrained sensor nodes that communicate wirelessly, making them vulnerable to security threats. Although cryptographic methods, time synchronisation and error-correcting codes (ECCs) offer some protection, they often struggle with the computational and energy limitations of sensor nodes. Among ECCs, Hamming codes combined with quadratic residue (H-QR) techniques have shown promise in enhancing network security and improving performance metrics such as packet delivery ratio (PDR) and throughput (TP). However, existing H-QR implementations are limited in scalability, supporting only small networks with up to 15 nodes. To address this limitation, this study introduces an enhanced security architecture for clustered WSNs using Hamming codes with quadratic residue and nonresidue (H-QRN) properties. The proposed H-QRN scheme supports an arbitrary number of sensor nodes, making it suitable for large-scale and diverse industrial applications. Simulation results demonstrate that H-QRN significantly improves PDR and TP over traditional H-QR methods while maintaining similar end-to-end delay (E2E) and control overhead (CO). This work offers a scalable and efficient security solution for WSNs and provides practical insights for selecting security protocols tailored to specific application requirements.

Livestock monitoring systems have been significantly transformed by the implementation of Internet of Things (IoT) technology in agriculture. This integration enables the collection and analysis of data in real time, which contributes to improved animal welfare and productivity. This paper showcases the integration of various microcontrollers, sensors and sophisticated algorithms to give an extensive assessment of the most recent IoT-based livestock health monitoring systems. A wide array of sensors, including accelerometers, temperature sensors, heart rate sensors and more, coupled with various microcontrollers, such as Raspberry Pi, ESP8266, Arduino and ESP32, are primarily used in monitoring systems. Internet of Things (IoT) platforms such as ThingSpeak and Blynk, as well as the development of online interfaces and mobile applications, provide extensive user input. The integration of state-of-the-art algorithms is explored in detail, including support vector machines (SVM), decision trees, artificial neural networks (ANN), YOLOv5 object detection, different machine learning algorithms, random forest classifiers and ThingSpeak IoT analytics platform. Particular attention is given to algorithms that detect various parameters, including acetone levels, cow location, hormone release, body temperature, activity level, bellowing, jaw movement, heart rate, temperature, oestrous cycle, detection and tracking, action recognition, size variations, motion deformation, heat stress, ambient temperature, sleep tracking and more. The purpose of this review article is to facilitate the adoption of Internet of Things (IoT) solutions for sustainable and effective livestock management practices by offering a thorough analysis of existing IoT technologies used in livestock monitoring.

Two stacked layers of graphene on a typical dielectric with a back reflector are proposed. The structure is designed to stabilise the absorption response against probable mismatches. Additionally, the proposed absorber is modelled by an equivalent circuit model. Based on the optimised response, the design parameters can be selected by known algorithms. The finding suggests that the proposed structure is able to show absorption peaks in THz gap. Furthermore, the appropriate convergence of the circuit model approach with the full-wave simulation is a motivating reason to interact more deeply with the impedance matching concept. According to the simulation results, the proposed absorber express acceptable reliability against the design parameters while it can cover almost all of the THz gap and beyond (0.1 THz–20 THz). Design simplicity with an alternative modelling approach is leveraged in this work which can be exploited in several applications ranging from healthcare to the indoor communication.