Search Results (80)

The integration of radar and Automatic Dependent Surveillance–Broadcast (ADS-B) surveillance data is critical for increasing the accuracy of air traffic monitoring; however, effective track associations remain challenging due to inherent sensor discrepancies and computational constraints. To achieve accurate identification and association between radar tracks and ADS-B tracks, this study proposes an adaptive feature extraction method based on the longest common subsequence (LCSS) combined with classification theory to address the limitations inherent in traditional machine learning-based track association approaches. These limitations encompass challenges in acquiring training samples, extended training times, and limited model generalization performance. The proposed method employs LCSS to measure the similarity between two types of trajectories and categorizes tracks into three groups—definite associations, definite nonassociations, and fuzzy associations—using a similarity matrix and an adaptive sample classification model (adaptive classification model). Fuzzy mathematical techniques are subsequently applied to extract discriminative features from both definite association and nonassociation sets, followed by training a support vector machine (SVM) model. Finally, the SVM performs classification and association of trajectories in the fuzzy association group. The computational results show that, compared with conventional statistical methods, the proposed methodology achieves both superior precision and recall rates while maintaining computational efficiency threefold that of traditional machine learning algorithms. Full article

This study investigates the accuracy of traffic flow and speed measurements using two radar-based devices, Sierzega and SDR, against manual video-based traffic counts. The measurements were conducted over a 12-h period on a congested urban road section characterized by variable traffic conditions and frequent vehicle stops. The results revealed that the SDR device generally provided lower deviations compared to manual counting, especially in measuring traffic flow. In contrast, the Sierzega device demonstrated greater and more inconsistent deviations, particularly in vehicle categorization and traffic density estimation. The observed discrepancies were primarily attributed to vehicle stopping and queuing, influencing length estimation and classification errors. Despite these limitations, SDR provided sufficient accuracy for practical applications, such as monitoring traffic trends or supporting long-term traffic planning in urban environments. Full article

With the rapid advancement of modern society, the demand for efficient and convenient transportation has increased significantly, making traffic congestion a pressing challenge that must be addressed in the process of urban expansion. To effectively mitigate this issue, we propose an approach that leverages sensor networks to monitor real-time traffic data across road networks, enabling the precise characterization of traffic flow dynamics. This method integrates the Webster algorithm with a proportional–integral–derivative (PID) controller, whose parameters are optimized using a genetic algorithm, thereby facilitating scientifically informed traffic signal timing strategies for enhanced traffic regulation. Geomagnetic sensors are deployed along the roads at a ratio of 1:50–1:60, and radar sensors are deployed on the roadsides of key sections. This can effectively detect changes in road traffic flow and provide early warnings for possible accidents. The integration of the Webster method with a genetically optimized PID controller enables adaptive traffic signal timing with minimal energy consumption, effectively reducing road occupancy rates and mitigating congestion-related risks. Compared to conventional fixed-time control schemes, the proposed approach improves traffic regulation efficiency by 17.3%. Furthermore, it surpasses traditional real-time adaptive control strategies by 3% while significantly lowering communication energy expenditure. Notably, during peak hours, the genetically optimized PID controller enhances traffic control effectiveness by 13% relative to its non-optimized counterpart. A framework is proposed to improve the efficiency of road operation under the condition of random traffic changes. The k-means method is used to mark key roads, and weights are assigned based on this to coordinate and regulate traffic conditions. These findings underscore our contribution to the field of intelligent transportation systems by presenting a novel, energy-efficient, and highly effective traffic management solution. The proposed method not only advances the scientific understanding of dynamic traffic control but also offers a robust technical foundation for alleviating urban traffic congestion and improving overall travel efficiency. Full article

This paper presents the design, implementation, and testing of a traffic monitoring platform based on 5G-connected Ultra-Wideband (UWB) radars deployed on a university campus. The development of both connected radars and an IoT platform is detailed. The connected radars integrate commercial components, including a Raspberry Pi (RPi), a UWB radar, a standard enclosure, and a custom communication board featuring a 5G module. The IoT platform, which receives data from the radars via MQTT, is scalable, easily deployable, and supports radar management, data visualization, and external data access via an API. The solution was deployed and tested on campus, demonstrating real-time operation over a commercial 5G network with an estimated median latency between the radar and server of 75 ms. A preliminary evaluation conducted on a single radar during peak-hour traffic on a double-lane road, representing a challenging scenario, indicated a high detection rate of 94.81%, a low false detection rate of 1.02%, a high classification accuracy of 97.29%, and a high direction accuracy of 99.66%. These results validate the system’s capability to deliver accurate traffic monitoring. Full article

Synthetic aperture radar (SAR) ship detection holds significant application value in maritime monitoring, marine traffic management, and safety maintenance. Despite remarkable advances in deep-learning-based detection methods, performance remains constrained by the vast size differences between ships, limited feature information of small targets, and complex environmental interference in SAR imagery. Although many studies have separately tackled small target identification and multi-scale detection in SAR imagery, integrated approaches that jointly address both challenges within a unified framework for SAR ship detection are still relatively scarce. This study presents MC-ASFF-ShipYOLO (Monte Carlo Attention—Adaptively Spatial Feature Fusion—ShipYOLO), a novel framework addressing both small target recognition and multi-scale ship detection challenges. Two key innovations distinguish our approach: (1) We introduce a Monte Carlo Attention (MCAttn) module into the backbone network that employs random sampling pooling operations to generate attention maps for feature map weighting, enhancing focus on small targets and improving their detection performance. (2) We add Adaptively Spatial Feature Fusion (ASFF) modules to the detection head that adaptively learn spatial fusion weights across feature layers and perform dynamic feature fusion, ensuring consistent ship representations across scales and mitigating feature conflicts, thereby enhancing multi-scale detection capability. Experiments are conducted on a newly constructed dataset combining HRSID and SSDD. Ablation experiment results demonstrate that, compared to the baseline, MC-ASFF-ShipYOLO achieves improvements of 1.39% in precision, 2.63% in recall, 2.28% in ${AP}_{50}$, and 3.04% in $AP$, indicating a significant enhancement in overall detection performance. Furthermore, comparative experiments show that our method outperforms mainstream models. Even under high-confidence thresholds, MC-ASFF-ShipYOLO is capable of predicting more high-quality detection boxes, offering a valuable solution for advancing SAR ship detection technology. Full article

In air traffic systems, aircraft trajectories between airports are monitored by the radar networking system forming dynamic air traffic flow. Accurate airport arrival flow prediction is significant in implementing large-scale intelligent air traffic flow management. Despite years of studies to improve prediction precision, most existing methods only focus on a single airport or simplify the traffic network as a static and simple graph. To mitigate this shortage, we propose a hybrid neural network method, called Dynamic Multi-graph Convolutional Spatial-Temporal Network (DMCSTN), to predict network-level airport arrival flow considering the multiple operation constraints and flight interactions among airport nodes. Specifically, in the spatial dimension, a novel dynamic multi-graph convolutional network is designed to adaptively model the heterogeneous and dynamic airport networks. It enables the proposed model to dynamically capture informative spatial correlations according to the input traffic features. In the temporal dimension, an enhanced self-attention mechanism is utilized to mine the arrival flow evolution patterns. Experiments on a real-world dataset from an ATFM system validate the effectiveness of DMCSTN for arrival flow forecasting tasks. Full article

Traditional 3D radar can only detect the planar characteristic information of a target. Thus, it cannot describe its spatial geometric characteristics, which is critical for accurate vehicle classification. To overcome these limitations, this paper investigates elevation features using 4D millimeter-wave radar data and presents a maximum likelihood estimation (MLE)-based vehicle classification method. The elevation data collected by 4D radar from a real road scenario are applied for further analysis. By establishing radar coordinate systems and geodetic coordinate systems, the distribution feature of vehicles’ elevation is analyzed by spatial geometric transformation referring to the radar installation parameters, and a Gaussian-based probability distribution model is subsequently proposed. Further, the data-driven parameter optimization on likelihood probabilities of different vehicle samples is performed using a large-scale elevation dataset, and an MLE-based vehicle classification method is presented for identifying small and large vehicles. The experimental results show that there are significant differences in elevation distribution from two different vehicle types, where large vehicles exhibit a wider range of left-skewed distribution in different cross-sections, while small vehicles are more concentrated with a right-skewed distribution. The Gaussian-based MLE method achieves an accuracy of 92%, precision of 87% and recall of 98%, demonstrating excellent performance for traffic monitoring and related applications. Full article

Soil degradation is a critical challenge to global agricultural sustainability, driven by intensive land use, unsustainable farming practices, and climate change. Conventional soil monitoring techniques often rely on invasive sampling methods, which can be labor-intensive, disruptive, and limited in spatial coverage. In contrast, non-invasive geophysical techniques, particularly ground-penetrating radar, have gained attention as tools for assessing soil properties. However, an assessment of ground-penetrating radar’s applications in agricultural soil research—particularly for detecting soil structural changes related to degradation—remains undetermined. To address this issue, a systematic literature review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guidelines. A search was conducted across Scopus and Web of Science databases, as well as relevant review articles and study reference lists, up to 31 December 2024. This process resulted in 86 potentially relevant studies, of which 24 met the eligibility criteria and were included in the final review. The analysis revealed that the ground-penetrating radar allows the detection of structural changes associated with tillage practices and heavy machinery traffic in agricultural lands, namely topsoil disintegration and soil compaction, both of which are important indicators of soil degradation. These variations are reflected in changes in electrical permittivity and reflectivity, particularly above the tillage horizon. These shifts are associated with lower soil water content, increased soil homogeneity, and heightened wave reflectivity at the upper boundary of compacted soil. The latter is linked to density contrasts and waterlogging above this layer. Additionally, ground-penetrating radar has demonstrated its potential in mapping alterations in electrical permittivity related to preferential water flow pathways, detecting shifts in soil organic carbon distribution, identifying disruptions in root systems due to tillage, and assessing soil conditions potentially affected by excessive fertilization in iron oxide-rich soils. Future research should focus on refining methodologies to improve the ground-penetrating radar’s ability to quantify soil degradation processes with greater accuracy. In particular, there is a need for standardized experimental protocols to evaluate the effects of monocultures on soil fertility, assess the impact of excessive fertilization effects on soil acidity, and integrate ground-penetrating radar with complementary geophysical and remote sensing techniques for a holistic approach to soil health monitoring. Full article

Dams are crucial for ensuring water and electricity supply, while also providing significant flood protection. Regular monitoring of dam deformations is of vital socio-economic and ecological significance. In Germany, dams must be constructed and operated according to generally accepted rules of engineering. The safety concept for dams based on these rules relies on structural safety, professional operation and maintenance, safety monitoring, and precautionary measures. Rather time-consuming in situ techniques have been employed for these measurements, which permit monitoring deformations with either high spatial or temporal resolution, but not both. As a means of measuring large-scale deformations in the millimeter range, the Differential Interferometric Synthetic Aperture Radar (DInSAR) technique of Persistent Scatterer Interferometry (PSI) is already being applied in various fields. However, when considering the operational monitoring of dams using PSI, specific characteristics need to be considered. For example, the geographical location of the dam in space, as well as its shape, size, and land cover. All these factors can affect the visibility of the structure for the use with PSI and, in certain cases, limit the applicability of SAR data. The visibility of dams for PSI monitoring is often limited, particularly in cases where observation is typically not feasible due to factors such as geographical and structural characteristics. While corner reflectors can improve visibility, their large size often makes them unsuitable for dam infrastructure and may raise concerns with heritage protection for listed dams. Addressing these challenges, electronic corner reflectors (ECRs) offer an effective alternative due to their small and compact size. In this study, we analyzed the strategic placement of ECRs on dam structures. We developed a new CR Index, which identifies areas where PSI alone is insufficient due to unfavorable geometric or land use conditions. This index categorizes visibility potential into three classes, presented in a ‘traffic light’ map, and is instrumental in selecting optimal installation sites. We furthermore investigated the signal stability of ECRs over an extended observation period, considering the Amplitude Dispersion Index (ADI). It showed values between 0.1 and 0.4 for many dam structures, which is comparable to normal corner reflectors (CRs), confirming the reliability of these signals for PSI analysis. This work underscores the feasibility of using ECRs to enhance monitoring capabilities at dam infrastructure. Full article

Small target ship detection and anomaly analysis play a pivotal role in ocean remote sensing technologies, offering critical capabilities for maritime surveillance, enhancing maritime safety, and improving traffic management. However, existing methodologies in the field of detection are predominantly based on deep learning models with complex network architectures, which may fail to accurately detect smaller targets. In the classification domain, most studies focus on synthetic aperture radar (SAR) images combined with Automatic Identification System (AIS) data, but these approaches have significant limitations: first, they often overlook further analysis of anomalies arising from mismatched data; second, there is a lack of research on small target ship classification using wide-area optical remote sensing imagery. In this paper, we develop SVIADF, a multi-source information fusion framework for small vessel identification and anomaly detection. The framework consists of two main steps: detection and classification. To address challenges in the detection domain, we introduce the YOLOv8x-CA-CFAR framework. In this approach, YOLOv8x is first utilized to detect suspicious objects and generate image patches, which are then subjected to secondary analysis using CA-CFAR. Experimental results demonstrate that this method achieves improvements in Recall and F1-score by 2.9% and 1.13%, respectively, compared to using YOLOv8x alone. By integrating structural and pixel-based approaches, this method effectively mitigates the limitations of traditional deep learning techniques in small target detection, providing more practical and reliable support for real-time maritime monitoring and situational assessment. In the classification domain, this study addresses two critical challenges. First, it investigates and resolves anomalies arising from mismatched data. Second, it introduces an unsupervised domain adaptation model, Multi-CDT, for heterogeneous multi-source data. This model effectively transfers knowledge from SAR–AIS data to optical remote sensing imagery, thereby enabling the development of a small target ship classification model tailored for optical imagery. Experimental results reveal that, compared to the CDTrans method, Multi-CDT not only retains a broader range of classification categories but also improves target domain accuracy by 0.32%. The model extracts more discriminative and robust features, making it well suited for complex and dynamic real-world scenarios. This study offers a novel perspective for future research on domain adaptation and its application in maritime scenarios. Full article

In response to the limitations of existing vehicle detection technologies when applied to distributed sensor networks for road traffic holographic perception, this paper proposes a vehicle information detection technology based on single-point laser ranging. The system uses two single-point laser radars with fixed angles, combined with an adaptive threshold state machine and waveform segmentation fusion, to accurately detect vehicle speed, lane position, and other parameters. Compared with traditional methods, this technology offers advantages such as richer detection dimensions, low cost, and ease of installation and maintenance, making it suitable for large-scale traffic monitoring on secondary roads, highways, and suburban roads. Experimental results show that the system achieves high accuracy and reliability in low-to-medium-traffic flow scenarios, demonstrating its potential for intelligent road traffic applications. Full article

Passive radar has drawn a lot of attention due to its applications across military and civilian sectors. Under this working paradigm, the utilization of antenna arrays is instrumental, as it increases the signal quality and enables precise target positioning. These promising features rely, however, on the precise calibration of the antenna array, as the different hardware components introduce impairments that compromise the beamforming capabilities of the system. We propose a technique that employs a low-power external beacon signal to produce precise information about the target location, avoiding the angular ambiguities present in other solutions in the literature. The experimental results demonstrate the method’s ability to effectively correct the amplitude and phase inconsistencies while compensating for frequency drifts, enabling beamforming capabilities and direction-of-arrival estimation. Among the tested beacon waveforms, the pseudo-random noise-based signals proved the most robust, especially in low-power scenarios. Additionally, the method was validated in a passive radar setup, where it successfully detected a vessel using opportunistic signals. These findings highlight the method’s potential to enhance passive radar performance while maintaining a low probability of detection, a key aspect in military applications, as well as its applicability to civilian purposes, such as infrastructure monitoring, environmental observation, and traffic management. Full article

Autonomous vehicles (AVs) have revolutionized the automotive industry by leveraging data to perceive and interact with their environment effectively. Data safety is essential for supporting AV decision-making and ensuring reliability in complex environments. AVs continuously collect data from multiple sources like LiDAR, RADAR, cameras, and ultrasonic sensors to monitor road conditions, traffic signals, and pedestrian movements. An effective data classification framework is crucial for managing vast amounts of information and securing AV systems against cyber threats. This paper proposes a comprehensive framework for AV data classification, categorizing data by sensitivity, usage, and source. By integrating a review of the literature, real-world cases, and practical insights, this study introduces a novel data classification model and explores sensitivity criteria. The findings aim to assist industry stakeholders in creating secure, efficient, and sustainable AV ecosystems. Full article

Synthetic aperture radar (SAR) is an essential tool for monitoring and managing maritime traffic and ensuring safety. It is particularly valuable because it can provide surveillance in all weather conditions. Ship wake detection has attracted considerable attention in offshore management as it has potential for widespread use in ship positioning and motion parameter inversion, surpassing conventional ship detection methods. Traditional wake detection methods depend on linear feature extraction through image transformation processing techniques, which are often ineffective and time-consuming when applied to large-scale SAR data. Conversely, deep learning (DL) algorithms have been infrequently utilized in wake detection and encounter significant challenges due to the complex ocean background and the effect of the sea state. In this study, we propose a lightweight rotating target detection network designed for detecting ship wakes under various sea states. For this purpose, we initially analyzed the features of wake samples across various frequency domains. In the framework, a YOLO structure-based deep learning is implemented to achieve wake detection. Our network design enhances the YOLOv8’s structure by incorporating advanced techniques such as deep separation convolution and combined frequency domain–spatial feature extraction modules. These modules are used to replace the usual convolutional layer. Furthermore, it integrates an attention technique to extract diverse features. By conducting experiments on the OpenSARWake dataset, our network exhibited outstanding performance, achieving a wake detection accuracy of 66.3% while maintaining a compact model size of 51.5 MB and time of 14 ms. This model size is notably less than the existing techniques employed for rotating target detection and wake detection. Additionally, the algorithm exhibits excellent generalization ability across different sea states, addressing to a certain extent the challenge of wake detection being easily influenced by varying sea states. Full article

Avalanches present substantial hazard risk in mountainous regions, particularly when avalanches obstruct roads, either hitting vehicles directly or leaving traffic exposed to subsequent avalanches during cycles. Traditional detection methods often are designed to cover only a limited section of a road stretch, hampering effective risk management. This research introduces a novel approach using Distributed Acoustic Sensing (DAS) for avalanche detection. The monitoring site in Northern Norway is known to be frequently impacted by avalanches. Between 2022–2024, we continuously monitored the road for avalanches blocking the traffic. The automated alert system identifies avalanches affecting the road and estimates accumulated snow. The system provides continuous, real-time monitoring with competitive sensitivity and accuracy over large areas (up to 170 km) and for multiple sites on parallel. DAS powered alert system can work unaffected by visual barriers or adverse weather conditions. The system successfully identified 10 road-impacting avalanches (100% detection rate). Our results via DAS align with the previous works and indicate that low frequency part of the signal (<20 Hz) is crucial for detection and size estimation of avalanche events. Alternative fiber installation methods are evaluated for optimal sensitivity to avalanches. Consequently, this study demonstrates its durability and lower maintenance requirements, especially when compared to the high setup costs and coverage limitations of radar systems, or the weather and lighting vulnerabilities of cameras. Furthermore the system can detect vehicles on the road as important supplemental information for search and rescue operations, and thus the authorities can be alerted, thereby playing a vital role in urgent rescue efforts. Full article