Explore the vast range of titles available.

In this article, the problem of Bayesian detecting rank-one distributed targets under subspace interference and compound Gaussian clutter with inverse Gaussian texture is investigated. Due to the clutter heterogeneity, the training data may be insufficient. To tackle this problem, the clutter speckle covariance matrix (CM) is assumed to obey the complex inverse Wishart distribution, and the Bayesian theory is utilized to obtain an effective estimation. Moreover, the target echo is assumed to be with a known steering vector and unknown amplitudes across range cells. The interference is regarded as a steering matrix that is linearly independent of the target steering vector. By utilizing the generalized likelihood ratio test (GLRT), a Bayesian interference-canceling detector that can work in the absence of training data is derived. Moreover, five interference-cancelling detectors based on the maximum a posteriori (MAP) estimate of the speckle CM are proposed with the two-step GLRT, the Rao, Wald, Gradient, and Durbin tests. Experiments with simulated and measured sea clutter data indicate that the Bayesian interference-canceling detectors show better performance than the competitor in scenarios with limited training data.

Fuel cell microgrid, as an efficient and clean energy system, are widely used in industrial, commercial and residential sectors. However, due to the complexity of their operating environment, there are certain safety hazards, especially the risk of fire. Traditional fuel cell microgrid fire detection has weak ability to adapt to complex scenes, poor interactivity, untimely flame detection and cannot give detailed information about the fire, and target detection is difficult to detect small target flames. Aiming at the above problems, this paper proposes an improved flame detection algorithm for fuel cell microgrids with YOLOv5s, which can effectively reduce the interference of complex environment on flame detection and significantly improve the small target detection capability. In this study, the attention mechanism FLA (focused linear attention) is firstly added to YOLOv5s backbone network, which helps the network model to extract the flame multi-scale spatial information and important features more fully. Next, the Neck part of the neck network is improved by incorporating the weighted BiFPN (bidirectional feature pyramid network) structure to improve the small target detection capability and feature fusion capability. Then WloU (wise inter-section over union) as a new bounding box loss function to enhance the model generalisation ability. Finally, the inter-frame difference method is added to the detection head part of the model, which makes it able to effectively distinguish non-flame units when applied to flame detection cameras. Comparative experiments on the flame detection dataset show that the improved model improves the precision by 11.3%, the accuracy by 12.6%, and the recall by 3.8%.Moreover, the detection method proposed in this study has significantly enhanced detection capability for small target flames, and can be connected to detection cameras and alarm systems for effective detection of fuel cell microgrids, which can help to improve hydrogen energy safety and promote the demonstration application of fuel cell microgrids.

Statistical inference is a mature research area, but distributed inference problems that arise in the context of modern wireless sensor networks (WSNs) have new and unique features that have revitalized research in this area in recent years. The goal of this paper is to introduce the readers to these novel features and to summarize recent research developments in this area. In particular, results on distributed detection, parameter estimation and tracking in WSNs will be discussed, with a special emphasis on solutions to these inference problems that take into account the communication network connecting the sensors and the resource constraints at the sensors.

This study proposes a novel technique for detecting aerial moving targets using multiple satellite radars. The approach enhances the image contrast of fused local three-dimensional (3D) profiles. Exploiting global navigation satellite system (GNSS) satellites as illuminators of opportunity (IOs) has brought remarkable innovations to multistatic radar. However, target detection is restricted by radiation sources since IOs are often uncontrollable. To address this, we utilize satellite radars operating in an active self-transmitting and self-receiving mode for controllability. The main challenge of multiradar target detection lies in effectively fusing the target echoes from individual radars, as the target ranges and Doppler histories differ. To this end, two periods, namely the integration period and detection period, are precisely designed. In the integration period, we propose a range difference-based positive and negative second-order Keystone transform (SOKT) method to make range compensation accurate. This method compensates for the range difference rather than the target range. In the detection period, we develop two weighting functions, i.e., the Doppler frequency rate (DFR) variance function and smooth spatial filtering function, to extract prominent areas and make efficient detection, respectively. Finally, the results from simulation datasets confirm the effectiveness of our proposed technique.

Uncrewed aerial vehicles (UAVs), also known as drones, are ubiquitous and their use cases extend today from governmental applications to civil applications such as the agricultural, medical, and transport sectors, etc. In accordance with the requirements in terms of demand, it is possible to carry out various missions involving several types of UAVs as well as various onboard sensors. According to the complexity of the mission, some configurations are required both in terms of hardware and software. This task becomes even more complex when the system is composed of autonomous UAVs that collaborate with each other without the assistance of an operator. Several factors must be considered, such as the complexity of the mission, the types of UAVs, the communication architecture, the routing protocol, the coordination of tasks, and many other factors related to the environment. Unfortunately, although there are many research works that address the use cases of multi-UAV systems, there is a gap in the literature regarding the difficulties involved with the implementation of these systems from scratch. This review article seeks to examine and understand the communication issues related to the implementation from scratch of autonomous multi-UAV systems for collaborative decisions. The manuscript will also provide a formal definition of the ecosystem of a multi-UAV system, as well as a comparative study of UAV types and related works that highlight the use cases of multi-UAV systems. In addition to the mathematical modeling of the collaborative target detection problem in distributed environments, this article establishes a comparative study of communication architectures and routing protocols in a UAV network. After reading this review paper, readers will benefit from the multicriteria decision-making roadmaps to choose the right architectures and routing protocols adapted for specific missions. The open challenges and future directions described in this manuscript can be used to understand the current limitations and how to overcome them to effectively exploit autonomous swarms in future trends.

Target tracking is one of the important tasks of UAV (Unmanned Aerial Vehicle). The indicators for completing the UAV tracking task are the system’s tracking speed, tracking error, and tracking of specific targets. This paper proposes a distributed system based on the YOLO (You Only Look Once) detection algorithm and the DIMP (Discriminative Model Prediction) tracking algorithm for real-time tracking of specific targets. The system separates algorithm initialization and online operation, coordinates detection and tracking, and gives full play to the accuracy and speed advantages of both. Then, on the basis of the YOLO detection algorithm, the CBAM (Convolutional Block Attention Module) module is integrated to improve the network structure, which improves the detection accuracy of the YOLO algorithm for specific targets. Then, on the basis of the DIMP tracking algorithm, the tracking speed and stability of the DIMP algorithm are improved by distributed design, replacing the backbone network, optimizing training and updating, and integrating the attention mechanism. Finally, simulation and experiments verified that the distributed system can achieve long-term and stable tracking of specific targets. The detection accuracy (mAP0.5:0.95) of the YOLO algorithm integrated with CBAM was improved by 5.2%, and the tracking speed of the improved DIMP algorithm was increased by 10FPS.

In distributed radar, a coherent system often gains attention for its higher detection potential in contrast to its non-coherent counterpart. However, even for a distributed coherent radar, it is difficult to coherently accumulate local observations in the searching mode if target returns in local channels are decorrelated. In order to obtain the superiority of coherent processing while overcoming the real implementation difficulties of a coherent framework, this paper studies a distributed coherent detection algorithm for fusion detection. It is utilized in detecting a target during tracking while a target is searched for in a non-coherent manner. From historic observations on target tracking, relative phase delays in different channels are predicted by a phase lock loop and then used to compensate phases for observations in the current frame. Moreover, to enhance the detection performance of distributed radar during tracking, a switching rule between phase prediction-based coherent and non-coherent processing is proposed based on their detection performance. Numerical results indicate that the switching operation can improve the detection probability during tracking, and the non-coherent operation can still provide a moderate detection performance if the phase prediction is unreliable.

Unmanned aerial vehicles (UAVs) are effective tools for identifying road anomalies with limited detection coverage due to the discrete spatial distribution of roads. Despite computational, storage, and transmission challenges, existing detection algorithms can be improved to support this task with robustness and efficiency. In this study, the K-means clustering algorithm was used to calculate the best prior anchor boxes; Faster R-CNN (region-based convolutional neural network), YOLOX-s (You Only Look Once version X-small), YOLOv5-s, YOLOv7-tiny, YOLO-MobileNet, and YOLO-RDD models were built based on image data collected by UAVs. YOLO-MobileNet has the most lightweight model but performed worst in accuracy, but greatly reduces detection accuracy. YOLO-RDD (road distress detection) performed best with a mean average precision (mAP) of 0.701 above the Intersection over Union (IoU) value of 0.5 and achieved relatively high accuracy in detecting all four types of distress. The YOLO-RDD model most successfully detected potholes with an AP of 0.790. Significant or severe distresses were better identified, and minor cracks were relatively poorly identified. The YOLO-RDD model achieved an 85% computational reduction compared to YOLOv7-tiny while maintaining high detection accuracy.

This paper considers the detection of fluctuating targets in heavy-tailed clutter through the use of dynamic programming based on track-before-detect (DP–TBD) in radar systems. The clutter is modeled in terms of K-distribution, which can be widely used to describe non-Gaussian clutter received from high-resolution radars and radars working at small grazing angles. Swerling type 1 is considered to describe the target fluctuation between scans. Conventional TBD techniques suffer from significant performance loss in heavy-tailed environments due to the more frequent occurrences of target-like outliers. In this paper, we resort to a DP–TBD algorithm based on prior information, which can enhance the detection performance by using the environment and target fluctuating information during the integration process of TBD. Under non-Gaussian background, the expressions of the likelihood ratio merit function for Swerling type 1 targets are derived first. However, the closed-form of the merit function is difficult to obtain. In order to reduce the complexity of evaluating the merit function and the computational load, an efficient approximation method as well as a two-stage detection approach is proposed and used in the integration process. Finally, several numerical simulations of the new strategy and the comparisons are presented to verify that the proposed algorithm can improve the detection performance, especially for fluctuating targets in heavy-tailed clutter.

With the continuous development of remote sensing technology, the data volume of remote sensing images has increased exponentially, resulting in many difficulties in the storage, management, transmission, calculation, and other processes of remote sensing images. In order to solve the above problems, this paper studies the use of the Hadoop Distributed File System (HDFS) and related technologies to design and implement a browser/server (B/S) architecture for a massive, multisource, remote sensing images distributed storage management system. The image data are stored in the HDFS, and the image metadata are stored in a MySQL database. The distributed parallel construction of the image pyramid is completed based on the Spark computing engine, and the Akka framework is used to construct WMTS (Web Map Tile Service) to realize the release of remote sensing images. Finally, the rapid visual display of remote sensing images is carried out using Leaflet. The system also supports image data management, image target detection, user management, and other functions. After testing, this system can support the storage and management of multisource remote sensing image data, and can solve perfectly the problems of insufficient storage space and insufficient computing power of a single server. It is found that the upload and download speeds of a large amount of remote sensing images can be close to the maximum speed of a gigabit local area network (LAN). In the gigabit LAN environment, the average upload speed of a single remote sensing image is 97.74 MB/s, and the average download speed is 87.62 MB/s. In terms of image pyramid construction, the speed of a multi-node parallel construction based on Spark is two times higher than that of a single-node construction. Additionally, compared to similar systems, this system has better data transmission and retrieval speed, better data computing ability, and higher concurrency processing ability.