China Electronics Standardization Institute

INTRODUCTION: Next to existing terminology of the lower urinary tract, due to its increasing complexity, the terminology for pelvic floor dysfunction in women may be better updated by a female-specific approach and clinically based consensus report. METHODS: This report combines the input of members of the Standardization and Terminology Committees of two international organizations, the International Urogynecological Association (IUGA), and the International Continence Society (ICS), assisted at intervals by many external referees. Appropriate core clinical categories and a subclassification were developed to give an alphanumeric coding to each definition. An extensive process of 15 rounds of internal and external review was developed to exhaustively examine each definition, with decision-making by collective opinion (consensus). RESULTS: A terminology report for female pelvic floor dysfunction, encompassing over 250 separate definitions, has been developed. It is clinically based with the six most common diagnoses defined. Clarity and user-friendliness have been key aims to make it interpretable by practitioners and trainees in all the different specialty groups involved in female pelvic floor dysfunction. Female-specific imaging (ultrasound, radiology, and MRI) has been a major addition while appropriate figures have been included to supplement and help clarify the text. Ongoing review is not only anticipated but will be required to keep the document updated and as widely acceptable as possible. CONCLUSION: A consensus-based terminology report for female pelvic floor dysfunction has been produced aimed at being a significant aid to clinical practice and a stimulus for research.

With the rapid expansion in the number of Unmanned Aircraft Vehicles (UAVs) available and the development of modern technologies, the commercial applications of UAVs in urban areas, such as urban remote sensing (RS), express services, urban road traffic monitoring, urban police security, urban air shows and so on, have increased greatly. However, most UAVs, especially light and small civil UAVs, have been operating in low-altitude airspace, and a conflict may exist between increasing the number of UAVs and the limited low airspace. To promote low-altitude airspace resource development and to standardize the operation and management of UAVs in urban regions, some global laws and regulations and key technologies for urban low-altitude applications of UAVs have been implemented. This paper reviews the development of current policies and key technologies concerning safe and efficient operations of the light-and-small civil UAVs in low altitude in urban areas. Discussions are made progressively on measures and methods of airspace restriction, airspace structuring and air route planning in China primarily and the rest of world. After surveying the practical industry tests and the initial studies of air routes, the survey results indicate that the construction of air route networks is a scientific and effective measure to standardize and improve the efficiency of low-altitude UAV operations. From the view point of safety and efficiency, the most valuable direction for UAV regulation in urban regions involves deepening the research which largely relies on urban RS and Geographic Information System (GIS) technology, and application demonstrations of low-altitude public air route networks.

Grid-connected renewable energy conversion systems (RECSs) are usually required by grid codes to possess the low voltage ride through (LVRT) and reactive power support capabilities so as to cope with grid voltage sags. During LVRT, RECS's terminal voltage becomes sensitive and changeable with its output current, which brings a great challenge for the RECS to resynchronize with the grid by means of phase-locked loops (PLLs). This paper indicates that loss of synchronism (LOS) of PLLs is responsible for the transient instability of grid-connected RECSs during LVRT, and the LOS is essentially due to the transient interaction between the PLL and the weak terminal voltage. For achieving a quantitative analysis, an equivalent swing equation model is developed to describe the transient interaction. Based on the model, the transient instability mechanism of RECSs during LVRT is clarified. Furthermore, a transient stability enhancement method is proposed to avoid the possibility of transient instability. Simulations performed on the New England 39-bus test system verify the effectiveness of the method.

In this paper, a novel model-free adaptive control (MFAC) algorithm based on a dual successive projection (DuSP)-MFAC method is proposed, and it is analyzed using the introduced DuSP method and the symmetrically similar structures of the controller and its parameter estimator of MFAC. Then, the proposed DuSP-MFAC scheme is successfully implemented in an autonomous car "Ruilong" for the lateral tracking control problem via converting the trajectory tracking problem into a stabilization problem by using the proposed preview-deviation-yaw angle. This MFAC-based lateral tracking control method was tested and demonstrated satisfactory performance on real roads in Fengtai, Beijing, China, and through successful participation in the Chinese Smart Car Future Challenge Competition held in 2015 and 2016.

Alzheimer's disease (AD) is a neurodegenerative disorder that causes memory degradation and cognitive function impairment in elderly people. The irreversible and devastating cognitive decline brings large burdens on patients and society. So far, there is no effective treatment that can cure AD, but the process of early-stage AD can slow down. Early and accurate detection is critical for treatment. In recent years, deep-learning-based approaches have achieved great success in Alzheimer's disease diagnosis. The main objective of this paper is to review some popular conventional machine learning methods used for the classification and prediction of AD using Magnetic Resonance Imaging (MRI). The methods reviewed in this paper include support vector machine (SVM), random forest (RF), convolutional neural network (CNN), autoencoder, deep learning, and transformer. This paper also reviews pervasively used feature extractors and different types of input forms of convolutional neural network. At last, this review discusses challenges such as class imbalance and data leakage. It also discusses the trade-offs and suggestions about pre-processing techniques, deep learning, conventional machine learning methods, new techniques, and input type selection.

As a mode of processing task request, edge computing paradigm can reduce task delay and effectively alleviate network congestion caused by the proliferation of Internet of things(IoT) devices compared with cloud computing. However, in the actual construction of the network, there are various edge autonomous subnets in the adjacent areas, which leads to the possibility of unbalance of server load among autonomous subnets during the peak period of task request. In this paper, a deep reinforcement learning algorithm is proposed to solve the complex computation offloading problem for the heterogeneous Edge Computing Server(ECS) collaborative computing. The problem is solved based on the real-time state of the network and the attributes of the task, which adopts Actor Critic and Policy Gradient's Deep Deterministic Policy Gradient(DDPG) to make optimized decisions of computation offloading. Considering multi-task, the heterogeneity of edge subnet and mobility of edge devices, the proposed algorithm can learn the network environment and generate the computation offloading decision to minimize the task delay.The simulation results show that the proposed DDPG-based algorithm is competitive compared with the Deep Q Network(DQN) algorithm and Asynchronous Advantage Actor-Critic(A3C) algorithm. Moreover, the optimal solutions are leveraged to analyze the influence of edge network parameters on task delay.

• Establish a comprehensive framework for model robustness, containing 23 different metrics. • Provide an open-sourced platform to support easy-to-use robustness evaluation and continuous integration. • Conduct large-scale experiments and provide preliminary suggestions to the evaluation of model robustness. Deep neural networks (DNNs) have achieved remarkable performance across a wide range of applications, while they are vulnerable to adversarial examples , which motivates the evaluation and benchmark of model robustness. However, current evaluations usually use simple metrics to study the performance of defenses, which are far from understanding the limitation and weaknesses of these defense methods. Thus, most proposed defenses are quickly shown to be attacked successfully, which results in the “arm race” phenomenon between attack and defense. To mitigate this problem, we establish a model robustness evaluation framework containing 23 comprehensive and rigorous metrics, which consider two key perspectives of adversarial learning (i.e., data and model). Through neuron coverage and data imperceptibility , we use data-oriented metrics to measure the integrity of test examples; by delving into model structure and behavior, we exploit model-oriented metrics to further evaluate robustness in the adversarial setting . To fully demonstrate the effectiveness of our framework, we conduct large-scale experiments on multiple datasets including CIFAR-10, SVHN, and ImageNet using different models and defenses with our open-source platform. Overall, our paper provides a comprehensive evaluation framework, where researchers could conduct comprehensive and fast evaluations using the open-source toolkit, and the analytical results could inspire deeper understanding and further improvement to the model robustness.

Single image-based dehazing has achieved remarkable progress with the development of deep learning technologies. End-to-end neural networks have been proposed to learn a direct hazy-to-clear image translation to recover the clear structures and edges cues from the hazy inputs. However, the frequency domain information is explored insufficiently and lots of intermediate structure and texture related cues of current dehazing networks are ignored, which limits the performances of current approaches. To handle these limitations mentioned above, a wavelet spatial attention based multi-stream feedback network (WSAMF-Net) is proposed for effective single image dehazing. Specifically, the proposed wavelet spatial attention utilizes both frequency-domain and spatial-domain information to enhance the extracted features for better structures and edges. Meanwhile, an enhanced multi-stream based cross feature fusion strategy, including vertical and horizontal attentions, is proposed to reweight and fuse the intermediate features of each stream to acquire more meaningful aggregated features, while the weight sharing strategy is used to achieve a good trade-off between performance and parameters. Besides, feedback mechanism is also designed to provide strong reconstruction ability. Furthermore, we propose a critical real-world industrial dataset (IDS) with images captured in real-world industrial quarry scenarios for research uses. Extensive experiments on various benchmarking datasets, including both synthetic and real-world datasets, demonstrate the superiority of our WSAMF-Net over state-of-the-art single image dehazing methods. The IDS dataset will be available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/XBSong/IDS-Datasethttps://github.com/XBSong/IDS-Dataset</uri> .

Abstract Manipulating light polarization is of fundamental importance in the modern photonic applications such as spectroscopy, laser science, optical communication, quantum information processing, chemical and biological sensing. Polarization control of light is typically achieved by natural chiral or birefringent materials with macroscopic volume due to the weak light–matter interaction. Here, a folded eta‐shaped metamaterial that is capable of generating gigantic optical chirality and spectrally breaking the spin degeneracy of optical transmission at multiband is experimentally demonstrated. The intrinsic chiral configuration is achieved by folding the eta‐shaped metasurface along the vertical direction to break the mirror symmetries. A remarkable circular dichroism approaching unity is experimentally achieved, with the maximum transmittance exceeding 93%. This is the record high value demonstrated to date for single‐layer metasurfaces without diffraction in infrared region. The folded metamaterial provides a straightforward strategy for achieving intrinsic 3D chirality and has great potential for applications in photon‐spin selective devices and chiral biomolecule identification.

With the increasing availability of diverse healthcare data sources, such as medical images and electronic health records, there is a growing need to effectively integrate and fuse this multimodal data for comprehensive analysis and decision-making. However, despite its potential, multimodal data fusion in healthcare remains limited. This review paper provides an overview of existing literature on multimodal data fusion in healthcare, covering 69 relevant works published between 2018 and 2024. It focuses on methodologies that integrate different data types to enhance medical analysis, including techniques for integrating medical images with structured and unstructured data, combining multiple image modalities, and other features. Additionally, the paper reviews various approaches to multimodal data fusion, such as early, intermediate, and late fusion methods, and examines the challenges and limitations associated with these techniques. The potential benefits and applications of multimodal data fusion in various diseases are highlighted, illustrating specific strategies employed in healthcare artificial intelligence (AI) model development. This research synthesizes existing information to facilitate progress in using multimodal data for improved medical diagnosis and treatment planning.

With the development of information technology, the network connection of industrial control system (ICS) and information technology (IT) is becoming more and more closely. What's more, the security issues of traditional IT systems in industrial control system are also more prominent. Early industrial control system mainly uses physical isolation approach to protect security. In this paper, we review the characteristics and reference models of industrial control system and analyze the current security status of industrial control system. Moreover, we propose a defense-in-depth system, security policies of active protection and passive monitoring for these security issues. Besides, we also discuss the key technologies and summarize the full text.

The shuttle effect of dissolved polysulfides produced during the operation of lithium–sulfur batteries is the most serious and fundamental problem among many challenges. We propose a strategy via in situ formation of a functionalized molecule with a dual-terminal coupling function to bind the dissolved polysulfide intermediates, thus turning them back into solid-state organopolysulfide complexes by molecule binding, and then the polysulfides can be pinned on the cathode firmly. The dual-terminal coupling functional molecule binder (MB), which is formed in situ by reaction between quinhydrone (QH) and lithium, can not only bind polysulfides by reversible chemical coordination but also promote the conversion of polysulfides during cycling synchronously. In theory, with the dual-terminal coupling function, MB can bind polysulfide intermediates to copolymerize them, forming −[MB-Li2Sn]– that has faster reaction activity and redox conversion kinetics in comparison with simple Li2Sn. With the MB, the Li–S battery exhibits a large initial capacity of 1347 mAh g–1 at 0.1 C. The remaining capacity of 963 mAh g–1 at 1 C shows no obvious decay for more than 400 cycles, and the retention of the first 300 cycles can reach 96.9%, in particular. This study delivers an alternative approach to resolving the shuttle effect and achieving excellent Li–S battery performance, with the potential significance going way beyond battery systems.

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In the <emphasis emphasistype="bold"><emphasis emphasistype="smcaps">IEEE Sensors Journal</emphasis></emphasis>, vol. 7, no. 8, pp. 1102–1109, 2007, we presented an analyzing method for a threshold accelerometer based on the fully compliant bistable mechanism composed of post-buckling beams with both ends fixed. The critical buckling phenomenon and bistability of the compliant buckling structures are considerably suitable for mechanically sensing the threshold acceleration, and eliminating the electrical power needed for retaining the stable state. Based on the bistability of the fully compliant clamped-clamped mechanism, a threshold accelerometer which can sense the threshold deceleration (cutoff acceleration) after being triggered is designed and fabricated in this paper. The accelerometer mainly consists of a contact, an inertial mass, and two parallel clamped-clamped beams with a length of <formula formulatype="inline"> <tex Notation="TeX">$L$</tex></formula> and an initial angle <formula formulatype="inline"> <tex Notation="TeX">$\alpha$</tex></formula>. A mini rotating test bench is designed to test the sensing capability of threshold acceleration. The threshold acceleration and deceleration for triggering and cutting-off the accelerometer are 9.7 g and <formula formulatype="inline"><tex Notation="TeX">${-}3.1$</tex> </formula> g in experiments, which deviate from the design values of 10.0 g and <formula formulatype="inline"><tex Notation="TeX">${-}3.0$</tex></formula> g with only 3.0% and 3.33%, respectively. The capabilities of accurate threshold sensing and stable state retaining validate the feasibility of introducing the post-buckling compliant beams in designing threshold accelerometers. And, the accordance in the five experiment results also shows that the repeatability and structural simplicity of the designed threshold accelerometer ensure its reliability and applicability in engineering. </para>

Augmented Intelligence of Things (AIoT) combines augmented intelligence algorithms with the massive data collected by IoT devices, enabling more advanced decision-making. The typical application of AIoT is edge computing (EC), which provides computational and storage resources at the edge to support vehicle decision making for computation tasks. With the development of EC, task offloading has become a hopeful paradigm for assisting the time-sensitive tasks of resource-limited vehicles, such as emergency rescue vehicles by deploying at roadside units (RSUs). However, the effectiveness of task offloading for emergency vehicles is hindered by the timeliness of trajectory data and the concern regarding vehicle location. Therefore, this study introduces a secure task offloading scheme relying on the real-time trajectory prediction, named STODRL. First, this study proposes a temporal differential privacy method to disturb vehicular location information to avoid suffering from malicious stealing. Second, a vehicular trajectory prediction method using the temporal convolutional network (TCN) is designed to improve the task offloading precision by offering supplemental trajectory information. Moreover, the scheme employs a reinforcement learning method to offload computational requests effectively and avoid dimensional disasters. Simulated results validate that the STODRL outperforms the existing methods, significantly reducing task completion delays and ensuring the security of location information.

An organo-cobalt coordination complex (Co-H4APD), based on phosphonitrile-azacycles, was prepared by hydrothermal method. The flame retardancy, smoke suppression and thermal stability of epoxy (EP) composites were investigate by means of limited oxygen index (LOI), cone calorimeter test (CONE) and thermogravimetric analysis (TGA). The flame retardant modes of action of Co-H4APD in EP were confirmed by experiments, such as thermogravimetry-Fourier transform infrared spectroscopy-gas chromatograph/mass spectrometer (TG-FTIR-GC/MS), exploring condense and gas-phase products after composites pyrolysis or combustion. Results revealed that the introduction of 6 wt.% Co-H4APD increased LOI value to 29.8% and effectively suppressed heat/smoke release of EP composites. The synergistic charring effect of Co-H4APD improved the thermal stability and char-forming ability of composites. The char strength may be well-correlated with gas release for EP/Co-H4APD, conducive to form dense and regularly expanded char layer, with more phosphorus-rich graphitic structures and cross-linking structures catalyzed by cobalt ions. This high-quality char layer was regarded as the most critical side in improving the flame retardant and smoke suppression performance of EP composites. The gas-phase function of Co-H4APD should not be overlooked due to releasing phosphorous-based radicals during pyrolysis, exhibiting flame inhibition effect in gas phase. More efficient interactions between phosphazene and cobalt within one molecule unit of Co-H4APD contributed to its more obvious reduction of the combustion and smoke production than those of the physical mixing system of CoO + H4APD.

To solve those traffic accidents during conscious lane change of vehicles on highway under dangerous conditions, a new safety lane change model is established on the basis of the lane departure warning system which Jilin University has developed. This model is studied under a typical scenario on highway, and also considered with the actual driver behavior that most vehicles always accelerate during lane changing process. Simulation software is developed to testify the model performance based on Matlab7.0. The results show that the new lane change model can offer certain technical foundation for actualization of safety lane change system of vehicle assistant driving on highway.

In this paper, a multi-agent deep reinforcement learning method was adopted to realize cooperative spectrum sensing in cognitive radio networks. Each secondary user learns an efficient sensing strategy from the sensing results of some of the selected spectra to avoid interference to the primary users and to coordinate with other secondary users. It is necessary to balance exploration and exploitation in the learning process when using deep reinforcement learning methods, helping explain that upper confidence bound with Hoeffding-style bonus has been adopted in this paper to improve the efficiency of exploration. The simulation results verify that the proposed algorithm, when compared with the conventional reinforcement learning methods with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\varepsilon $ </tex-math></inline-formula> -greedy exploration, is much easier to achieve faster convergence speed and better reward performance.

Self-supervised representation learning (SSL) typically suffers from inadequate data utilization and feature-specificity due to the suboptimal sampling strategy and the monotonous optimization method. Existing contrastive-based methods alleviate these issues through exceedingly long training time and large batch size, resulting in non-negligible computational consumption and memory usage. In this paper, we present an efficient self-supervised framework, called GLNet. The key insights of this work are the novel sampling and ensemble learning strategies embedded in the self-supervised framework. We first propose a location-based sampling strategy to integrate the complementary advantages of semantic and spatial characteristics. Whereafter, a Siamese network with momentum update is introduced to generate representative vectors, which are used to optimize the feature extractor. Finally, we particularly embed global contrastive and local location tasks in the framework, which aims to leverage the complementarity between the high-level semantic features and low-level texture features. Such complementarity is significant for mitigating the feature-specificity and improving the generalizability, thus effectively improving the performance of downstream tasks. Extensive experiments on representative benchmark datasets demonstrate that GLNet performs favorably against the state-of-the-art SSL methods. Specifically, GLNet improves MoCo-v3 by 2.4% accuracy on ImageNet dataset, while improves 2% accuracy and consumes only 75% training time on the ImageNet-100 dataset. In addition, GLNet is appealing in its compatibility with popular SSL frameworks. Code is available at GLNet.

In view of the low detection efficiency and high missed detection rate in the current printed circuit board (PCB), this paper proposes an improved YOLOv3 PCB surface defect detection method. This method is based on the YOLOv3 network model. The improvement of its network structure mainly includes: 1. Combine the batch normalization (BN, Batch Normalization) layer to the convolutional layer, improve the forward reasoning speed of the model, and reduce the model's PCB defects the training time of the dataset. 2. Aiming at the problem that the objective function and evaluation metric are not uniform in the YOLOv3 object detection algorithm, the GIoU performance metric and loss function are used to improve the detection effect of the model on small and medium targets of PCB defects. 3. Use the K-means++ clustering algorithm to optimize the K-means clustering algorithm, and determine the appropriate anchor boxes for the PCB defect dataset. 4. Multiscale training is used to enhance the robustness of the model for image detection with different resolutions. The experimental results show that mAP (Mean Average Precision) reaches 92.13%, and the detection rate is increased to 63f/s, which is improved compared to the YOLOv3 model, and has a better application prospect in PCB surface defect detection.

Modbus over TCP/IP is one of the most popular industrial network protocol that are widely used in critical infrastructures. However, vulnerability of Modbus TCP protocol has attracted widely concern in the public. The traditional intrusion detection methods can identify some intrusion behaviors, but there are still some problems. In this paper, we present an innovative approach, SD-IDS (Stereo Depth IDS), which is designed for perform real-time deep inspection for Modbus TCP traffic. SD-IDS algorithm is composed of two parts: rule extraction and deep inspection. The rule extraction module not only analyzes the characteristics of industrial traffic, but also explores the semantic relationship among the key field in the Modbus TCP protocol. The deep inspection module is based on rule-based anomaly intrusion detection. Furthermore, we use the online test to evaluate the performance of our SD-IDS system. Our approach get a low rate of false positive and false negative.