With the rapid development of 5G NR(New Radio),the explosive increment of traffic amount is calling the utilization of unlicensed band.3GPP has proposed LAA(Licensed Assisted Access)to use LTE in unlicensed band and p...With the rapid development of 5G NR(New Radio),the explosive increment of traffic amount is calling the utilization of unlicensed band.3GPP has proposed LAA(Licensed Assisted Access)to use LTE in unlicensed band and pointed out that NR-U(NR-Unlicensed)can reuse most designs of it.However,the existing channel access mechanism of LAA is conservative under the coexistence scenario of NR-U,which leads to the waste of time resource.To address the problem this paper proposes a hybrid channel access mechanism to take advantage of the LBT(Listen-Before-Talk)mechanism of LAA when channel is quite busy and transmit directly with reduced power when it is relatively idle.The channel busy degree is judged by a series of periodically updated adaptive thresholds.System-level simulation verifies that under the coexistence scenario of NR-U the proposed mechanism can achieve higher UPT(User Perceived Throughput)and lower delay than other channel access mechanisms.展开更多
In many existing multi-view gait recognition methods based on images or video sequences,gait sequences are usually used to superimpose and synthesize images and construct energy-like template.However,information may b...In many existing multi-view gait recognition methods based on images or video sequences,gait sequences are usually used to superimpose and synthesize images and construct energy-like template.However,information may be lost during the process of compositing image and capture EMG signals.Errors and the recognition accuracy may be introduced and affected respectively by some factors such as period detection.To better solve the problems,a multi-view gait recognition method using deep convolutional neural network and channel attention mechanism is proposed.Firstly,the sliding time window method is used to capture EMG signals.Then,the back-propagation learning algorithm is used to train each layer of convolution,which improves the learning ability of the convolutional neural network.Finally,the channel attention mechanism is integrated into the neural network,which will improve the ability of expressing gait features.And a classifier is used to classify gait.As can be shown from experimental results on two public datasets,OULP and CASIA-B,the recognition rate of the proposed method can be achieved at 88.44%and 97.25%respectively.As can be shown from the comparative experimental results,the proposed method has better recognition effect than several other newer convolutional neural network methods.Therefore,the combination of convolutional neural network and channel attention mechanism is of great value for gait recognition.展开更多
The propagation mechanism of steady cellular detonations in curved channels is investigated numerically with a detailed chemical reaction mechanism, The numerical results demonstrate that as the radius of the curvatur...The propagation mechanism of steady cellular detonations in curved channels is investigated numerically with a detailed chemical reaction mechanism, The numerical results demonstrate that as the radius of the curvature decreases, detonation fails near the inner wall due to the strong expansion effect. As the radius of the curvature increases, the detonation front near the inner wall can sustain an underdriven detonation. In the case where deto- nation fails, a transverse detonation downstream forms and re-initiates the quenched detonation as it propagates toward the inner wall. Two kinds of propagation modes exist as the detonation is propagating in the curved channel. One is that the detonation fails first, and then a following transverse detonation initiates the quenched detonation and this process repeats itself. The other one is that without detonation failure and re-initiation, a steady detonation exists which consists of an underdriven detonation front near the inner wall subject to the diffraction and an overdriven detonation near the outer wall subject to the compression.展开更多
Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to ach...Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to achieve better results in VQA tasks.Analysis of all features may cause information redundancy and heavy computational burden.Attention mechanism is a wise way to solve this problem.However,using single attention mechanism may cause incomplete concern of features.This paper improves the attention mechanism method and proposes a hybrid attention mechanism that combines the spatial attention mechanism method and the channel attention mechanism method.In the case that the attention mechanism will cause the loss of the original features,a small portion of image features were added as compensation.For the attention mechanism of text features,a selfattention mechanism was introduced,and the internal structural features of sentences were strengthened to improve the overall model.The results show that attention mechanism and feature compensation add 6.1%accuracy to multimodal low-rank bilinear pooling network.展开更多
Vehicle detection plays a crucial role in the field of autonomous driving technology.However,directly applying deep learning-based object detection algorithms to complex road scene images often leads to subpar perform...Vehicle detection plays a crucial role in the field of autonomous driving technology.However,directly applying deep learning-based object detection algorithms to complex road scene images often leads to subpar performance and slow inference speeds in vehicle detection.Achieving a balance between accuracy and detection speed is crucial for real-time object detection in real-world road scenes.This paper proposes a high-precision and fast vehicle detector called the feature-guided bidirectional pyramid network(FBPN).Firstly,to tackle challenges like vehicle occlusion and significant background interference,the efficient feature filtering module(EFFM)is introduced into the deep network,which amplifies the disparities between the features of the vehicle and the background.Secondly,the proposed global attention localization module(GALM)in the model neck effectively perceives the detailed position information of the target,improving both the accuracy and inference speed of themodel.Finally,the detection accuracy of small-scale vehicles is further enhanced through the utilization of a four-layer feature pyramid structure.Experimental results show that FBPN achieves an average precision of 60.8% and 97.8% on the BDD100K and KITTI datasets,respectively,with inference speeds reaching 344.83 frames/s and 357.14 frames/s.FBPN demonstrates its effectiveness and superiority by striking a balance between detection accuracy and inference speed,outperforming several state-of-the-art methods.展开更多
Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,...Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.展开更多
Based on the quantum information theory, we have investigated the entropy squeezing of a moving two-level atom interacting with the coherent field via the quantum mechanical channel of the two-photon process. The resu...Based on the quantum information theory, we have investigated the entropy squeezing of a moving two-level atom interacting with the coherent field via the quantum mechanical channel of the two-photon process. The results are compared with those of atomic squeezing based on the Heisenberg uncertainty relation. The influences of the atomic motion and field-mode structure parameter on the atomic entropy squeezing and on the control of noise of the quantum mechanical channel via the two-photon process are examined. Our results show that the squeezed period, duration of optimal entropy squeezing of a two-level atom and the noise of the quantum mechanical channel can be controlled by appropriately choosing the atomic motion and the field-mode structure parameter, respectively. The quantum mechanical channel of two-photon process is an ideal channel for quantum information (atomic quantum state) transmission. Quantum information entropy is a remarkably accurate measure of the atomic squeezing.展开更多
For the past few decades,the internet of underwater things(IoUT)otained a lot of attention in mobile aquatic applications such as oceanography,diver network monitoring,unmanned underwater exploration,underwater survei...For the past few decades,the internet of underwater things(IoUT)otained a lot of attention in mobile aquatic applications such as oceanography,diver network monitoring,unmanned underwater exploration,underwater surveillance,location tracking system,etc.Most of the IoUT applications rely on acoustic medium.The current IoUT applications face difficulty in delivering a reliable communication system due to the various technical limitations of IoUT environment such as low data rate,attenuation,limited bandwidth,limited battery,limited memory,connectivity problem,etc.One of the significant applications of IoUT include monitoring underwater diver networks.In order to perform a reliable and energy-efficient communication system in the underwater diver networks,a smart underwater hybrid softwaredefined modem(UHSDM)for the mobile ad-hoc network was developed that is used for selecting the best channel/medium among acoustic,visible light communication(VLC),and infrared(IR)based on the criteria established within the system.However,due to the mobility of underwater divers,the developed UHSDMmeets the challenges such as connectivity errors,frequent link failure,transmission delay caused by re-routing,etc.During emergency,the divers are most at the risk of survival.To deal with diver mobility,connectivity,energy efficiency,and reducing the latency in ADN,a handover mechanism based on pre-built UHSDM is proposed in this paper.This paper focuses on(1)design of UHSDM for ADN(2)propose the channel selection mechanism in UHSDM for selecting the best medium for handover and(3)propose handover protocol inADN.The implementation result shows that the proposed mechanism can be used to find the new route for divers in advance and the latency can be reduced significantly.Additionally,this paper shows the real field experiment of air tests and underwater tests with various distances.This research will contribute much to the profit of researchers in underwater diver networks and underwater networks,for improving the quality of services(QoS)of underwater applications.展开更多
A two-step equal channel angular extrusion(ECAE) procedure was used to process pure Mg. The effects of ECAE processing temperature on the microstructure, mechanical properties, and corrosion behavior of pure Mg were...A two-step equal channel angular extrusion(ECAE) procedure was used to process pure Mg. The effects of ECAE processing temperature on the microstructure, mechanical properties, and corrosion behavior of pure Mg were studied. The results show that the average grain size of pure Mg decreases with decreasing extrusion temperature. After ECAE processing at 473 K, fine and equiaxed grains(~9 μm) are obtained. The sample processed at 473 K exhibits the excellent mechanical properties, whereas the sample processed at 633 K has the lowest corrosion rate. The improved corrosion resistance and mechanical properties of pure Mg by ECAE are ascribed to grain refinement and microstructural modification.展开更多
To address the problems of low detection accuracy and slow speed of traditional vision in the pharmaceutical industry,a YOLOv5s-EBD defect detection algorithm:Based on YOLOv5 network,firstly,the channel attention mech...To address the problems of low detection accuracy and slow speed of traditional vision in the pharmaceutical industry,a YOLOv5s-EBD defect detection algorithm:Based on YOLOv5 network,firstly,the channel attention mechanism is introduced into the network to focus the network on defects similar to the pill background,reducing the time-consuming scanning of invalid backgrounds;the PANet module in the network is then replaced with BiFPN for differential fusion of different features;finally,Depth-wise separable convolution is used instead of standard convolution to achieve the output Finally,Depth-wise separable convolution is used instead of standard convolution to achieve the output feature map requirements of standard convolution with less number of parameters and computation,and improve detection speed.the improved model is able to detect all types of defects in tablets with an accuracy of over 94%and a detection speed of 123.8 fps,which is 4.27%higher than the unimproved YOLOv5 network model with 5.2 fps.展开更多
Based on the measured results that wall pressure fluctuations are mainly de- cided by coherent structures of turbulence, the relationship between root-mean- square wall pressure and wall shear stress in turbulent shea...Based on the measured results that wall pressure fluctuations are mainly de- cided by coherent structures of turbulence, the relationship between root-mean- square wall pressure and wall shear stress in turbulent shear flow and that between the intensities of pressure and fluctuating velocity in homogeneous and isotropic turbulence are established in this paper. These relationships are consistent with former works, and have good agreement with experimental data. The paper also dis- cusses the concept of 'apparent pressure' on the wall in mean flow.展开更多
Ryanodine receptors (RyR) are the major Ca2+ release channels in both cardiac and skeletal muscle, they play a crucial role in the Ca2+ signaling pathway that govern the
To improve the performance of the forest fire smoke detection model and achieve a better balance between detection accuracy and speed, an improved YOLOv4 detection model (MoAm-YOLOv4) that combines a lightweight netwo...To improve the performance of the forest fire smoke detection model and achieve a better balance between detection accuracy and speed, an improved YOLOv4 detection model (MoAm-YOLOv4) that combines a lightweight network and attention mechanism was proposed. Based on the YOLOv4 algorithm, the backbone network CSPDarknet53 was replaced with a lightweight network MobilenetV1 to reduce the model’s size. An attention mechanism was added to the three channels before the output to increase its ability to extract forest fire smoke effectively. The algorithm used the K-means clustering algorithm to cluster the smoke dataset, and obtained candidate frames that were close to the smoke images;the dataset was expanded to 2000 images by the random flip expansion method to avoid overfitting in training. The experimental results show that the improved YOLOv4 algorithm has excellent detection effect. Its mAP can reach 93.45%, precision can get 93.28%, and the model size is only 45.58 MB. Compared with YOLOv4 algorithm, MoAm-YOLOv4 improves the accuracy by 1.3% and reduces the model size by 80% while sacrificing only 0.27% mAP, showing reasonable practicability.展开更多
The field of finance heavily relies on cybersecurity to safeguard its systems and clients from harmful software.The identification of malevolent code within financial software is vital for protecting both the financia...The field of finance heavily relies on cybersecurity to safeguard its systems and clients from harmful software.The identification of malevolent code within financial software is vital for protecting both the financial system and individual clients.Nevertheless,present detection models encounter limitations in their ability to identify malevolent code and its variations,all while encompassing a multitude of parameters.To overcome these obsta-cles,we introduce a lean model for classifying families of malevolent code,formulated on Ghost-DenseNet-SE.This model integrates the Ghost module,DenseNet,and the squeeze-and-excitation(SE)channel domain attention mechanism.It substitutes the standard convolutional layer in DenseNet with the Ghost module,thereby diminishing the model’s size and augmenting recognition speed.Additionally,the channel domain attention mechanism assigns distinctive weights to feature channels,facilitating the extraction of pivotal characteristics of malevolent code and bolstering detection precision.Experimental outcomes on the Malimg dataset indicate that the model attained an accuracy of 99.14%in discerning families of malevolent code,surpassing AlexNet(97.8%)and The visual geometry group network(VGGNet)(96.16%).The proposed model exhibits reduced parameters,leading to decreased model complexity alongside enhanced classification accuracy,rendering it a valuable asset for categorizing malevolent code.展开更多
In the presence of an MMC-HVDC system,current differential protection(CDP)has the risk of failure in operation under an internal fault.In addition,CDP may also incur security issues in the presence of current transfor...In the presence of an MMC-HVDC system,current differential protection(CDP)has the risk of failure in operation under an internal fault.In addition,CDP may also incur security issues in the presence of current transformer(CT)saturation and outliers.In this paper,a current trajectory image-based protection algorithm is proposed for AC lines connected to MMC-HVDC stations using a convolution neural network improved by a channel attention mechanism(CA-CNN).Taking the dual differential currents as two-dimensional coordinates of the moving point,the moving-point trajectories formed by differential currents have significant differences under internal and external faults.Therefore,internal faults can be identified using image recognition based on CA-CNN.This is improved by a channel attention mechanism,data augmentation,and adaptive learning rate.In comparison with other machine learning algorithms,the feature extraction ability and accuracy of CA-CNN are greatly improved.Various fault conditions like different net-work structures,operation modes,fault resistances,outliers,and current transformer saturation,are fully considered to verify the superiority of the proposed protection algorithm.The results confirm that the proposed current trajectory image-based protection algorithm has strong learning and generalizability,and can identify internal faults reliably.展开更多
With the rapid development of information technology,information system security and insider threat detection have become important topics for organizational management.In the current network environment,user behavior...With the rapid development of information technology,information system security and insider threat detection have become important topics for organizational management.In the current network environment,user behavioral bio-data presents the characteristics of nonlinearity and temporal sequence.Most of the existing research on authentication based on user behavioral biometrics adopts the method of manual feature extraction.They do not adequately capture the nonlinear and time-sequential dependencies of behavioral bio-data,and also do not adequately reflect the personalized usage characteristics of users,leading to bottlenecks in the performance of the authentication algorithm.In order to solve the above problems,this paper proposes a Temporal Convolutional Network method based on an Efficient Channel Attention mechanism(ECA-TCN)to extract user mouse dynamics features and constructs an one-class Support Vector Machine(OCSVM)for each user for authentication.Experimental results show that compared with four existing deep learning algorithms,the method retains more adequate key information and improves the classification performance of the neural network.In the final authentication,the Area Under the Curve(AUC)can reach 96%.展开更多
基金the Project “Evaluation and verification of candidate solutions for international standardization of 5G” supported by National Science and Technology Major Project of the Ministry of Science and Technology (2018ZX03001024-006)
文摘With the rapid development of 5G NR(New Radio),the explosive increment of traffic amount is calling the utilization of unlicensed band.3GPP has proposed LAA(Licensed Assisted Access)to use LTE in unlicensed band and pointed out that NR-U(NR-Unlicensed)can reuse most designs of it.However,the existing channel access mechanism of LAA is conservative under the coexistence scenario of NR-U,which leads to the waste of time resource.To address the problem this paper proposes a hybrid channel access mechanism to take advantage of the LBT(Listen-Before-Talk)mechanism of LAA when channel is quite busy and transmit directly with reduced power when it is relatively idle.The channel busy degree is judged by a series of periodically updated adaptive thresholds.System-level simulation verifies that under the coexistence scenario of NR-U the proposed mechanism can achieve higher UPT(User Perceived Throughput)and lower delay than other channel access mechanisms.
基金This work was supported by the Natural Science Foundation of China(No.61902133)Fujian natural science foundation project(No.2018J05106)Xiamen Collaborative Innovation projects of Produces study grinds(3502Z20173046)。
文摘In many existing multi-view gait recognition methods based on images or video sequences,gait sequences are usually used to superimpose and synthesize images and construct energy-like template.However,information may be lost during the process of compositing image and capture EMG signals.Errors and the recognition accuracy may be introduced and affected respectively by some factors such as period detection.To better solve the problems,a multi-view gait recognition method using deep convolutional neural network and channel attention mechanism is proposed.Firstly,the sliding time window method is used to capture EMG signals.Then,the back-propagation learning algorithm is used to train each layer of convolution,which improves the learning ability of the convolutional neural network.Finally,the channel attention mechanism is integrated into the neural network,which will improve the ability of expressing gait features.And a classifier is used to classify gait.As can be shown from experimental results on two public datasets,OULP and CASIA-B,the recognition rate of the proposed method can be achieved at 88.44%and 97.25%respectively.As can be shown from the comparative experimental results,the proposed method has better recognition effect than several other newer convolutional neural network methods.Therefore,the combination of convolutional neural network and channel attention mechanism is of great value for gait recognition.
基金Supported by the National Natural Science Foundation of China under Grant Nos 11390363 and 11325209
文摘The propagation mechanism of steady cellular detonations in curved channels is investigated numerically with a detailed chemical reaction mechanism, The numerical results demonstrate that as the radius of the curvature decreases, detonation fails near the inner wall due to the strong expansion effect. As the radius of the curvature increases, the detonation front near the inner wall can sustain an underdriven detonation. In the case where deto- nation fails, a transverse detonation downstream forms and re-initiates the quenched detonation as it propagates toward the inner wall. Two kinds of propagation modes exist as the detonation is propagating in the curved channel. One is that the detonation fails first, and then a following transverse detonation initiates the quenched detonation and this process repeats itself. The other one is that without detonation failure and re-initiation, a steady detonation exists which consists of an underdriven detonation front near the inner wall subject to the diffraction and an overdriven detonation near the outer wall subject to the compression.
基金This work was supported by the Sichuan Science and Technology Program(2021YFQ0003).
文摘Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to achieve better results in VQA tasks.Analysis of all features may cause information redundancy and heavy computational burden.Attention mechanism is a wise way to solve this problem.However,using single attention mechanism may cause incomplete concern of features.This paper improves the attention mechanism method and proposes a hybrid attention mechanism that combines the spatial attention mechanism method and the channel attention mechanism method.In the case that the attention mechanism will cause the loss of the original features,a small portion of image features were added as compensation.For the attention mechanism of text features,a selfattention mechanism was introduced,and the internal structural features of sentences were strengthened to improve the overall model.The results show that attention mechanism and feature compensation add 6.1%accuracy to multimodal low-rank bilinear pooling network.
基金funded by Ministry of Science and Technology of the People’s Republic of China,Grant Numbers 2022YFC3800502Chongqing Science and Technology Commission,Grant Number cstc2020jscx-dxwtBX0019,CSTB2022TIAD-KPX0118,cstc2020jscx-cylhX0005 and cstc2021jscx-gksbX0058.
文摘Vehicle detection plays a crucial role in the field of autonomous driving technology.However,directly applying deep learning-based object detection algorithms to complex road scene images often leads to subpar performance and slow inference speeds in vehicle detection.Achieving a balance between accuracy and detection speed is crucial for real-time object detection in real-world road scenes.This paper proposes a high-precision and fast vehicle detector called the feature-guided bidirectional pyramid network(FBPN).Firstly,to tackle challenges like vehicle occlusion and significant background interference,the efficient feature filtering module(EFFM)is introduced into the deep network,which amplifies the disparities between the features of the vehicle and the background.Secondly,the proposed global attention localization module(GALM)in the model neck effectively perceives the detailed position information of the target,improving both the accuracy and inference speed of themodel.Finally,the detection accuracy of small-scale vehicles is further enhanced through the utilization of a four-layer feature pyramid structure.Experimental results show that FBPN achieves an average precision of 60.8% and 97.8% on the BDD100K and KITTI datasets,respectively,with inference speeds reaching 344.83 frames/s and 357.14 frames/s.FBPN demonstrates its effectiveness and superiority by striking a balance between detection accuracy and inference speed,outperforming several state-of-the-art methods.
基金supported by the National Natural Science Foundation of China(61877067)the Foundation of Science and Technology on Near-Surface Detection Laboratory(TCGZ2019A002,TCGZ2021C003,6142414200511)the Natural Science Basic Research Program of Shaanxi(2021JZ-19)。
文摘Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.
基金Project supported by the National Natural Science Foundation of China (Grant No 10374025), the Natural Science Foundation of Hunan Province, China (Grant No 05JJ30004) and the Scientific Research Fund of Hunan Provincial Education Department, China (Grant No 03c543)
文摘Based on the quantum information theory, we have investigated the entropy squeezing of a moving two-level atom interacting with the coherent field via the quantum mechanical channel of the two-photon process. The results are compared with those of atomic squeezing based on the Heisenberg uncertainty relation. The influences of the atomic motion and field-mode structure parameter on the atomic entropy squeezing and on the control of noise of the quantum mechanical channel via the two-photon process are examined. Our results show that the squeezed period, duration of optimal entropy squeezing of a two-level atom and the noise of the quantum mechanical channel can be controlled by appropriately choosing the atomic motion and the field-mode structure parameter, respectively. The quantum mechanical channel of two-photon process is an ideal channel for quantum information (atomic quantum state) transmission. Quantum information entropy is a remarkably accurate measure of the atomic squeezing.
基金This research was a part of the project titled“Development of the wide-area underwater mobile communication systems”funded by the Ministry of Oceans and Fisheries,Korea.
文摘For the past few decades,the internet of underwater things(IoUT)otained a lot of attention in mobile aquatic applications such as oceanography,diver network monitoring,unmanned underwater exploration,underwater surveillance,location tracking system,etc.Most of the IoUT applications rely on acoustic medium.The current IoUT applications face difficulty in delivering a reliable communication system due to the various technical limitations of IoUT environment such as low data rate,attenuation,limited bandwidth,limited battery,limited memory,connectivity problem,etc.One of the significant applications of IoUT include monitoring underwater diver networks.In order to perform a reliable and energy-efficient communication system in the underwater diver networks,a smart underwater hybrid softwaredefined modem(UHSDM)for the mobile ad-hoc network was developed that is used for selecting the best channel/medium among acoustic,visible light communication(VLC),and infrared(IR)based on the criteria established within the system.However,due to the mobility of underwater divers,the developed UHSDMmeets the challenges such as connectivity errors,frequent link failure,transmission delay caused by re-routing,etc.During emergency,the divers are most at the risk of survival.To deal with diver mobility,connectivity,energy efficiency,and reducing the latency in ADN,a handover mechanism based on pre-built UHSDM is proposed in this paper.This paper focuses on(1)design of UHSDM for ADN(2)propose the channel selection mechanism in UHSDM for selecting the best medium for handover and(3)propose handover protocol inADN.The implementation result shows that the proposed mechanism can be used to find the new route for divers in advance and the latency can be reduced significantly.Additionally,this paper shows the real field experiment of air tests and underwater tests with various distances.This research will contribute much to the profit of researchers in underwater diver networks and underwater networks,for improving the quality of services(QoS)of underwater applications.
基金financially supported by the National Natural Science Foundation of China (Nos. 81330031 and 81271701)
文摘A two-step equal channel angular extrusion(ECAE) procedure was used to process pure Mg. The effects of ECAE processing temperature on the microstructure, mechanical properties, and corrosion behavior of pure Mg were studied. The results show that the average grain size of pure Mg decreases with decreasing extrusion temperature. After ECAE processing at 473 K, fine and equiaxed grains(~9 μm) are obtained. The sample processed at 473 K exhibits the excellent mechanical properties, whereas the sample processed at 633 K has the lowest corrosion rate. The improved corrosion resistance and mechanical properties of pure Mg by ECAE are ascribed to grain refinement and microstructural modification.
文摘To address the problems of low detection accuracy and slow speed of traditional vision in the pharmaceutical industry,a YOLOv5s-EBD defect detection algorithm:Based on YOLOv5 network,firstly,the channel attention mechanism is introduced into the network to focus the network on defects similar to the pill background,reducing the time-consuming scanning of invalid backgrounds;the PANet module in the network is then replaced with BiFPN for differential fusion of different features;finally,Depth-wise separable convolution is used instead of standard convolution to achieve the output Finally,Depth-wise separable convolution is used instead of standard convolution to achieve the output feature map requirements of standard convolution with less number of parameters and computation,and improve detection speed.the improved model is able to detect all types of defects in tablets with an accuracy of over 94%and a detection speed of 123.8 fps,which is 4.27%higher than the unimproved YOLOv5 network model with 5.2 fps.
文摘Based on the measured results that wall pressure fluctuations are mainly de- cided by coherent structures of turbulence, the relationship between root-mean- square wall pressure and wall shear stress in turbulent shear flow and that between the intensities of pressure and fluctuating velocity in homogeneous and isotropic turbulence are established in this paper. These relationships are consistent with former works, and have good agreement with experimental data. The paper also dis- cusses the concept of 'apparent pressure' on the wall in mean flow.
文摘Ryanodine receptors (RyR) are the major Ca2+ release channels in both cardiac and skeletal muscle, they play a crucial role in the Ca2+ signaling pathway that govern the
文摘To improve the performance of the forest fire smoke detection model and achieve a better balance between detection accuracy and speed, an improved YOLOv4 detection model (MoAm-YOLOv4) that combines a lightweight network and attention mechanism was proposed. Based on the YOLOv4 algorithm, the backbone network CSPDarknet53 was replaced with a lightweight network MobilenetV1 to reduce the model’s size. An attention mechanism was added to the three channels before the output to increase its ability to extract forest fire smoke effectively. The algorithm used the K-means clustering algorithm to cluster the smoke dataset, and obtained candidate frames that were close to the smoke images;the dataset was expanded to 2000 images by the random flip expansion method to avoid overfitting in training. The experimental results show that the improved YOLOv4 algorithm has excellent detection effect. Its mAP can reach 93.45%, precision can get 93.28%, and the model size is only 45.58 MB. Compared with YOLOv4 algorithm, MoAm-YOLOv4 improves the accuracy by 1.3% and reduces the model size by 80% while sacrificing only 0.27% mAP, showing reasonable practicability.
基金funded by National Natural Science Foundation of China(under Grant No.61905201)。
文摘The field of finance heavily relies on cybersecurity to safeguard its systems and clients from harmful software.The identification of malevolent code within financial software is vital for protecting both the financial system and individual clients.Nevertheless,present detection models encounter limitations in their ability to identify malevolent code and its variations,all while encompassing a multitude of parameters.To overcome these obsta-cles,we introduce a lean model for classifying families of malevolent code,formulated on Ghost-DenseNet-SE.This model integrates the Ghost module,DenseNet,and the squeeze-and-excitation(SE)channel domain attention mechanism.It substitutes the standard convolutional layer in DenseNet with the Ghost module,thereby diminishing the model’s size and augmenting recognition speed.Additionally,the channel domain attention mechanism assigns distinctive weights to feature channels,facilitating the extraction of pivotal characteristics of malevolent code and bolstering detection precision.Experimental outcomes on the Malimg dataset indicate that the model attained an accuracy of 99.14%in discerning families of malevolent code,surpassing AlexNet(97.8%)and The visual geometry group network(VGGNet)(96.16%).The proposed model exhibits reduced parameters,leading to decreased model complexity alongside enhanced classification accuracy,rendering it a valuable asset for categorizing malevolent code.
基金supported in part by the Fundamental Research Funds for the Central Universities under Grant 2022JCCXJD01in part by Training Program of Innovation and Entrepreneurship for Undergraduates of China University of Mining and Technology(Beijing)under Grant 202204009.
文摘In the presence of an MMC-HVDC system,current differential protection(CDP)has the risk of failure in operation under an internal fault.In addition,CDP may also incur security issues in the presence of current transformer(CT)saturation and outliers.In this paper,a current trajectory image-based protection algorithm is proposed for AC lines connected to MMC-HVDC stations using a convolution neural network improved by a channel attention mechanism(CA-CNN).Taking the dual differential currents as two-dimensional coordinates of the moving point,the moving-point trajectories formed by differential currents have significant differences under internal and external faults.Therefore,internal faults can be identified using image recognition based on CA-CNN.This is improved by a channel attention mechanism,data augmentation,and adaptive learning rate.In comparison with other machine learning algorithms,the feature extraction ability and accuracy of CA-CNN are greatly improved.Various fault conditions like different net-work structures,operation modes,fault resistances,outliers,and current transformer saturation,are fully considered to verify the superiority of the proposed protection algorithm.The results confirm that the proposed current trajectory image-based protection algorithm has strong learning and generalizability,and can identify internal faults reliably.
基金supported by the National Natural Science Foundation of China(61962015)the Guangxi Key Laboratory of Cryptography and Information Security Research Project,China(GCIS202127)+2 种基金the Central Guidance on Local Science and Technology Development Fund of Guangxi Province,China(ZY23055008)the Scientific Research and Technological Development Planning Project of Guilin,China(20220124-12)the Innovation Project of Guangxi Graduate Education,China(2023YCXS043).
文摘With the rapid development of information technology,information system security and insider threat detection have become important topics for organizational management.In the current network environment,user behavioral bio-data presents the characteristics of nonlinearity and temporal sequence.Most of the existing research on authentication based on user behavioral biometrics adopts the method of manual feature extraction.They do not adequately capture the nonlinear and time-sequential dependencies of behavioral bio-data,and also do not adequately reflect the personalized usage characteristics of users,leading to bottlenecks in the performance of the authentication algorithm.In order to solve the above problems,this paper proposes a Temporal Convolutional Network method based on an Efficient Channel Attention mechanism(ECA-TCN)to extract user mouse dynamics features and constructs an one-class Support Vector Machine(OCSVM)for each user for authentication.Experimental results show that compared with four existing deep learning algorithms,the method retains more adequate key information and improves the classification performance of the neural network.In the final authentication,the Area Under the Curve(AUC)can reach 96%.