To study the fluid dynamic response mechanism under the working condition of water injection well borehole,based on the microelement analysis of fluid mechanics and the classical theory of hydrodynamics,a fluid microe...To study the fluid dynamic response mechanism under the working condition of water injection well borehole,based on the microelement analysis of fluid mechanics and the classical theory of hydrodynamics,a fluid microelement pressure-flow rate relationship model is built to derive and solve the dynamic distribution of fluid pressure and flow rate in the space of well borehole.Combined with the production data of a typical deviated well in China,numerical simulations and analyses are carried out to analyze the dynamic distribution of wellbore pressure at different injection pressures and injection volumes,the delayed and attenuated characteristics of fluid transmission in tube,and the dynamic distribution of wellbore pressure amplitude under the fluctuation of wellhead pressure.The pressure loss along the wellbore has nothing to do with the absolute pressure,and the design of the coding and decoding scheme for wave code communication doesn’t need to consider the absolute pressure during injecting.When the injection pressure is constant,the higher the injection flow rate at the wellhead,the larger the pressure loss along the wellbore.The fluid wave signal delay amplitude mainly depends on the length of the wellbore.The smaller the tubing diameter,the larger the fluid wave signal attenuation amplitude.The higher the target wave code amplitude(differential pressure identification root mean square)generated at the same well depth,the greater the wellhead pressure wave amplitude required to overcome the wellbore pressure loss.展开更多
In the process of identifying parameters for a permanent magnet synchronous motor,the particle swarm optimization method is prone to being stuck in local optima in the later stages of iteration,resulting in low parame...In the process of identifying parameters for a permanent magnet synchronous motor,the particle swarm optimization method is prone to being stuck in local optima in the later stages of iteration,resulting in low parameter accuracy.This work proposes a fuzzy particle swarm optimization approach based on the transformation function and the filled function.This approach addresses the topic of particle swarmoptimization in parameter identification from two perspectives.Firstly,the algorithm uses a transformation function to change the form of the fitness function without changing the position of the extreme point of the fitness function,making the extreme point of the fitness function more prominent and improving the algorithm’s search ability while reducing the algorithm’s computational burden.Secondly,on the basis of themulti-loop fuzzy control systembased onmultiplemembership functions,it is merged with the filled function to improve the algorithm’s capacity to skip out of the local optimal solution.This approach can be used to identify the parameters of permanent magnet synchronous motors by sampling only the stator current,voltage,and speed data.The simulation results show that the method can effectively identify the electrical parameters of a permanent magnet synchronous motor,and it has superior global convergence performance and robustness.展开更多
With the rapid development of machine learning,the demand for high-efficient computing becomes more and more urgent.To break the bottleneck of the traditional Von Neumann architecture,computing-in-memory(CIM)has attra...With the rapid development of machine learning,the demand for high-efficient computing becomes more and more urgent.To break the bottleneck of the traditional Von Neumann architecture,computing-in-memory(CIM)has attracted increasing attention in recent years.In this work,to provide a feasible CIM solution for the large-scale neural networks(NN)requiring continuous weight updating in online training,a flash-based computing-in-memory with high endurance(10^(9) cycles)and ultrafast programming speed is investigated.On the one hand,the proposed programming scheme of channel hot electron injection(CHEI)and hot hole injection(HHI)demonstrate high linearity,symmetric potentiation,and a depression process,which help to improve the training speed and accuracy.On the other hand,the low-damage programming scheme and memory window(MW)optimizations can suppress cell degradation effectively with improved computing accuracy.Even after 109 cycles,the leakage current(I_(off))of cells remains sub-10pA,ensuring the large-scale computing ability of memory.Further characterizations are done on read disturb to demonstrate its robust reliabilities.By processing CIFAR-10 tasks,it is evident that~90%accuracy can be achieved after 109 cycles in both ResNet50 and VGG16 NN.Our results suggest that flash-based CIM has great potential to overcome the limitations of traditional Von Neumann architectures and enable high-performance NN online training,which pave the way for further development of artificial intelligence(AI)accelerators.展开更多
Detection of maneuvering small targets has always been an important yet challenging task for radar signal processing.One primary reason is that target variable motions within coherent processing interval generate ener...Detection of maneuvering small targets has always been an important yet challenging task for radar signal processing.One primary reason is that target variable motions within coherent processing interval generate energy migrations across multiple resolution bins,which severely deteriorate the parameter estimation performance.A coarse-to-fine strategy for the detection of maneuvering small targets is proposed.Integration of small points segmented coherently is performed first,and then an optimal inter-segment integration is utilized to derive the coarse estimation of the chirp rate.Sparse fractional Fourier transform(FrFT)is then employed to refine the coarse estimation at a significantly reduced computational complexity.Simulation results verify the proposed scheme that achieves an efficient and reliable maneuvering target detection with-16dB input signal-to-noise ratio(SNR),while requires no exact a priori knowledge on the motion parameters.展开更多
In converged heterogeneous wireless networks, vertical handoff is an important issue in radio resource management and occurs when an end user switches from one network to another (e.g., from wireless local area networ...In converged heterogeneous wireless networks, vertical handoff is an important issue in radio resource management and occurs when an end user switches from one network to another (e.g., from wireless local area network to wideband code division multiple access). Efficient vertical handoff should allocate network resources efficiently and maintain good quality of service (QoS) for the end users. The objective of this work is to determine conditions under which vertical handoff can be performed. The channel usage situation of each access network is formulated as a birth-death process with the objective of predicting the available bandwidth and the blocking probability. A reward function is used to capture the network bandwidth and the blocking probability is expressed as a cost function. An end user will access the certain network which maximizes the total function defined as the combination of the reward function and the cost function. Simulation results show that the proposed algorithm can significantly improve the network performance, including higher bandwidth for end users and lower new call blocking and handoff call blocking probability for networks.展开更多
This paper proposes a model of neural networks consisting of populations of perceptive neurons, inter-neurons, and motor neurons according to the theory of stochastic phase resetting dynamics. According to this model,...This paper proposes a model of neural networks consisting of populations of perceptive neurons, inter-neurons, and motor neurons according to the theory of stochastic phase resetting dynamics. According to this model, the dynamical characteristics of neural networks are studied in three coupling cases, namely, series and parallel coupling, series coupling, and unilateral coupling. The results show that the indentified structure of neural networks enables the basic characteristics of neural information processing to be described in terms of the actions of both the optional motor and the reflected motor. The excitation of local neural networks is caused by the action of the optional motor. In particular, the excitation of the neural population caused by the action of the optional motor in the motor cortex is larger than that caused by the action of the reflected motor. This phenomenon indicates that there are more neurons participating in the neural information processing and the excited synchronization motion under the action of the optional motor.展开更多
For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce i...For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce indicators of the capability according to the theory of rough set. The new indicator system can be determined. Attribute reduction can also reduce the workload and remove the redundant information,when there are too many indicators or the indicators have strong correlation. The research complexity can be reduced and the efficiency can be improved. Entropy weighting method is used to determine the weights of the remaining indicators,and the importance of indicators is analyzed. The information fusion model based on nearest neighbor method is developed and utilized to evaluate the capability of multiple agent coalitions,compared to cloud evaluation model and D-S evidence method. Simulation results are reasonable and with obvious distinction. Thus they verify the effectiveness and feasibility of the model. The information fusion model can provide more scientific,rational decision support for choosing the best agent coalition,and provide innovative steps for the evaluation process of capability of agent coalitions.展开更多
The general frame for the system of intelligence science was proposed, the common features of the researching objects of the intelligence science were summarized. The intelligence science consists of three portions: s...The general frame for the system of intelligence science was proposed, the common features of the researching objects of the intelligence science were summarized. The intelligence science consists of three portions: scientific foundation, technical methodology and application fields. The common features of intelligence science include complexity, intersection, nonlinearity, anthropomorphic property, uncertainty, incompleteness and distribution etc. The new proposed scientific branch would reflect the new height, new thought and new way for developing the control science and intelligent systems from one angle, and present a strong wish for establishing a new branch of intelligence science.展开更多
In this research, a near infrared multi-wavelength noninvasive blood glucose monitoring system with distributed laser multi-sensors is applied to monitor human blood glucose concentration. In order to improve the moni...In this research, a near infrared multi-wavelength noninvasive blood glucose monitoring system with distributed laser multi-sensors is applied to monitor human blood glucose concentration. In order to improve the monitoring accuracy, a multi-sensors information fusion model based on Back Propagation Artificial Neural Network is proposed. The Root- Mean-Square Error of Prediction for noninvasive blood glucose measurement is 0.088mmol/L, and the correlation coefficient is 0.94. The noninvasive blood glucose monitoring system based on distributed multi-sensors information fusion of multi-wavelength NIR is proved to be of great efficient. And the new proposed idea of measurement based on distri- buted multi-sensors, shows better prediction accuracy.展开更多
ChatG PT,an artificial intelligence generated content (AIGC) model developed by OpenAI,has attracted worldwide attention for its capability of dealing with challenging language understanding and generation tasks in th...ChatG PT,an artificial intelligence generated content (AIGC) model developed by OpenAI,has attracted worldwide attention for its capability of dealing with challenging language understanding and generation tasks in the form of conversations.This paper briefly provides an overview on the history,status quo and potential future development of ChatGPT,helping to provide an entry point to think about ChatGPT.Specifically,from the limited open-accessed resources,we conclude the core techniques of ChatGPT,mainly including large-scale language models,in-context learning,reinforcement learning from human feedback and the key technical steps for developing ChatGPT.We further analyze the pros and cons of ChatGPT and we rethink the duality of ChatGPT in various fields.Although it has been widely acknowledged that ChatGPT brings plenty of opportunities for various fields,mankind should still treat and use ChatG PT properly to avoid the potential threat,e.g.,academic integrity and safety challenge.Finally,we discuss several open problems as the potential development of ChatGPT.展开更多
Inspired by the design philosophy of information metasurfaces based on the digital coding concept,a planar 4-bit reconfigurable antenna array with low profile of 0.15λ0(whereλ0is the free-space wavelength)is present...Inspired by the design philosophy of information metasurfaces based on the digital coding concept,a planar 4-bit reconfigurable antenna array with low profile of 0.15λ0(whereλ0is the free-space wavelength)is presented.The array is based on a digital coding radiation element consisting of a 1-bit magnetoelectric(ME)dipole and a miniaturized reflection-type phase shifter(RTPS).The proposed 1-bit ME dipole can provide two digital states of"0"and"1"(with 0°and 180°phase responses)over a wide frequency band by individually exciting its two symmetrical feeding ports.The designed RTPS is able to realize a relative phase shift of 173°.By digitally quantizing its phase in the range of 157.5°,additional eight digital states at intervals of 22.5°are obtained.To achieve low sidelobe levels,a 1:16 power divider based on the Taylor line source method is employed to feed the array,A prototype of the proposed 4-bit antenna array has been fabricated and tested,and the experimental results are in good agreement with the simulations.Scanning beams within a±45°range were measured with a maximum realized gain of 13.4 dBi at12 GHz.The sidelobe and cross-polarization levels are below-14.3 and-23.0 dB,respectively.Furthermore,the beam pointing error is within 0.8°,and the 3 dB gain bandwidth of the broadside beam is 25%.Due to its outstanding performance,the array holds potential for significant applications in radar and wireless communication systems.展开更多
High purity SiC crystal was used as a passive monitor to measure neutron irradiation temperature in the 49-2 research reactor.The SiC monitors were irradiated with fast neutrons at elevated temperatures to 3.2×10...High purity SiC crystal was used as a passive monitor to measure neutron irradiation temperature in the 49-2 research reactor.The SiC monitors were irradiated with fast neutrons at elevated temperatures to 3.2×10^(20)n/cm^(2).The isochronal and isothermal annealing behaviors of the irradiated SiC were investigated by x-ray diffraction and four-point probe techniques.Invisible point defects and defect clusters are found to be the dominating defect types in the neutron-irradiated SiC.The amount of defect recovery in SiC reaches a maximum value after isothermal annealing for 30 min.Based on the annealing temperature dependences of both lattice swelling and material resistivity,the irradiation temperature of the SiC monitors is determined to be~410℃,which is much higher than the thermocouple temperature of 275℃ recorded during neutron irradiation.The possible reasons for the difference are carefully discussed.展开更多
Aiming at the problems of low target image resolution,insufficient target feature extraction,low detection accuracy and poor real time in remote engineering vehicle detection,an improved CenterNet target detection mod...Aiming at the problems of low target image resolution,insufficient target feature extraction,low detection accuracy and poor real time in remote engineering vehicle detection,an improved CenterNet target detection model is proposed in this paper.Firstly,EfficientNet-B0 with Efficient Channel Attention(ECA)module is used as the basic network,which increases the quality and speed of feature extraction and reduces the number of model parameters.Then,the proposed Adaptive Fusion Bidirectional Feature Pyramid Network(AF-BiFPN)module is applied to fuse the features of different feature layers.Furthermore,the feature information of engineering vehicle targets is added by making full use of the high-level semantic and low-level fine-grained feature information of the target,which overcomes the problem that the original CenterNet network did not perform well in small target detection and improve the detection accuracy of the network.Finally,the tag coding strategy and bounding box regression method of CenterNet are optimized by introducing positioning quality loss.The accuracy of target prediction is increased by joint prediction of center position and target size.Experimental results show that the mean Average Precision(mAP)of the improved CenterNet model is 94.74%on the engineering vehicle dataset,and the detection rate is 29 FPS.Compared with the original CenterNet model based on ResNet-18,the detection accuracy of this model is improved by 16.29%,the detection speed is increased by 9 FPS,and the memory usage is reduced by 43 MB.Compared with YOLOv3 and YOLOv4,the mAP of this model is improved by 19.9%and 5.61%respectively.The proposed method can detect engineering vehicles more quickly and accurately in far distance.It has obvious advantages in target detection compared with traditional methods.展开更多
WISA 2006 has received 581 submissions and has accepted 65 papers for publication of this issue. These papers are involved in 8 research areas, including Web Information Mining and Retrieval, Semantic Web and Intellig...WISA 2006 has received 581 submissions and has accepted 65 papers for publication of this issue. These papers are involved in 8 research areas, including Web Information Mining and Retrieval, Semantic Web and Intelligent Web, Web Data Management and Information Integration, Web Application Framework and Architecture, Web Information Security, Web Services and Workflow Models, Text Processing and Decision Support, and Grid and Networking Technology. This paper gives an introduction to previous WISA conferences and a survey on the papers to be published in this issue.展开更多
Interactivity is the most significant feature of network data,especially in social networks.Existing network embedding methods have achieved remarkable results in learning network structure and node attributes,but do ...Interactivity is the most significant feature of network data,especially in social networks.Existing network embedding methods have achieved remarkable results in learning network structure and node attributes,but do not pay attention to the multi-interaction between nodes,which limits the extraction and mining of potential deep interactions between nodes.To tackle the problem,we propose a method called Multi-Interaction heterogeneous information Network Embedding(MINE).Firstly,we introduced the multi-interactions heterogeneous information network and extracted complex heterogeneous relation sequences by the multi-interaction extraction algorithm.Secondly,we use a well-designed multi-relationship network fusion model based on the attention mechanism to fuse multiple interactional relationships.Finally,applying a multitasking model makes the learned vector contain richer semantic relationships.A large number of practical experiments prove that our proposed method outperforms existing methods on multiple data sets.展开更多
Vehicular Edge Computing(VEC)is a promising technique to accommodate the computation-intensive and delaysensitive tasks through offloading the tasks to the RoadSide-Unit(RSU)equipped with edge computing servers or nei...Vehicular Edge Computing(VEC)is a promising technique to accommodate the computation-intensive and delaysensitive tasks through offloading the tasks to the RoadSide-Unit(RSU)equipped with edge computing servers or neighboring vehicles.Nevertheless,the limited computation resources of edge computing servers and the mobility of vehicles make the offloading policy design very challenging.In this context,through considering the potential transmission gains brought by the mobility of vehicles,we propose an efficient computation offloading and resource allocation scheme in VEC networks with two kinds of offloading modes,i.e.,Vehicle to Vehicle(V2V)and Vehicle to RSU(V2R).We define a new cost function for vehicular users by incorporating the vehicles’offloading delay,energy consumption,and expenses with a differentiated pricing strategy,as well as the transmission gain.An optimization problem is formulated to minimize the average cost of all the task vehicles under the latency and computation capacity constraints.A distributed iterative algorithm is proposed by decoupling the problem into two subproblems for the offloading mode selection and the resource allocation.Matching theorybased and Lagrangian-based algorithms are proposed to solve the two subproblems,respectively.Simulation results show the proposed algorithm achieves low complexity and significantly improves the system performance compared with three benchmark schemes.展开更多
In order to improve the registration accuracy of brain magnetic resonance images(MRI),some deep learning registration methods use segmentation images for training model.How-ever,the segmentation values are constant fo...In order to improve the registration accuracy of brain magnetic resonance images(MRI),some deep learning registration methods use segmentation images for training model.How-ever,the segmentation values are constant for each label,which leads to the gradient variation con-centrating on the boundary.Thus,the dense deformation field(DDF)is gathered on the boundary and there even appears folding phenomenon.In order to fully leverage the label information,the morphological opening and closing information maps are introduced to enlarge the non-zero gradi-ent regions and improve the accuracy of DDF estimation.The opening information maps supervise the registration model to focus on smaller,narrow brain regions.The closing information maps supervise the registration model to pay more attention to the complex boundary region.Then,opening and closing morphology networks(OC_Net)are designed to automatically generate open-ing and closing information maps to realize the end-to-end training process.Finally,a new registra-tion architecture,VM_(seg+oc),is proposed by combining OC_Net and VoxelMorph.Experimental results show that the registration accuracy of VM_(seg+oc) is significantly improved on LPBA40 and OASIS1 datasets.Especially,VM_(seg+oc) can well improve registration accuracy in smaller brain regions and narrow regions.展开更多
Optoelectronic materials are essential for today's scientific and technological development,and machine learning provides new ideas and tools for their research.In this paper,we first summarize the development his...Optoelectronic materials are essential for today's scientific and technological development,and machine learning provides new ideas and tools for their research.In this paper,we first summarize the development history of optoelectronic materials and how materials informatics drives the innovation and progress of optoelectronic materials and devices.Then,we introduce the development of machine learning and its general process in optoelectronic materials and describe the specific implementation methods.We focus on the cases of machine learning in several application scenarios of optoelectronic materials and devices,including the methods related to crystal structure,properties(defects,electronic structure)research,materials and devices optimization,material characterization,and process optimization.In summarizing the algorithms and feature representations used in different studies,it is noted that prior knowledge can improve optoelectronic materials design,research,and decision-making processes.Finally,the prospect of machine learning applications in optoelectronic materials is discussed,along with current challenges and future directions.This paper comprehensively describes the application value of machine learning in optoelectronic materials research and aims to provide reference and guidance for the continuous development of this field.展开更多
In this paper, we rewrote the equation of algebraic curve segmentswith the geometric informationonboth ends. The optimal or nearly optimal rationalparametric equation is determinedbythe principle that parametricspeeds...In this paper, we rewrote the equation of algebraic curve segmentswith the geometric informationonboth ends. The optimal or nearly optimal rationalparametric equation is determinedbythe principle that parametricspeedsat both endsareequal. Comparing withotherliteratures, the methodofthis paper has advantage in efficiency andiseasy to realize. The equation of optimal rational parameterization can be obtained directly by the information of both ends. Large numbers ofexperimental data show that our method hasbeen given withmore self-adaptability and accuracy than that ofotherliteratures, and if the parametricspeedat any end reaches its maximum or minimum value, the parameterization is optimal; otherwise itis close tooptimal rational parameterization.展开更多
This paper examines the prediction of film ratings.Firstly,in the data feature engineering,feature construction is performed based on the original features of the film dataset.Secondly,the clustering algorithm is util...This paper examines the prediction of film ratings.Firstly,in the data feature engineering,feature construction is performed based on the original features of the film dataset.Secondly,the clustering algorithm is utilized to remove singular film samples,and feature selections are carried out.When solving the problem that film samples of the target domain are unlabelled,it is impossible to train a model and address the inconsistency in the feature dimension for film samples from the source domain.Therefore,the domain adaptive transfer learning model combined with dimensionality reduction algorithms is adopted in this paper.At the same time,in order to reduce the prediction error of models,the stacking ensemble learning model for regression is also used.Finally,through comparative experiments,the effectiveness of the proposed method is verified,which proves to be better predicting film ratings in the target domain.展开更多
基金Supported by the National Natural Science Foundation of China(52074345)CNPC Research and Technology Development Project(2021ZG12).
文摘To study the fluid dynamic response mechanism under the working condition of water injection well borehole,based on the microelement analysis of fluid mechanics and the classical theory of hydrodynamics,a fluid microelement pressure-flow rate relationship model is built to derive and solve the dynamic distribution of fluid pressure and flow rate in the space of well borehole.Combined with the production data of a typical deviated well in China,numerical simulations and analyses are carried out to analyze the dynamic distribution of wellbore pressure at different injection pressures and injection volumes,the delayed and attenuated characteristics of fluid transmission in tube,and the dynamic distribution of wellbore pressure amplitude under the fluctuation of wellhead pressure.The pressure loss along the wellbore has nothing to do with the absolute pressure,and the design of the coding and decoding scheme for wave code communication doesn’t need to consider the absolute pressure during injecting.When the injection pressure is constant,the higher the injection flow rate at the wellhead,the larger the pressure loss along the wellbore.The fluid wave signal delay amplitude mainly depends on the length of the wellbore.The smaller the tubing diameter,the larger the fluid wave signal attenuation amplitude.The higher the target wave code amplitude(differential pressure identification root mean square)generated at the same well depth,the greater the wellhead pressure wave amplitude required to overcome the wellbore pressure loss.
基金the Natural Science Foundation of China under Grant 52077027in part by the Liaoning Province Science and Technology Major Project No.2020JH1/10100020.
文摘In the process of identifying parameters for a permanent magnet synchronous motor,the particle swarm optimization method is prone to being stuck in local optima in the later stages of iteration,resulting in low parameter accuracy.This work proposes a fuzzy particle swarm optimization approach based on the transformation function and the filled function.This approach addresses the topic of particle swarmoptimization in parameter identification from two perspectives.Firstly,the algorithm uses a transformation function to change the form of the fitness function without changing the position of the extreme point of the fitness function,making the extreme point of the fitness function more prominent and improving the algorithm’s search ability while reducing the algorithm’s computational burden.Secondly,on the basis of themulti-loop fuzzy control systembased onmultiplemembership functions,it is merged with the filled function to improve the algorithm’s capacity to skip out of the local optimal solution.This approach can be used to identify the parameters of permanent magnet synchronous motors by sampling only the stator current,voltage,and speed data.The simulation results show that the method can effectively identify the electrical parameters of a permanent magnet synchronous motor,and it has superior global convergence performance and robustness.
基金This work was supported by the National Natural Science Foundation of China(Nos.62034006,92264201,and 91964105)the Natural Science Foundation of Shandong Province(Nos.ZR2020JQ28 and ZR2020KF016)the Program of Qilu Young Scholars of Shandong University.
文摘With the rapid development of machine learning,the demand for high-efficient computing becomes more and more urgent.To break the bottleneck of the traditional Von Neumann architecture,computing-in-memory(CIM)has attracted increasing attention in recent years.In this work,to provide a feasible CIM solution for the large-scale neural networks(NN)requiring continuous weight updating in online training,a flash-based computing-in-memory with high endurance(10^(9) cycles)and ultrafast programming speed is investigated.On the one hand,the proposed programming scheme of channel hot electron injection(CHEI)and hot hole injection(HHI)demonstrate high linearity,symmetric potentiation,and a depression process,which help to improve the training speed and accuracy.On the other hand,the low-damage programming scheme and memory window(MW)optimizations can suppress cell degradation effectively with improved computing accuracy.Even after 109 cycles,the leakage current(I_(off))of cells remains sub-10pA,ensuring the large-scale computing ability of memory.Further characterizations are done on read disturb to demonstrate its robust reliabilities.By processing CIFAR-10 tasks,it is evident that~90%accuracy can be achieved after 109 cycles in both ResNet50 and VGG16 NN.Our results suggest that flash-based CIM has great potential to overcome the limitations of traditional Von Neumann architectures and enable high-performance NN online training,which pave the way for further development of artificial intelligence(AI)accelerators.
基金supported in part by the National Natural Science Foundation of China (Nos.62171029,61931015,U1833203)Natural Science Foundation of Beijing Municipality (No.4172052)supported in part by the Basic Research Program of Jiangsu Province (No.SBK2019042353)。
文摘Detection of maneuvering small targets has always been an important yet challenging task for radar signal processing.One primary reason is that target variable motions within coherent processing interval generate energy migrations across multiple resolution bins,which severely deteriorate the parameter estimation performance.A coarse-to-fine strategy for the detection of maneuvering small targets is proposed.Integration of small points segmented coherently is performed first,and then an optimal inter-segment integration is utilized to derive the coarse estimation of the chirp rate.Sparse fractional Fourier transform(FrFT)is then employed to refine the coarse estimation at a significantly reduced computational complexity.Simulation results verify the proposed scheme that achieves an efficient and reliable maneuvering target detection with-16dB input signal-to-noise ratio(SNR),while requires no exact a priori knowledge on the motion parameters.
基金Project(20040533035) supported by the National Research Foundation for the Doctoral Program of Higher Education of ChinaProject (50275150) supported by the National Natural Science Foundation of China
文摘In converged heterogeneous wireless networks, vertical handoff is an important issue in radio resource management and occurs when an end user switches from one network to another (e.g., from wireless local area network to wideband code division multiple access). Efficient vertical handoff should allocate network resources efficiently and maintain good quality of service (QoS) for the end users. The objective of this work is to determine conditions under which vertical handoff can be performed. The channel usage situation of each access network is formulated as a birth-death process with the objective of predicting the available bandwidth and the blocking probability. A reward function is used to capture the network bandwidth and the blocking probability is expressed as a cost function. An end user will access the certain network which maximizes the total function defined as the combination of the reward function and the cost function. Simulation results show that the proposed algorithm can significantly improve the network performance, including higher bandwidth for end users and lower new call blocking and handoff call blocking probability for networks.
基金supported by the National Natural Science Foundation of China (Nos.10872068,10672057)
文摘This paper proposes a model of neural networks consisting of populations of perceptive neurons, inter-neurons, and motor neurons according to the theory of stochastic phase resetting dynamics. According to this model, the dynamical characteristics of neural networks are studied in three coupling cases, namely, series and parallel coupling, series coupling, and unilateral coupling. The results show that the indentified structure of neural networks enables the basic characteristics of neural information processing to be described in terms of the actions of both the optional motor and the reflected motor. The excitation of local neural networks is caused by the action of the optional motor. In particular, the excitation of the neural population caused by the action of the optional motor in the motor cortex is larger than that caused by the action of the reflected motor. This phenomenon indicates that there are more neurons participating in the neural information processing and the excited synchronization motion under the action of the optional motor.
基金Sponsored by the National Natural Science Foundation of China(Grant No.61173052)the China Postdoctoral Scinece Foundation(Grant No.2014M561363)
文摘For the issue of evaluation of capability of enterprise agent coalition,an evaluation model based on information fusion and entropy weighting method is presented. The attribute reduction method is utilized to reduce indicators of the capability according to the theory of rough set. The new indicator system can be determined. Attribute reduction can also reduce the workload and remove the redundant information,when there are too many indicators or the indicators have strong correlation. The research complexity can be reduced and the efficiency can be improved. Entropy weighting method is used to determine the weights of the remaining indicators,and the importance of indicators is analyzed. The information fusion model based on nearest neighbor method is developed and utilized to evaluate the capability of multiple agent coalitions,compared to cloud evaluation model and D-S evidence method. Simulation results are reasonable and with obvious distinction. Thus they verify the effectiveness and feasibility of the model. The information fusion model can provide more scientific,rational decision support for choosing the best agent coalition,and provide innovative steps for the evaluation process of capability of agent coalitions.
基金Projects (60234030 60404021) supported by the National Natural Science Foundation of China
文摘The general frame for the system of intelligence science was proposed, the common features of the researching objects of the intelligence science were summarized. The intelligence science consists of three portions: scientific foundation, technical methodology and application fields. The common features of intelligence science include complexity, intersection, nonlinearity, anthropomorphic property, uncertainty, incompleteness and distribution etc. The new proposed scientific branch would reflect the new height, new thought and new way for developing the control science and intelligent systems from one angle, and present a strong wish for establishing a new branch of intelligence science.
文摘In this research, a near infrared multi-wavelength noninvasive blood glucose monitoring system with distributed laser multi-sensors is applied to monitor human blood glucose concentration. In order to improve the monitoring accuracy, a multi-sensors information fusion model based on Back Propagation Artificial Neural Network is proposed. The Root- Mean-Square Error of Prediction for noninvasive blood glucose measurement is 0.088mmol/L, and the correlation coefficient is 0.94. The noninvasive blood glucose monitoring system based on distributed multi-sensors information fusion of multi-wavelength NIR is proved to be of great efficient. And the new proposed idea of measurement based on distri- buted multi-sensors, shows better prediction accuracy.
基金supported by National Key Research and Development Program of China (2021YFB1714300)National Natural Science Foundation of China (62293502, 61831022, 61976211)Youth Innovation Promotion Association CAS。
文摘ChatG PT,an artificial intelligence generated content (AIGC) model developed by OpenAI,has attracted worldwide attention for its capability of dealing with challenging language understanding and generation tasks in the form of conversations.This paper briefly provides an overview on the history,status quo and potential future development of ChatGPT,helping to provide an entry point to think about ChatGPT.Specifically,from the limited open-accessed resources,we conclude the core techniques of ChatGPT,mainly including large-scale language models,in-context learning,reinforcement learning from human feedback and the key technical steps for developing ChatGPT.We further analyze the pros and cons of ChatGPT and we rethink the duality of ChatGPT in various fields.Although it has been widely acknowledged that ChatGPT brings plenty of opportunities for various fields,mankind should still treat and use ChatG PT properly to avoid the potential threat,e.g.,academic integrity and safety challenge.Finally,we discuss several open problems as the potential development of ChatGPT.
基金supported in part by the National Key Research and Development Program of China(2017YFA0700201,2017YFA0700202,and 2017YFA0700203)the National Natural Science Foundation of China(61631007,61571117,61138001,61371035,61722106,61731010,11227904,and 62171124)+1 种基金the 111 Project(111-2-05)the Scientific Research Foundation of Graduate School of Southeast University(YBYP2119)。
文摘Inspired by the design philosophy of information metasurfaces based on the digital coding concept,a planar 4-bit reconfigurable antenna array with low profile of 0.15λ0(whereλ0is the free-space wavelength)is presented.The array is based on a digital coding radiation element consisting of a 1-bit magnetoelectric(ME)dipole and a miniaturized reflection-type phase shifter(RTPS).The proposed 1-bit ME dipole can provide two digital states of"0"and"1"(with 0°and 180°phase responses)over a wide frequency band by individually exciting its two symmetrical feeding ports.The designed RTPS is able to realize a relative phase shift of 173°.By digitally quantizing its phase in the range of 157.5°,additional eight digital states at intervals of 22.5°are obtained.To achieve low sidelobe levels,a 1:16 power divider based on the Taylor line source method is employed to feed the array,A prototype of the proposed 4-bit antenna array has been fabricated and tested,and the experimental results are in good agreement with the simulations.Scanning beams within a±45°range were measured with a maximum realized gain of 13.4 dBi at12 GHz.The sidelobe and cross-polarization levels are below-14.3 and-23.0 dB,respectively.Furthermore,the beam pointing error is within 0.8°,and the 3 dB gain bandwidth of the broadside beam is 25%.Due to its outstanding performance,the array holds potential for significant applications in radar and wireless communication systems.
文摘High purity SiC crystal was used as a passive monitor to measure neutron irradiation temperature in the 49-2 research reactor.The SiC monitors were irradiated with fast neutrons at elevated temperatures to 3.2×10^(20)n/cm^(2).The isochronal and isothermal annealing behaviors of the irradiated SiC were investigated by x-ray diffraction and four-point probe techniques.Invisible point defects and defect clusters are found to be the dominating defect types in the neutron-irradiated SiC.The amount of defect recovery in SiC reaches a maximum value after isothermal annealing for 30 min.Based on the annealing temperature dependences of both lattice swelling and material resistivity,the irradiation temperature of the SiC monitors is determined to be~410℃,which is much higher than the thermocouple temperature of 275℃ recorded during neutron irradiation.The possible reasons for the difference are carefully discussed.
基金This research was funded by College Student Innovation and Entrepreneurship Training Program,Grant Number 2021055Z and S202110082031the Special Project for Cultivating Scientific and Technological Innovation Ability of College and Middle School Students in Hebei Province,Grant Number 2021H011404.
文摘Aiming at the problems of low target image resolution,insufficient target feature extraction,low detection accuracy and poor real time in remote engineering vehicle detection,an improved CenterNet target detection model is proposed in this paper.Firstly,EfficientNet-B0 with Efficient Channel Attention(ECA)module is used as the basic network,which increases the quality and speed of feature extraction and reduces the number of model parameters.Then,the proposed Adaptive Fusion Bidirectional Feature Pyramid Network(AF-BiFPN)module is applied to fuse the features of different feature layers.Furthermore,the feature information of engineering vehicle targets is added by making full use of the high-level semantic and low-level fine-grained feature information of the target,which overcomes the problem that the original CenterNet network did not perform well in small target detection and improve the detection accuracy of the network.Finally,the tag coding strategy and bounding box regression method of CenterNet are optimized by introducing positioning quality loss.The accuracy of target prediction is increased by joint prediction of center position and target size.Experimental results show that the mean Average Precision(mAP)of the improved CenterNet model is 94.74%on the engineering vehicle dataset,and the detection rate is 29 FPS.Compared with the original CenterNet model based on ResNet-18,the detection accuracy of this model is improved by 16.29%,the detection speed is increased by 9 FPS,and the memory usage is reduced by 43 MB.Compared with YOLOv3 and YOLOv4,the mAP of this model is improved by 19.9%and 5.61%respectively.The proposed method can detect engineering vehicles more quickly and accurately in far distance.It has obvious advantages in target detection compared with traditional methods.
文摘WISA 2006 has received 581 submissions and has accepted 65 papers for publication of this issue. These papers are involved in 8 research areas, including Web Information Mining and Retrieval, Semantic Web and Intelligent Web, Web Data Management and Information Integration, Web Application Framework and Architecture, Web Information Security, Web Services and Workflow Models, Text Processing and Decision Support, and Grid and Networking Technology. This paper gives an introduction to previous WISA conferences and a survey on the papers to be published in this issue.
基金This work is supported by the Fundamental Research Funds for the Central Universities(Grant No.HIT.NSRIF.201714)Weihai Science and Technology Development Program(2016DXGJMS15)Key Research and Development Program in Shandong Provincial(2017GGX90103).
文摘Interactivity is the most significant feature of network data,especially in social networks.Existing network embedding methods have achieved remarkable results in learning network structure and node attributes,but do not pay attention to the multi-interaction between nodes,which limits the extraction and mining of potential deep interactions between nodes.To tackle the problem,we propose a method called Multi-Interaction heterogeneous information Network Embedding(MINE).Firstly,we introduced the multi-interactions heterogeneous information network and extracted complex heterogeneous relation sequences by the multi-interaction extraction algorithm.Secondly,we use a well-designed multi-relationship network fusion model based on the attention mechanism to fuse multiple interactional relationships.Finally,applying a multitasking model makes the learned vector contain richer semantic relationships.A large number of practical experiments prove that our proposed method outperforms existing methods on multiple data sets.
基金The work was supported in part by the National Natural Science Foundation of China(No.62271295,U22A2003,62201329)Shandong Provincial Natural Science Foundation(ZR2020QF002,ZR2022QF002).
文摘Vehicular Edge Computing(VEC)is a promising technique to accommodate the computation-intensive and delaysensitive tasks through offloading the tasks to the RoadSide-Unit(RSU)equipped with edge computing servers or neighboring vehicles.Nevertheless,the limited computation resources of edge computing servers and the mobility of vehicles make the offloading policy design very challenging.In this context,through considering the potential transmission gains brought by the mobility of vehicles,we propose an efficient computation offloading and resource allocation scheme in VEC networks with two kinds of offloading modes,i.e.,Vehicle to Vehicle(V2V)and Vehicle to RSU(V2R).We define a new cost function for vehicular users by incorporating the vehicles’offloading delay,energy consumption,and expenses with a differentiated pricing strategy,as well as the transmission gain.An optimization problem is formulated to minimize the average cost of all the task vehicles under the latency and computation capacity constraints.A distributed iterative algorithm is proposed by decoupling the problem into two subproblems for the offloading mode selection and the resource allocation.Matching theorybased and Lagrangian-based algorithms are proposed to solve the two subproblems,respectively.Simulation results show the proposed algorithm achieves low complexity and significantly improves the system performance compared with three benchmark schemes.
基金supported by Shandong Provincial Natural Science Foundation(No.ZR2023MF062)the National Natural Science Foundation of China(No.61771230).
文摘In order to improve the registration accuracy of brain magnetic resonance images(MRI),some deep learning registration methods use segmentation images for training model.How-ever,the segmentation values are constant for each label,which leads to the gradient variation con-centrating on the boundary.Thus,the dense deformation field(DDF)is gathered on the boundary and there even appears folding phenomenon.In order to fully leverage the label information,the morphological opening and closing information maps are introduced to enlarge the non-zero gradi-ent regions and improve the accuracy of DDF estimation.The opening information maps supervise the registration model to focus on smaller,narrow brain regions.The closing information maps supervise the registration model to pay more attention to the complex boundary region.Then,opening and closing morphology networks(OC_Net)are designed to automatically generate open-ing and closing information maps to realize the end-to-end training process.Finally,a new registra-tion architecture,VM_(seg+oc),is proposed by combining OC_Net and VoxelMorph.Experimental results show that the registration accuracy of VM_(seg+oc) is significantly improved on LPBA40 and OASIS1 datasets.Especially,VM_(seg+oc) can well improve registration accuracy in smaller brain regions and narrow regions.
基金Project supported by the National Natural Science Foundation of China (Grant No.61601198)the University of Jinan PhD Foundation (Grant No.XBS1714)。
文摘Optoelectronic materials are essential for today's scientific and technological development,and machine learning provides new ideas and tools for their research.In this paper,we first summarize the development history of optoelectronic materials and how materials informatics drives the innovation and progress of optoelectronic materials and devices.Then,we introduce the development of machine learning and its general process in optoelectronic materials and describe the specific implementation methods.We focus on the cases of machine learning in several application scenarios of optoelectronic materials and devices,including the methods related to crystal structure,properties(defects,electronic structure)research,materials and devices optimization,material characterization,and process optimization.In summarizing the algorithms and feature representations used in different studies,it is noted that prior knowledge can improve optoelectronic materials design,research,and decision-making processes.Finally,the prospect of machine learning applications in optoelectronic materials is discussed,along with current challenges and future directions.This paper comprehensively describes the application value of machine learning in optoelectronic materials research and aims to provide reference and guidance for the continuous development of this field.
文摘In this paper, we rewrote the equation of algebraic curve segmentswith the geometric informationonboth ends. The optimal or nearly optimal rationalparametric equation is determinedbythe principle that parametricspeedsat both endsareequal. Comparing withotherliteratures, the methodofthis paper has advantage in efficiency andiseasy to realize. The equation of optimal rational parameterization can be obtained directly by the information of both ends. Large numbers ofexperimental data show that our method hasbeen given withmore self-adaptability and accuracy than that ofotherliteratures, and if the parametricspeedat any end reaches its maximum or minimum value, the parameterization is optimal; otherwise itis close tooptimal rational parameterization.
基金Supported by the Scientific Research Foundation of Liaoning Provincial Department of Education(No.LJKZ0139).
文摘This paper examines the prediction of film ratings.Firstly,in the data feature engineering,feature construction is performed based on the original features of the film dataset.Secondly,the clustering algorithm is utilized to remove singular film samples,and feature selections are carried out.When solving the problem that film samples of the target domain are unlabelled,it is impossible to train a model and address the inconsistency in the feature dimension for film samples from the source domain.Therefore,the domain adaptive transfer learning model combined with dimensionality reduction algorithms is adopted in this paper.At the same time,in order to reduce the prediction error of models,the stacking ensemble learning model for regression is also used.Finally,through comparative experiments,the effectiveness of the proposed method is verified,which proves to be better predicting film ratings in the target domain.