Non-uniform linear array(NULA)configurations are well renowned due to their structural ability for providing increased degrees of freedom(DOF)and wider array aperture than uniform linear arrays(ULAs).These characteris...Non-uniform linear array(NULA)configurations are well renowned due to their structural ability for providing increased degrees of freedom(DOF)and wider array aperture than uniform linear arrays(ULAs).These characteristics play a significant role in improving the direction-of-arrival(DOA)estimation accuracy.However,most of the existing NULA geometries are primarily applicable to circular sources(CSs),while they limitedly improve the DOF and continuous virtual aperture for noncircular sources(NCSs).Toward this purpose,we present a triaddisplaced ULAs(Tdis-ULAs)configuration for NCS.The TdisULAs structure generally consists of three ULAs,which are appropriately placed.The proposed antenna array approach fully exploits the non-circular characteristics of the sources.Given the same number of elements,the Tdis-ULAs design achieves more DOF and larger hole-free co-array aperture than its sparse array competitors.Advantageously,the number of uniform DOF,optimal distribution of elements among the ULAs,and precise element positions are uniquely determined by the closed-form expressions.Moreover,the proposed array also produces a filled resulting co-array.Numerical simulations are conducted to show the performance advantages of the proposed Tdis-ULAs configuration over its counterpart designs.展开更多
With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors we...With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors were widely applied due to their low cost. This paper explored the implementation of a human hand posture recognition system using ToF sensors and residual neural networks. Firstly, this paper reviewed the typical applications of human hand recognition. Secondly, this paper designed a hand gesture recognition system using a ToF sensor VL53L5. Subsequently, data preprocessing was conducted, followed by training the constructed residual neural network. Then, the recognition results were analyzed, indicating that gesture recognition based on the residual neural network achieved an accuracy of 98.5% in a 5-class classification scenario. Finally, the paper discussed existing issues and future research directions.展开更多
The direct target of engineering education is to cultivate engineers and industry leaders with the qualifi ed skills and strong competencies. To cultivate skilled technological talents in a local engineering college, ...The direct target of engineering education is to cultivate engineers and industry leaders with the qualifi ed skills and strong competencies. To cultivate skilled technological talents in a local engineering college, the fi rst subject is to establish the discipline according to the major strategy of the local economic development requirements, so that the professional regional service function is relatively strong. Secondly, the individual work orientation and development direction are considered as well in the cultivated process. These aspects are required such as renewing education idea, strengthening the education of humanity quality, perfecting the system of continuing education, establishing the mechanism of lifelong learning.展开更多
Model predictive control is widely used in the design of autonomous driving algorithms.However,its parameters are sensitive to dynamically varying driving conditions,making it difficult to be implemented into practice...Model predictive control is widely used in the design of autonomous driving algorithms.However,its parameters are sensitive to dynamically varying driving conditions,making it difficult to be implemented into practice.As a result,this study presents a self-learning algorithm based on reinforcement learning to tune a model predictive controller.Specifically,the proposed algorithm is used to extract features of dynamic traffic scenes and adjust the weight coefficients of the model predictive controller.In this method,a risk threshold model is proposed to classify the risk level of the scenes based on the scene features,and aid in the design of the reinforcement learning reward function and ultimately improve the adaptability of the model predictive controller to real-world scenarios.The proposed algorithm is compared to a pure model predictive controller in car-following case.According to the results,the proposed method enables autonomous vehicles to adjust the priority of performance indices reasonably in different scenarios according to risk variations,showing a good scenario adaptability with safety guaranteed.展开更多
With the development of urbanization,the problem of neurological diseases brought about by population aging has gradually become a social problem of worldwide concern.Aging leads to gradual degeneration of the central...With the development of urbanization,the problem of neurological diseases brought about by population aging has gradually become a social problem of worldwide concern.Aging leads to gradual degeneration of the central nervous system,shrinkage of brain tissue,and decline in physical function in many elderlies,making them susceptible to neurological diseases such as Alzheimer’s disease(AD),stroke,Parkinson’s and major depressive disorder(MDD).Due to the influence of these neurological diseases,the elderly have troubles such as memory loss,inability to move,falling,and getting lost,which seriously affect their quality of life.Tracking and positioning of elderly with neurological diseases and keeping track of their location in real-time are necessary and crucial in order to detect and treat dangerous and unexpected situations in time.Considering that the elderly with neurological diseases forget to wear a positioning device or have mobility problems due to carrying a positioning device,device-free positioning as a passive positioning technology that detects device-free individuals is more suitable than traditional active positioning for the home-based care of the elderly with neurological diseases.This paper provides an extensive and in-depth survey of device-free indoor positioning technology for home-based care and an in-depth analysis of the main features of current positioning systems,as well as the techniques,technologies andmethods they employ,fromthe perspective of the needs of the elderly with neurological conditions.Moreover,evaluation criteria and possible solutions of positioning techniques for the home-based care of the elderly with neurological conditions are proposed.Finally,the opportunities and challenges for the development of indoor positioning technology in 6G mobile networks for home-based care of the elderly with neurological diseases are discussed.This review has provided comprehensive and effective tracking and positioning techniques,technologies and methods for the elderly,by which we can obtain the location information of the elderly in real-time and make home-based care more comfortable and safer for the elderly with neurological diseases.展开更多
Decision-making and motion planning are extremely important in autonomous driving to ensure safe driving in a real-world environment.This study proposes an online evolutionary decision-making and motion planning frame...Decision-making and motion planning are extremely important in autonomous driving to ensure safe driving in a real-world environment.This study proposes an online evolutionary decision-making and motion planning framework for autonomous driving based on a hybrid data-and model-driven method.First,a data-driven decision-making module based on deep reinforcement learning(DRL)is developed to pursue a rational driving performance as much as possible.Then,model predictive control(MPC)is employed to execute both longitudinal and lateral motion planning tasks.Multiple constraints are defined according to the vehicle’s physical limit to meet the driving task requirements.Finally,two principles of safety and rationality for the self-evolution of autonomous driving are proposed.A motion envelope is established and embedded into a rational exploration and exploitation scheme,which filters out unreasonable experiences by masking unsafe actions so as to collect high-quality training data for the DRL agent.Experiments with a high-fidelity vehicle model and MATLAB/Simulink co-simulation environment are conducted,and the results show that the proposed online-evolution framework is able to generate safer,more rational,and more efficient driving action in a real-world environment.展开更多
Colored dissolved organic matter(CDOM)is an important optically active substance in marine environment.Its biochemical conservatism makes it an important indicator to offshore pollution process.Monitoring the content,...Colored dissolved organic matter(CDOM)is an important optically active substance in marine environment.Its biochemical conservatism makes it an important indicator to offshore pollution process.Monitoring the content,composition,and diffusion process of CDOM is a good approach to analyze the terrestrial input,optical properties,and ecological environment of offshore areas.The spatiotemporal characteristics of CDOM around the Leizhou Peninsula were analyzed based on field observation data collected in the autumn 2020 and spring 2021,and an empirical inversion model of the ag(355)(absorption coefficient at 355 nm)and spectral slope(Sg)(g stands for gelbstoff/gilvin,which is called CDOM)based on Sentinel-3A ocean and land color instrument(OLCI)images was constructed.The results show(1)the order of average ag(355)value around the Leizhou Peninsula was east coast(0.503/m)>Qiongzhou Strait(0.502/m)>west coast(0.365/m);(2)the best band combinations of CDOM inversion in spring and autumn were(B4+B11)/B3 and(B7-B1)/B6(B stands for the band of spectral images),and the final inversion results are close to the measured results,indicating that the model has good accuracy;(3)the S_(g)value of the CDOM absorption spectrum was fitted to the hyperbolicexponential model.The fitting accuracy of the model was higher than those of the exponential model and the hyperbolic model,and the best S_(g)inversion model was constructed by selecting Sg(275-295)and Sg(250-295)in spring and autumn;(4)the spatial distributions of ag(355)and S_(g)were inverted,and CDOM in the waters around the Leizhou Peninsula originated from terrestrial organic matter carried by coastal aquaculture zones and runoff in the northeast Zhanjiang and the northern Beibu Gulf.展开更多
By pushing computation,cache,and network control to the edge,mobile edge computing(MEC)is expected to play a leading role in fifth generation(5G)and future sixth generation(6G).Nevertheless,facing ubiquitous fast-grow...By pushing computation,cache,and network control to the edge,mobile edge computing(MEC)is expected to play a leading role in fifth generation(5G)and future sixth generation(6G).Nevertheless,facing ubiquitous fast-growing computational demands,it is impossible for a single MEC paradigm to effectively support high-quality intelligent services at end user equipments(UEs).To address this issue,we propose an air-ground collaborative MEC(AGCMEC)architecture in this article.The proposed AGCMEC integrates all potentially available MEC servers within air and ground in the envisioned 6G,by a variety of collaborative ways to provide computation services at their best for UEs.Firstly,we introduce the AGC-MEC architecture and elaborate three typical use cases.Then,we discuss four main challenges in the AGC-MEC as well as their potential solutions.Next,we conduct a case study of collaborative service placement for AGC-MEC to validate the effectiveness of the proposed collaborative service placement strategy.Finally,we highlight several potential research directions of the AGC-MEC.展开更多
In this study,we present characteristics of post-sunset GHz scintillation occurrence and their correlations with ionospheric parameters derived from ionosonde observations in high solar activity years(2012−2013)of sol...In this study,we present characteristics of post-sunset GHz scintillation occurrence and their correlations with ionospheric parameters derived from ionosonde observations in high solar activity years(2012−2013)of solar cycle 24 at Sanya(18.3°N,109.6°E;dip lat.:12.8°N),China.The analyzed data include the F_(2)-layer’s critical frequency(foF_(2)),peak height(h_(m)F_(2)),and minimum virtual height(h’F),as well as the scale height around the F_(2)-layer peak(H_(m)),and virtual height(h’F_(5))and true height(hF_(5))measured at 5 MHz.We have investigated relationships between the equinoctial asymmetry of these scintillations and these ionospheric parameters.In addition,we calculate the growth rates of Rayleigh−Taylor instability on the basis of the ionosonde measurements and theoretical models,respectively.We find that the equinoctial asymmetry of scintillation onset time is associated with the scale length of the vertical electron density gradient(L),which has been shown to affect the growth of Rayleigh−Taylor instability at the bottom of the F-layer.The seasonal variations of foF_(2),H_(m)and scale length of vertical electron density gradient appear to cause the seasonal variations of scintillation occurrence;the equinoctial asymmetry of scintillation occurrence rate over low latitudes appears to be related to background electron density and vertical drifts in the F-layer around time of sunset.Further study is required to explain the observed correlational weakness in low latitudes between scintillation strength,represented by the daily maximum S4,and daily maximum values of foF_(2),h_(m)F_(2),h’F,H_(m),and also the drifts.展开更多
In this paper,we formulate the precoding problem of integrated sensing and communication(ISAC)waveform as a non-convex quadratically constrained quadratic programming(QCQP),in which the weighted sum of communication m...In this paper,we formulate the precoding problem of integrated sensing and communication(ISAC)waveform as a non-convex quadratically constrained quadratic programming(QCQP),in which the weighted sum of communication multi-user interference(MUI)and the gap between dual-use waveform and ideal radar waveform is minimized with peak-toaverage power ratio(PAPR)constraints.We propose an efficient algorithm based on alternating direction method of multipliers(ADMM),which is able to decouple multiple variables and provide a closed-form solution for each subproblem.In addition,to improve the sensing performance in both spatial and temporal domains,we propose a new criteria to design the ideal radar waveform,in which the beam pattern is made similar to the ideal one and the integrated sidelobe level of the ambiguity function in each target direction is minimized in the region of interest.The limited memory Broyden-Fletcher-Goldfarb-Shanno(LBFGS)algorithm is applied to the design of the ideal radar waveform which works as a reference in the design of the dual-function waveform.Numerical results indicate that the designed dual-function waveform is capable of offering good communication quality of service(QoS)and sensing performance.展开更多
The Nesterov accelerated dynamical approach serves as an essential tool for addressing convex optimization problems with accelerated convergence rates.Most previous studies in this field have primarily concentrated on...The Nesterov accelerated dynamical approach serves as an essential tool for addressing convex optimization problems with accelerated convergence rates.Most previous studies in this field have primarily concentrated on unconstrained smooth con-vex optimization problems.In this paper,on the basis of primal-dual dynamical approach,Nesterov accelerated dynamical approach,projection operator and directional gradient,we present two accelerated primal-dual projection neurodynamic approaches with time scaling to address convex optimization problems with smooth and nonsmooth objective functions subject to linear and set constraints,which consist of a second-order ODE(ordinary differential equation)or differential conclusion system for the primal variables and a first-order ODE for the dual vari-ables.By satisfying specific conditions for time scaling,we demonstrate that the proposed approaches have a faster conver-gence rate.This only requires assuming convexity of the objective function.We validate the effectiveness of our proposed two accel-erated primal-dual projection neurodynamic approaches through numerical experiments.展开更多
Collaborative edge computing is a promising direction to handle the computation intensive tasks in B5G wireless networks.However,edge computing servers(ECSs)from different operators may not trust each other,and thus t...Collaborative edge computing is a promising direction to handle the computation intensive tasks in B5G wireless networks.However,edge computing servers(ECSs)from different operators may not trust each other,and thus the incentives for collaboration cannot be guaranteed.In this paper,we propose a consortium blockchain enabled collaborative edge computing framework,where users can offload computing tasks to ECSs from different operators.To minimize the total delay of users,we formulate a joint task offloading and resource optimization problem,under the constraint of the computing capability of each ECS.We apply the Tammer decomposition method and heuristic optimization algorithms to obtain the optimal solution.Finally,we propose a reputation based node selection approach to facilitate the consensus process,and also consider a completion time based primary node selection to avoid monopolization of certain edge node and enhance the security of the blockchain.Simulation results validate the effectiveness of the proposed algorithm,and the total delay can be reduced by up to 40%compared with the non-cooperative case.展开更多
Age-related macular degeneration(AMD)ranks third among the most common causes of blindness.As the most conventional and direct method for identifying AMD,color fundus photography has become prominent owing to its cons...Age-related macular degeneration(AMD)ranks third among the most common causes of blindness.As the most conventional and direct method for identifying AMD,color fundus photography has become prominent owing to its consistency,ease of use,and good quality in extensive clinical practice.In this study,a convolutional neural network(CSPDarknet53)was combined with a transformer to construct a new hybrid model,HCSP-Net.This hybrid model was employed to tri-classify color fundus photography into the normal macula(NM),dry macular degeneration(DMD),and wet macular degeneration(WMD)based on clinical classification manifestations,thus identifying and resolving AMD as early as possible with color fundus photography.To further enhance the performance of this model,grouped convolution was introduced in this study without significantly increasing the number of parameters.HCSP-Net was validated using an independent test set.The average precision of HCSPNet in the diagnosis of AMD was 99.2%,the recall rate was 98.2%,the F1-Score was 98.7%,the PPV(positive predictive value)was 99.2%,and the NPV(negative predictive value)was 99.6%.Moreover,a knowledge distillation approach was also adopted to develop a lightweight student network(SCSP-Net).The experimental results revealed a noteworthy enhancement in the accuracy of SCSP-Net,rising from 94%to 97%,while remarkably reducing the parameter count to a quarter of HCSP-Net.This attribute positions SCSP-Net as a highly suitable candidate for the deployment of resource-constrained devices,which may provide ophthalmologists with an efficient tool for diagnosing AMD.展开更多
In recent times,various power control and clustering approaches have been proposed to enhance overall performance for cell-free massive multipleinput multiple-output(CF-mMIMO)networks.With the emergence of deep reinfo...In recent times,various power control and clustering approaches have been proposed to enhance overall performance for cell-free massive multipleinput multiple-output(CF-mMIMO)networks.With the emergence of deep reinforcement learning(DRL),significant progress has been made in the field of network optimization as DRL holds great promise for improving network performance and efficiency.In this work,our focus delves into the intricate challenge of joint cooperation clustering and downlink power control within CF-mMIMO networks.Leveraging the potent deep deterministic policy gradient(DDPG)algorithm,our objective is to maximize the proportional fairness(PF)for user rates,thereby aiming to achieve optimal network performance and resource utilization.Moreover,we harness the concept of“divide and conquer”strategy,introducing two innovative methods termed alternating DDPG(A-DDPG)and hierarchical DDPG(H-DDPG).These approaches aim to decompose the intricate joint optimization problem into more manageable sub-problems,thereby facilitating a more efficient resolution process.Our findings unequivo-cally showcase the superior efficacy of our proposed DDPG approach over the baseline schemes in both clustering and downlink power control.Furthermore,the A-DDPG and H-DDPG obtain higher performance gain than DDPG with lower computational complexity.展开更多
Point-of-interest(POI)recommendations in location-based social networks(LBSNs)have developed rapidly by incorporating feature information and deep learning methods.However,most studies have failed to accurately reflec...Point-of-interest(POI)recommendations in location-based social networks(LBSNs)have developed rapidly by incorporating feature information and deep learning methods.However,most studies have failed to accurately reflect different users’preferences,in particular,the short-term preferences of inactive users.To better learn user preferences,in this study,we propose a long-short-term-preference-based adaptive successive POI recommendation(LSTP-ASR)method by combining trajectory sequence processing,long short-term preference learning,and spatiotemporal context.First,the check-in trajectory sequences are adaptively divided into recent and historical sequences according to a dynamic time window.Subsequently,an adaptive filling strategy is used to expand the recent check-in sequences of users with inactive check-in behavior using those of similar active users.We further propose an adaptive learning model to accurately extract long short-term preferences of users to establish an efficient successive POI recommendation system.A spatiotemporal-context-based recurrent neural network and temporal-context-based long short-term memory network are used to model the users’recent and historical checkin trajectory sequences,respectively.Extensive experiments on the Foursquare and Gowalla datasets reveal that the proposed method outperforms several other baseline methods in terms of three evaluation metrics.More specifically,LSTP-ASR outperforms the previously best baseline method(RTPM)with a 17.15%and 20.62%average improvement on the Foursquare and Gowalla datasets in terms of the Fβmetric,respectively.展开更多
Extending the ionic conductivity is the pre-requisite of electrolytes in fuel cell technology for high-electrochemical performance.In this regard,the introduction of semiconductor-oxide materials and the approach of h...Extending the ionic conductivity is the pre-requisite of electrolytes in fuel cell technology for high-electrochemical performance.In this regard,the introduction of semiconductor-oxide materials and the approach of heterostructure formation by modulating energy bands to enhance ionic conduction acting as an electrolyte in fuel cell-device.Semiconductor(n-type;SnO_(2))plays a key role by introducing into p-type SrFe_(0.2)Ti_(0.8)O_(3-δ)(SFT)semiconductor perovskite materials to construct p-n heterojunction for high ionic conductivity.Therefore,two different composites of SFT and SnO_(2)are constructed by gluing p-and n-type SFT-SnO_(2),where the optimal composition of SFT-SnO_(2)(6∶4)heterostructure electrolyte-based fuel cell achieved excellent ionic conductivity 0.24 S cm^(-1)with power-output of 1004 mW cm^(-2)and high OCV 1.12 V at a low operational temperature of 500℃.The high power-output and significant ionic conductivity with durable operation of 54 h are accredited to SFT-SnO_(2)heterojunction formation including interfacial conduction assisted by a built-in electric field in fuel cell device.Moreover,the fuel conversion efficiency and considerable Faradaic efficiency reveal the compatibility of SFT-SnO_(2)heterostructure electrolyte and ruled-out short-circuiting issue.Further,the first principle calculation provides sufficient information on structure optimization and energy-band structure modulation of SFT-SnO_(2).This strategy will provide new insight into semiconductor-based fuel cell technology to design novel electrolytes.展开更多
Diagnosing individuals with autism spectrum disorder(ASD)accurately faces great chal-lenges in clinical practice,primarily due to the data's high heterogeneity and limited sample size.To tackle this issue,the auth...Diagnosing individuals with autism spectrum disorder(ASD)accurately faces great chal-lenges in clinical practice,primarily due to the data's high heterogeneity and limited sample size.To tackle this issue,the authors constructed a deep graph convolutional network(GCN)based on variable multi‐graph and multimodal data(VMM‐DGCN)for ASD diagnosis.Firstly,the functional connectivity matrix was constructed to extract primary features.Then,the authors constructed a variable multi‐graph construction strategy to capture the multi‐scale feature representations of each subject by utilising convolutional filters with varying kernel sizes.Furthermore,the authors brought the non‐imaging in-formation into the feature representation at each scale and constructed multiple population graphs based on multimodal data by fully considering the correlation between subjects.After extracting the deeper features of population graphs using the deep GCN(DeepGCN),the authors fused the node features of multiple subgraphs to perform node classification tasks for typical control and ASD patients.The proposed algorithm was evaluated on the Autism Brain Imaging Data Exchange I(ABIDE I)dataset,achieving an accuracy of 91.62%and an area under the curve value of 95.74%.These results demon-strated its outstanding performance compared to other ASD diagnostic algorithms.展开更多
Memristors as non-volatile memory devices have gained numerous attentions owing to their advantages in storage,in-memory computing, synaptic applications, etc. In recent years, two-dimensional(2D) materials with moder...Memristors as non-volatile memory devices have gained numerous attentions owing to their advantages in storage,in-memory computing, synaptic applications, etc. In recent years, two-dimensional(2D) materials with moderate defects have been discovered to exist memristive feature. However, it is very difficult to obtain moderate defect degree in 2D materials, and studied on modulation means and mechanism becomes urgent and essential. In this work, we realized memristive feature with a bipolar switching and a configurable on/off ratio in a two-terminal MoS_(2) device(on/off ratio ~100), for the first time, from absent to present using laser-modulation to few-layer defect-free MoS_(2)(about 10 layers), and its retention time in both high resistance state and low resistance state can reach 2×10^(4) s. The mechanism of the laser-induced memristive feature has been cleared by dynamic Monte Carlo simulations and first-principles calculations. Furthermore, we verified the universality of the laser-modulation by investigating other 2D materials of TMDs. Our work will open a route to modulate and optimize the performance of 2D semiconductor memristive devices.展开更多
full-time Master of engineering education has developed rapidly in recent years, but in the talent cultivation, professional degree of master of engineering characteristics is not obvious, there are such problems as b...full-time Master of engineering education has developed rapidly in recent years, but in the talent cultivation, professional degree of master of engineering characteristics is not obvious, there are such problems as backward curriculum system, practice teaching is weak, tutor guidance effectiveness is not high, thesis and practice related to small. Colleges and universities need to further adjust and optimize the master of engineering training mode and promote the cooperation between colleges and enterprises, the order-form talent-training model as a handle, promote the depth of integration of production, study and research, and highlight the characteristics of talent cultivation.展开更多
With technology advances and human requirements increasing, human-computer interaction plays an important role in our daily lives. Among these interactions, gesture-based recognition offers a natural and intuitive use...With technology advances and human requirements increasing, human-computer interaction plays an important role in our daily lives. Among these interactions, gesture-based recognition offers a natural and intuitive user experience that does not require physical contact and is becoming increasingly prevalent across various fields. Gesture recognition systems based on Frequency Modulated Continuous Wave (FMCW) millimeter-wave radar are receiving widespread attention due to their ability to operate without wearable sensors, their robustness to environmental factors, and the excellent penetrative ability of radar signals. This paper first reviews the current main gesture recognition applications. Subsequently, we introduce the system of gesture recognition based on FMCW radar and provide a general framework for gesture recognition, including gesture data acquisition, data preprocessing, and classification methods. We then discuss typical applications of gesture recognition systems and summarize the performance of these systems in terms of experimental environment, signal acquisition, signal processing, and classification methods. Specifically, we focus our study on four typical gesture recognition systems, including air-writing recognition, gesture command recognition, sign language recognition, and text input recognition. Finally, this paper addresses the challenges and unresolved problems in FMCW radar-based gesture recognition and provides insights into potential future research directions.展开更多
基金supported by the National Natural Science Foundation of China(62031017,61971221)the Fundamental Research Funds for the Central Universities of China(NP2020104)。
文摘Non-uniform linear array(NULA)configurations are well renowned due to their structural ability for providing increased degrees of freedom(DOF)and wider array aperture than uniform linear arrays(ULAs).These characteristics play a significant role in improving the direction-of-arrival(DOA)estimation accuracy.However,most of the existing NULA geometries are primarily applicable to circular sources(CSs),while they limitedly improve the DOF and continuous virtual aperture for noncircular sources(NCSs).Toward this purpose,we present a triaddisplaced ULAs(Tdis-ULAs)configuration for NCS.The TdisULAs structure generally consists of three ULAs,which are appropriately placed.The proposed antenna array approach fully exploits the non-circular characteristics of the sources.Given the same number of elements,the Tdis-ULAs design achieves more DOF and larger hole-free co-array aperture than its sparse array competitors.Advantageously,the number of uniform DOF,optimal distribution of elements among the ULAs,and precise element positions are uniquely determined by the closed-form expressions.Moreover,the proposed array also produces a filled resulting co-array.Numerical simulations are conducted to show the performance advantages of the proposed Tdis-ULAs configuration over its counterpart designs.
文摘With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors were widely applied due to their low cost. This paper explored the implementation of a human hand posture recognition system using ToF sensors and residual neural networks. Firstly, this paper reviewed the typical applications of human hand recognition. Secondly, this paper designed a hand gesture recognition system using a ToF sensor VL53L5. Subsequently, data preprocessing was conducted, followed by training the constructed residual neural network. Then, the recognition results were analyzed, indicating that gesture recognition based on the residual neural network achieved an accuracy of 98.5% in a 5-class classification scenario. Finally, the paper discussed existing issues and future research directions.
文摘The direct target of engineering education is to cultivate engineers and industry leaders with the qualifi ed skills and strong competencies. To cultivate skilled technological talents in a local engineering college, the fi rst subject is to establish the discipline according to the major strategy of the local economic development requirements, so that the professional regional service function is relatively strong. Secondly, the individual work orientation and development direction are considered as well in the cultivated process. These aspects are required such as renewing education idea, strengthening the education of humanity quality, perfecting the system of continuing education, establishing the mechanism of lifelong learning.
基金Supported by National Key R&D Program of China(Grant No.2022YFB2502900)Fundamental Research Funds for the Central Universities of China,Science and Technology Commission of Shanghai Municipality of China(Grant No.21ZR1465900)Shanghai Gaofeng&Gaoyuan Project for University Academic Program Development of China.
文摘Model predictive control is widely used in the design of autonomous driving algorithms.However,its parameters are sensitive to dynamically varying driving conditions,making it difficult to be implemented into practice.As a result,this study presents a self-learning algorithm based on reinforcement learning to tune a model predictive controller.Specifically,the proposed algorithm is used to extract features of dynamic traffic scenes and adjust the weight coefficients of the model predictive controller.In this method,a risk threshold model is proposed to classify the risk level of the scenes based on the scene features,and aid in the design of the reinforcement learning reward function and ultimately improve the adaptability of the model predictive controller to real-world scenarios.The proposed algorithm is compared to a pure model predictive controller in car-following case.According to the results,the proposed method enables autonomous vehicles to adjust the priority of performance indices reasonably in different scenarios according to risk variations,showing a good scenario adaptability with safety guaranteed.
基金supported by the National Natural Science Foundation of China under Grant No.61701284the Innovative Research Foundation of Qingdao under Grant No.19-6-2-1-CG+5 种基金the Elite Plan Project of Shandong University of Science and Technology under Grant No.skr21-3-B-048the Sci.&Tech.Development Fund of Shandong Province of China under Grant Nos.ZR202102230289,ZR202102250695,and ZR2019LZH001the Humanities and Social Science Research Project of the Ministry of Education under Grant No.18YJAZH017the Taishan Scholar Program of Shandong Province,the Shandong Chongqing Science and Technology Cooperation Project under Grant No.cstc2020jscx-lyjsAX0008the Sci.&Tech.Development Fund of Qingdao under Grant No.21-1-5-zlyj-1-zc,SDUST Research Fund under Grant No.2015TDJH102the Science and Technology Support Plan of Youth Innovation Team of Shandong higher School under Grant No.2019KJN024.
文摘With the development of urbanization,the problem of neurological diseases brought about by population aging has gradually become a social problem of worldwide concern.Aging leads to gradual degeneration of the central nervous system,shrinkage of brain tissue,and decline in physical function in many elderlies,making them susceptible to neurological diseases such as Alzheimer’s disease(AD),stroke,Parkinson’s and major depressive disorder(MDD).Due to the influence of these neurological diseases,the elderly have troubles such as memory loss,inability to move,falling,and getting lost,which seriously affect their quality of life.Tracking and positioning of elderly with neurological diseases and keeping track of their location in real-time are necessary and crucial in order to detect and treat dangerous and unexpected situations in time.Considering that the elderly with neurological diseases forget to wear a positioning device or have mobility problems due to carrying a positioning device,device-free positioning as a passive positioning technology that detects device-free individuals is more suitable than traditional active positioning for the home-based care of the elderly with neurological diseases.This paper provides an extensive and in-depth survey of device-free indoor positioning technology for home-based care and an in-depth analysis of the main features of current positioning systems,as well as the techniques,technologies andmethods they employ,fromthe perspective of the needs of the elderly with neurological conditions.Moreover,evaluation criteria and possible solutions of positioning techniques for the home-based care of the elderly with neurological conditions are proposed.Finally,the opportunities and challenges for the development of indoor positioning technology in 6G mobile networks for home-based care of the elderly with neurological diseases are discussed.This review has provided comprehensive and effective tracking and positioning techniques,technologies and methods for the elderly,by which we can obtain the location information of the elderly in real-time and make home-based care more comfortable and safer for the elderly with neurological diseases.
基金the financial support of the National Key Research and Development Program of China(2020AAA0108100)the Shanghai Municipal Science and Technology Major Project(2021SHZDZX0100)the Shanghai Gaofeng and Gaoyuan Project for University Academic Program Development for funding。
文摘Decision-making and motion planning are extremely important in autonomous driving to ensure safe driving in a real-world environment.This study proposes an online evolutionary decision-making and motion planning framework for autonomous driving based on a hybrid data-and model-driven method.First,a data-driven decision-making module based on deep reinforcement learning(DRL)is developed to pursue a rational driving performance as much as possible.Then,model predictive control(MPC)is employed to execute both longitudinal and lateral motion planning tasks.Multiple constraints are defined according to the vehicle’s physical limit to meet the driving task requirements.Finally,two principles of safety and rationality for the self-evolution of autonomous driving are proposed.A motion envelope is established and embedded into a rational exploration and exploitation scheme,which filters out unreasonable experiences by masking unsafe actions so as to collect high-quality training data for the DRL agent.Experiments with a high-fidelity vehicle model and MATLAB/Simulink co-simulation environment are conducted,and the results show that the proposed online-evolution framework is able to generate safer,more rational,and more efficient driving action in a real-world environment.
基金Supported by the Key Projects of the Guangdong Education Department(No.2019KZDXM019)the Fund of Southern Marine Science and Engineering Guangdong Laboratory(Zhanjiang)(No.ZJW-2019-08)+3 种基金the High-level Marine Discipline Team Project of Guangdong Ocean University(No.002026002009)the Guangdong Graduate Academic Forum Project(No.230420003)the“First Class”Discipline Construction Platform Project in 2019 of Guangdong Ocean University(No.231419026)the Innovation Projects of Colleges and Universities in Guangdong Province(No.2021KQNCX028)。
文摘Colored dissolved organic matter(CDOM)is an important optically active substance in marine environment.Its biochemical conservatism makes it an important indicator to offshore pollution process.Monitoring the content,composition,and diffusion process of CDOM is a good approach to analyze the terrestrial input,optical properties,and ecological environment of offshore areas.The spatiotemporal characteristics of CDOM around the Leizhou Peninsula were analyzed based on field observation data collected in the autumn 2020 and spring 2021,and an empirical inversion model of the ag(355)(absorption coefficient at 355 nm)and spectral slope(Sg)(g stands for gelbstoff/gilvin,which is called CDOM)based on Sentinel-3A ocean and land color instrument(OLCI)images was constructed.The results show(1)the order of average ag(355)value around the Leizhou Peninsula was east coast(0.503/m)>Qiongzhou Strait(0.502/m)>west coast(0.365/m);(2)the best band combinations of CDOM inversion in spring and autumn were(B4+B11)/B3 and(B7-B1)/B6(B stands for the band of spectral images),and the final inversion results are close to the measured results,indicating that the model has good accuracy;(3)the S_(g)value of the CDOM absorption spectrum was fitted to the hyperbolicexponential model.The fitting accuracy of the model was higher than those of the exponential model and the hyperbolic model,and the best S_(g)inversion model was constructed by selecting Sg(275-295)and Sg(250-295)in spring and autumn;(4)the spatial distributions of ag(355)and S_(g)were inverted,and CDOM in the waters around the Leizhou Peninsula originated from terrestrial organic matter carried by coastal aquaculture zones and runoff in the northeast Zhanjiang and the northern Beibu Gulf.
基金supported in part by the National Natural Science Foundation of China under Grant 62171465,62072303,62272223,U22A2031。
文摘By pushing computation,cache,and network control to the edge,mobile edge computing(MEC)is expected to play a leading role in fifth generation(5G)and future sixth generation(6G).Nevertheless,facing ubiquitous fast-growing computational demands,it is impossible for a single MEC paradigm to effectively support high-quality intelligent services at end user equipments(UEs).To address this issue,we propose an air-ground collaborative MEC(AGCMEC)architecture in this article.The proposed AGCMEC integrates all potentially available MEC servers within air and ground in the envisioned 6G,by a variety of collaborative ways to provide computation services at their best for UEs.Firstly,we introduce the AGC-MEC architecture and elaborate three typical use cases.Then,we discuss four main challenges in the AGC-MEC as well as their potential solutions.Next,we conduct a case study of collaborative service placement for AGC-MEC to validate the effectiveness of the proposed collaborative service placement strategy.Finally,we highlight several potential research directions of the AGC-MEC.
基金supported by the Fundamental Research Funds for the Central Universities,South-Central MinZu University(Grant Number:CPT22019).
文摘In this study,we present characteristics of post-sunset GHz scintillation occurrence and their correlations with ionospheric parameters derived from ionosonde observations in high solar activity years(2012−2013)of solar cycle 24 at Sanya(18.3°N,109.6°E;dip lat.:12.8°N),China.The analyzed data include the F_(2)-layer’s critical frequency(foF_(2)),peak height(h_(m)F_(2)),and minimum virtual height(h’F),as well as the scale height around the F_(2)-layer peak(H_(m)),and virtual height(h’F_(5))and true height(hF_(5))measured at 5 MHz.We have investigated relationships between the equinoctial asymmetry of these scintillations and these ionospheric parameters.In addition,we calculate the growth rates of Rayleigh−Taylor instability on the basis of the ionosonde measurements and theoretical models,respectively.We find that the equinoctial asymmetry of scintillation onset time is associated with the scale length of the vertical electron density gradient(L),which has been shown to affect the growth of Rayleigh−Taylor instability at the bottom of the F-layer.The seasonal variations of foF_(2),H_(m)and scale length of vertical electron density gradient appear to cause the seasonal variations of scintillation occurrence;the equinoctial asymmetry of scintillation occurrence rate over low latitudes appears to be related to background electron density and vertical drifts in the F-layer around time of sunset.Further study is required to explain the observed correlational weakness in low latitudes between scintillation strength,represented by the daily maximum S4,and daily maximum values of foF_(2),h_(m)F_(2),h’F,H_(m),and also the drifts.
基金supported in part by the National Natural Science Foundation of China under Grant 62271142in part by the Key Research and Development Program of Jiangsu Province BE2023021+2 种基金in part by the Jiangsu Key Research and Development Program Project under Grant BE2023011-2in part by the Young Scholar Funding of Southeast Universityin part by the Fundamental Research Funds for the Central Universities 2242022k60001。
文摘In this paper,we formulate the precoding problem of integrated sensing and communication(ISAC)waveform as a non-convex quadratically constrained quadratic programming(QCQP),in which the weighted sum of communication multi-user interference(MUI)and the gap between dual-use waveform and ideal radar waveform is minimized with peak-toaverage power ratio(PAPR)constraints.We propose an efficient algorithm based on alternating direction method of multipliers(ADMM),which is able to decouple multiple variables and provide a closed-form solution for each subproblem.In addition,to improve the sensing performance in both spatial and temporal domains,we propose a new criteria to design the ideal radar waveform,in which the beam pattern is made similar to the ideal one and the integrated sidelobe level of the ambiguity function in each target direction is minimized in the region of interest.The limited memory Broyden-Fletcher-Goldfarb-Shanno(LBFGS)algorithm is applied to the design of the ideal radar waveform which works as a reference in the design of the dual-function waveform.Numerical results indicate that the designed dual-function waveform is capable of offering good communication quality of service(QoS)and sensing performance.
基金supported by the National Natural Science Foundation of China(62176218,62176027)the Fundamental Research Funds for the Central Universities(XDJK2020TY003)the Funds for Chongqing Talent Plan(cstc2024ycjh-bgzxm0082)。
文摘The Nesterov accelerated dynamical approach serves as an essential tool for addressing convex optimization problems with accelerated convergence rates.Most previous studies in this field have primarily concentrated on unconstrained smooth con-vex optimization problems.In this paper,on the basis of primal-dual dynamical approach,Nesterov accelerated dynamical approach,projection operator and directional gradient,we present two accelerated primal-dual projection neurodynamic approaches with time scaling to address convex optimization problems with smooth and nonsmooth objective functions subject to linear and set constraints,which consist of a second-order ODE(ordinary differential equation)or differential conclusion system for the primal variables and a first-order ODE for the dual vari-ables.By satisfying specific conditions for time scaling,we demonstrate that the proposed approaches have a faster conver-gence rate.This only requires assuming convexity of the objective function.We validate the effectiveness of our proposed two accel-erated primal-dual projection neurodynamic approaches through numerical experiments.
基金supported in part by the National Key R&D Program of China under Grant 2020YFB1005900the National Natural Science Foundation of China under Grant 62001220+3 种基金the Jiangsu Provincial Key Research and Development Program under Grants BE2022068the Natural Science Foundation of Jiangsu Province under Grants BK20200440the Future Network Scientific Research Fund Project FNSRFP-2021-YB-03the Young Elite Scientist Sponsorship Program,China Association for Science and Technology.
文摘Collaborative edge computing is a promising direction to handle the computation intensive tasks in B5G wireless networks.However,edge computing servers(ECSs)from different operators may not trust each other,and thus the incentives for collaboration cannot be guaranteed.In this paper,we propose a consortium blockchain enabled collaborative edge computing framework,where users can offload computing tasks to ECSs from different operators.To minimize the total delay of users,we formulate a joint task offloading and resource optimization problem,under the constraint of the computing capability of each ECS.We apply the Tammer decomposition method and heuristic optimization algorithms to obtain the optimal solution.Finally,we propose a reputation based node selection approach to facilitate the consensus process,and also consider a completion time based primary node selection to avoid monopolization of certain edge node and enhance the security of the blockchain.Simulation results validate the effectiveness of the proposed algorithm,and the total delay can be reduced by up to 40%compared with the non-cooperative case.
基金Shenzhen Fund for Guangdong Provincial High-Level Clinical Key Specialties(SZGSP014)Sanming Project of Medicine in Shenzhen(SZSM202311012)Shenzhen Science and Technology Planning Project(KCXFZ20211020163813019).
文摘Age-related macular degeneration(AMD)ranks third among the most common causes of blindness.As the most conventional and direct method for identifying AMD,color fundus photography has become prominent owing to its consistency,ease of use,and good quality in extensive clinical practice.In this study,a convolutional neural network(CSPDarknet53)was combined with a transformer to construct a new hybrid model,HCSP-Net.This hybrid model was employed to tri-classify color fundus photography into the normal macula(NM),dry macular degeneration(DMD),and wet macular degeneration(WMD)based on clinical classification manifestations,thus identifying and resolving AMD as early as possible with color fundus photography.To further enhance the performance of this model,grouped convolution was introduced in this study without significantly increasing the number of parameters.HCSP-Net was validated using an independent test set.The average precision of HCSPNet in the diagnosis of AMD was 99.2%,the recall rate was 98.2%,the F1-Score was 98.7%,the PPV(positive predictive value)was 99.2%,and the NPV(negative predictive value)was 99.6%.Moreover,a knowledge distillation approach was also adopted to develop a lightweight student network(SCSP-Net).The experimental results revealed a noteworthy enhancement in the accuracy of SCSP-Net,rising from 94%to 97%,while remarkably reducing the parameter count to a quarter of HCSP-Net.This attribute positions SCSP-Net as a highly suitable candidate for the deployment of resource-constrained devices,which may provide ophthalmologists with an efficient tool for diagnosing AMD.
基金supported by Guangdong Basic and Applied Basic Research Foundation under Grant 2024A1515012015supported in part by the National Natural Science Foundation of China under Grant 62201336+4 种基金in part by Guangdong Basic and Applied Basic Research Foundation under Grant 2024A1515011541supported in part by the National Natural Science Foundation of China under Grant 62371344in part by the Fundamental Research Funds for the Central Universitiessupported in part by Knowledge Innovation Program of Wuhan-Shuguang Project under Grant 2023010201020316in part by Guangdong Basic and Applied Basic Research Foundation under Grant 2024A1515010247。
文摘In recent times,various power control and clustering approaches have been proposed to enhance overall performance for cell-free massive multipleinput multiple-output(CF-mMIMO)networks.With the emergence of deep reinforcement learning(DRL),significant progress has been made in the field of network optimization as DRL holds great promise for improving network performance and efficiency.In this work,our focus delves into the intricate challenge of joint cooperation clustering and downlink power control within CF-mMIMO networks.Leveraging the potent deep deterministic policy gradient(DDPG)algorithm,our objective is to maximize the proportional fairness(PF)for user rates,thereby aiming to achieve optimal network performance and resource utilization.Moreover,we harness the concept of“divide and conquer”strategy,introducing two innovative methods termed alternating DDPG(A-DDPG)and hierarchical DDPG(H-DDPG).These approaches aim to decompose the intricate joint optimization problem into more manageable sub-problems,thereby facilitating a more efficient resolution process.Our findings unequivo-cally showcase the superior efficacy of our proposed DDPG approach over the baseline schemes in both clustering and downlink power control.Furthermore,the A-DDPG and H-DDPG obtain higher performance gain than DDPG with lower computational complexity.
基金the National Natural Science Foundation of China(Grant Nos.62102347,62376041,62172352)Guangdong Ocean University Research Fund Project(Grant No.060302102304).
文摘Point-of-interest(POI)recommendations in location-based social networks(LBSNs)have developed rapidly by incorporating feature information and deep learning methods.However,most studies have failed to accurately reflect different users’preferences,in particular,the short-term preferences of inactive users.To better learn user preferences,in this study,we propose a long-short-term-preference-based adaptive successive POI recommendation(LSTP-ASR)method by combining trajectory sequence processing,long short-term preference learning,and spatiotemporal context.First,the check-in trajectory sequences are adaptively divided into recent and historical sequences according to a dynamic time window.Subsequently,an adaptive filling strategy is used to expand the recent check-in sequences of users with inactive check-in behavior using those of similar active users.We further propose an adaptive learning model to accurately extract long short-term preferences of users to establish an efficient successive POI recommendation system.A spatiotemporal-context-based recurrent neural network and temporal-context-based long short-term memory network are used to model the users’recent and historical checkin trajectory sequences,respectively.Extensive experiments on the Foursquare and Gowalla datasets reveal that the proposed method outperforms several other baseline methods in terms of three evaluation metrics.More specifically,LSTP-ASR outperforms the previously best baseline method(RTPM)with a 17.15%and 20.62%average improvement on the Foursquare and Gowalla datasets in terms of the Fβmetric,respectively.
基金supported by the National Natural Science Foundation of China(Grant No.32250410309 and 52105582)Natural Science Foundation of Guangdong Province(Grant No.2022A1515010894 and 2022B0303040002)+1 种基金Fundamental Research Foundation of Shenzhen(JCYJ20210324095210030 and JCYJ20220818095810023)Shenzhen-Hong Kong-Macao S&T Program(Category C:SGDX20210823103200004)
文摘Extending the ionic conductivity is the pre-requisite of electrolytes in fuel cell technology for high-electrochemical performance.In this regard,the introduction of semiconductor-oxide materials and the approach of heterostructure formation by modulating energy bands to enhance ionic conduction acting as an electrolyte in fuel cell-device.Semiconductor(n-type;SnO_(2))plays a key role by introducing into p-type SrFe_(0.2)Ti_(0.8)O_(3-δ)(SFT)semiconductor perovskite materials to construct p-n heterojunction for high ionic conductivity.Therefore,two different composites of SFT and SnO_(2)are constructed by gluing p-and n-type SFT-SnO_(2),where the optimal composition of SFT-SnO_(2)(6∶4)heterostructure electrolyte-based fuel cell achieved excellent ionic conductivity 0.24 S cm^(-1)with power-output of 1004 mW cm^(-2)and high OCV 1.12 V at a low operational temperature of 500℃.The high power-output and significant ionic conductivity with durable operation of 54 h are accredited to SFT-SnO_(2)heterojunction formation including interfacial conduction assisted by a built-in electric field in fuel cell device.Moreover,the fuel conversion efficiency and considerable Faradaic efficiency reveal the compatibility of SFT-SnO_(2)heterostructure electrolyte and ruled-out short-circuiting issue.Further,the first principle calculation provides sufficient information on structure optimization and energy-band structure modulation of SFT-SnO_(2).This strategy will provide new insight into semiconductor-based fuel cell technology to design novel electrolytes.
基金National Natural Science Foundation of China,Grant/Award Number:62172139Science Research Project of Hebei Province,Grant/Award Number:CXY2024031+3 种基金Natural Science Foundation of Hebei Province,Grant/Award Number:F2022201055Project Funded by China Postdoctoral,Grant/Award Number:2022M713361Natural Science Interdisciplinary Research Program of Hebei University,Grant/Award Number:DXK202102Open Project Program of the National Laboratory of Pattern Recognition,Grant/Award Number:202200007。
文摘Diagnosing individuals with autism spectrum disorder(ASD)accurately faces great chal-lenges in clinical practice,primarily due to the data's high heterogeneity and limited sample size.To tackle this issue,the authors constructed a deep graph convolutional network(GCN)based on variable multi‐graph and multimodal data(VMM‐DGCN)for ASD diagnosis.Firstly,the functional connectivity matrix was constructed to extract primary features.Then,the authors constructed a variable multi‐graph construction strategy to capture the multi‐scale feature representations of each subject by utilising convolutional filters with varying kernel sizes.Furthermore,the authors brought the non‐imaging in-formation into the feature representation at each scale and constructed multiple population graphs based on multimodal data by fully considering the correlation between subjects.After extracting the deeper features of population graphs using the deep GCN(DeepGCN),the authors fused the node features of multiple subgraphs to perform node classification tasks for typical control and ASD patients.The proposed algorithm was evaluated on the Autism Brain Imaging Data Exchange I(ABIDE I)dataset,achieving an accuracy of 91.62%and an area under the curve value of 95.74%.These results demon-strated its outstanding performance compared to other ASD diagnostic algorithms.
基金supported by the National Natural Science Foundation of China(Nos.51971070,10974037,and 62205011)the National Key Research and Development Program of China(No.2016YFA0200403)+6 种基金Eu-FP7 Project(No.247644)CAS Strategy Pilot Program(No.XDA 09020300)Fundamental Research Funds for the Central Universities(No.buctrc202122)the Open Research Project of Zhejiang province Key Laboratory of Quantum Technology and Device(No.20220401)the Open Research Project of Special Display and Imaging Technology Innovation Center of Anhui Province(No.2022AJ05001)funded by the Ph.D Foundation of Hebei University of Water Resources and Electric Engineering(No.SYBJ2202)Funded by Science and Technology Project of Hebei Education Department(No.BJK2022027)。
文摘Memristors as non-volatile memory devices have gained numerous attentions owing to their advantages in storage,in-memory computing, synaptic applications, etc. In recent years, two-dimensional(2D) materials with moderate defects have been discovered to exist memristive feature. However, it is very difficult to obtain moderate defect degree in 2D materials, and studied on modulation means and mechanism becomes urgent and essential. In this work, we realized memristive feature with a bipolar switching and a configurable on/off ratio in a two-terminal MoS_(2) device(on/off ratio ~100), for the first time, from absent to present using laser-modulation to few-layer defect-free MoS_(2)(about 10 layers), and its retention time in both high resistance state and low resistance state can reach 2×10^(4) s. The mechanism of the laser-induced memristive feature has been cleared by dynamic Monte Carlo simulations and first-principles calculations. Furthermore, we verified the universality of the laser-modulation by investigating other 2D materials of TMDs. Our work will open a route to modulate and optimize the performance of 2D semiconductor memristive devices.
文摘full-time Master of engineering education has developed rapidly in recent years, but in the talent cultivation, professional degree of master of engineering characteristics is not obvious, there are such problems as backward curriculum system, practice teaching is weak, tutor guidance effectiveness is not high, thesis and practice related to small. Colleges and universities need to further adjust and optimize the master of engineering training mode and promote the cooperation between colleges and enterprises, the order-form talent-training model as a handle, promote the depth of integration of production, study and research, and highlight the characteristics of talent cultivation.
文摘With technology advances and human requirements increasing, human-computer interaction plays an important role in our daily lives. Among these interactions, gesture-based recognition offers a natural and intuitive user experience that does not require physical contact and is becoming increasingly prevalent across various fields. Gesture recognition systems based on Frequency Modulated Continuous Wave (FMCW) millimeter-wave radar are receiving widespread attention due to their ability to operate without wearable sensors, their robustness to environmental factors, and the excellent penetrative ability of radar signals. This paper first reviews the current main gesture recognition applications. Subsequently, we introduce the system of gesture recognition based on FMCW radar and provide a general framework for gesture recognition, including gesture data acquisition, data preprocessing, and classification methods. We then discuss typical applications of gesture recognition systems and summarize the performance of these systems in terms of experimental environment, signal acquisition, signal processing, and classification methods. Specifically, we focus our study on four typical gesture recognition systems, including air-writing recognition, gesture command recognition, sign language recognition, and text input recognition. Finally, this paper addresses the challenges and unresolved problems in FMCW radar-based gesture recognition and provides insights into potential future research directions.
