This study focuses on the teaching reform of the communication application development course based on the core requirements of engineering education accreditation.To address key challenges such as the disconnection b...This study focuses on the teaching reform of the communication application development course based on the core requirements of engineering education accreditation.To address key challenges such as the disconnection between software and hardware teaching and insufficient practical skills among students,a project-driven“learning-practiceapplication”teaching model is proposed.By optimizing course content,innovating teaching methods,and introducing university-industry collaboration mechanisms,the reform aligns the curriculum more closely with engineering education standards and industry demands.The approach significantly enhances students’comprehensive skills,practical abilities,and employability.This study provides theoretical foundations and practical strategies for the teaching reform of courses in communication engineering.展开更多
The GaN-based heterostructures are widely used in optoelectronic devices,but the complex surface reconstructions and lattice mismatch greatly limit the applications.The stacking of two-dimensional transition metal dic...The GaN-based heterostructures are widely used in optoelectronic devices,but the complex surface reconstructions and lattice mismatch greatly limit the applications.The stacking of two-dimensional transition metal dichalcogenide(TMD=MoS_(2),MoSSe and MoSe_(2))monolayers on reconstructed GaN surface not only effectively overcomes the larger mismatch,but also brings about novel electronic and optical properties.By adopting the reconstructed GaN(0001)surface with adatoms(N-ter GaN and Ga-ter GaN),the influences of complicated surface conditions on the electronic properties of heterostructures have been investigated.The passivated N-ter and Ga-ter GaN surfaces push the mid-gap states to the valence bands,giving rise to small bandgaps in heterostructures.The charge transfer between Ga-ter GaN surface and TMD monolayers occurs much easier than that across the TMD/N-ter GaN interfaces,which induces stronger interfacial interaction and larger valence band offset(VBO).The band alignment can be switched between type-I and type-II by assembling different TMD monolayers,that is,MoS_(2)/N-ter GaN and MoS_(2)/Ga-ter GaN are type-II,and the others are type-I.The absorption of visible light is enhanced in all considered TMD/reconstructed GaN heterostructures.Additionally,MoSe_(2)/Ga-ter GaN and MoSSe/N-ter GaN have larger conductor band offset(CBO)of 1.32 eV and 1.29 eV,respectively,extending the range from deep ultraviolet to infrared regime.Our results revel that the TMD/reconstructed GaN heterostructures may be used for high-performance broadband photoelectronic devices.展开更多
The rapid expansion of the low-altitude economy is driving strong demand for highly accurate and reliable positioning technologies to support diverse aerial operations.This review examines core positioning methodologi...The rapid expansion of the low-altitude economy is driving strong demand for highly accurate and reliable positioning technologies to support diverse aerial operations.This review examines core positioning methodologies within the low-altitude intelligent network(LAIN)framework,beginning with an analysis of positioning requirements and performance metrics for low-altitude flight scenarios.It systematically assesses the principles,strengths,and limitations of mainstream positioning systems,including Global Navigation Satellite Systems(GNSS),terrestrial wireless positioning,and autonomous navigation,and it surveys prevalent integrated and cooperative positioning schemes.Our analysis demonstrates that standalone positioning technologies are inadequate in complex low-altitude settings,underscoring the pivotal role of multi-source fusion and unmanned aerial vehicle(UAV)swarm cooperative positioning as future trends.To address infrastructure gaps and high deployment costs in current LAIN systems,we propose a“space−air−ground”integrated and cooperative positioning architecture centered on GNSS and the 5th generation mobile communication technology(5G).The ground layer integrates 5G and GNSS for wide-area enhanced positioning.The aerial layer uses 5G aircraft-to-everything(A2X)and sidelink(SL)communications to build self-organizing networks for cooperative UAV localization.The space layer leverages low Earth orbit(LEO)satellites to overcome coverage limitations in communication and positioning.This hierarchical architecture reduces deployment costs through infrastructure reuse and enables deep integration of communication and navigation capabilities.By supporting collaborative enhancement across all three domains,the framework improves positioning robustness and delivers cost-effective,ubiquitous,and highly reliable positioning services.Finally,we outline promising research directions.This review aims to provide a systematic reference and a novel architectural perspective for the ongoing development of LAIN.展开更多
β-Ga_(2)O_(3) MOS inverter should play a crucial role in β-Ga_(2)O_(3) electronic circuits. Enhancement-mode(E-mode) MOSFET was fabricated based on β-Ga_(2)O_(3) film grown by atomic layer deposition technology, an...β-Ga_(2)O_(3) MOS inverter should play a crucial role in β-Ga_(2)O_(3) electronic circuits. Enhancement-mode(E-mode) MOSFET was fabricated based on β-Ga_(2)O_(3) film grown by atomic layer deposition technology, and the β-Ga_(2)O_(3) inverter was further monolithically integrated on this basis. The β-Ga_(2)O_(3) n MOSFET exhibits excellent electrical characteristics with an on/off current ratio reaching 10^(5). The logic inverter shows outstanding voltage inversion characteristics under low-frequency from 1 to 400 Hz operation. As the frequency continues to increase to 10 K, the reverse characteristic becomes worse due to parasitic capacitance induced by processes, and the difference between the highest and lowest values of VOUT has an exponential decay relationship with the frequency. This paper provides the practice for the development of β-Ga_(2)O_(3)-based circuits.展开更多
Ambient-air,moisture-assisted annealing is widely used in fabricating perovskite solar cells(PSCs).However,the inherent sensitivity of perovskite intermediate-phase to moisture—due to fast and spontaneous intermolecu...Ambient-air,moisture-assisted annealing is widely used in fabricating perovskite solar cells(PSCs).However,the inherent sensitivity of perovskite intermediate-phase to moisture—due to fast and spontaneous intermolecular exchange reaction—requires strict control of ambient humidity and immediate thermal annealing treatment,raising manufacturing costs and causing fast nucleation of perovskite films.We report herein a self-buffered molecular migration strategy to slow down the intermolecular exchange reaction by introducing a n-butylammonium bromide shielding layer,which limits moisture diffusion into intermediate-phase film.This further endows the notably wide nucleation time and humidity windows for perovskite crystallization in ambient air.Consequently,the optimized 1.68 e V-bandgap n-i-p structured PSC reaches a record-high reverse-scan(RS)PCE of 22.09%.Furthermore,the versatility and applicability of as-proposed self-buffered molecular migration strategy are certified by employing various shielding materials and 1.53 eV-/1.77 eV-bandgap perovskite materials.The n-i-p structured PSCs based on 1.53 eV-and 1.77 eV-bandgap perovskite films achieve outstanding RS PCEs of 25.23%and 19.09%,respectively,both of which are beyond of the state-of-the-art ambient-air processed PSCs.展开更多
A facile one-step hydrothermal method has been reported to synthesize theα-Fe_(2)O_(3)nanosheet arrays with the preferred orientation along the[104]direction on the ITO substrate.Theα-Fe_(2)O_(3)nanosheet arrays-bas...A facile one-step hydrothermal method has been reported to synthesize theα-Fe_(2)O_(3)nanosheet arrays with the preferred orientation along the[104]direction on the ITO substrate.Theα-Fe_(2)O_(3)nanosheet arrays-based W/α-Fe_(2)O_(3)/ITO memristor has been achieved by depositing the circular W top electrodes on theα-Fe_(2)O_(3)nanosheet arrays.The as-prepared W/α-Fe_(2)O_(3)/ITO memristor shows a reliable nonvolatile bipolar resistive switching behavior with the high resistance ratio of about 103at the reading voltage of 0.1 V,good resistance retention over 10~3s,ultralow set voltage of-0.6 V and reset voltage of 0.7 V,and good durability.In addition,the tunneling conduction mechanism modified by the oxygen vacancies has been proposed and suggested to be responsible for the nonvolatile resistive switching behavior of the as-prepared W/α-Fe_(2)O_(3)/ITO memristor.This work demonstrates that the as-preparedα-Fe_(2)O_(3)nanosheet arrays-based W/α-Fe_(2)O_(3)/ITO memristor would be a promising candidate for further ultralow power nonvolatile memory applications.展开更多
An advanced AlCrSiN/AlCrN/CrN/Cr multilayer coating was developed via hybrid multiarc ion plating and high-power impulse magnetron sputtering.The multilayer design enhanced the substrate-coating compatibility,achievin...An advanced AlCrSiN/AlCrN/CrN/Cr multilayer coating was developed via hybrid multiarc ion plating and high-power impulse magnetron sputtering.The multilayer design enhanced the substrate-coating compatibility,achieving a critical load of 87.8 N.Silicon doping induced nanocrystallization and amorphization,increasing the hardness to 26 GPa.At high temperatures,a nanoscale Cr-rich(Cr,Al)_(2)O_(3) layer was formed,effectively inhibiting oxygen diffusion.The coating underwent unique phase transformations,during which Cr_(2)N and amorphous Si3N4 were converted into dispersed SiCr_(3) nanoparticles,which stabilized Cr atoms and suppressed their outward diffusion.Ab initio molecular dynamics simulations revealed that Cr atoms exhibited higher chemical activity and oxygen-capture capability than Al atoms and Si atoms served as diffusion barriers by pinning onto the oxidized surface,considerably improving the oxidation resistance of the coating.展开更多
The prospect ofα-Ga2O3 in optical and electrical devices application is fascinating.In order to obtain better performance,Ge and F elements with similar electronegativity and atomic size are selected as dopants.Based...The prospect ofα-Ga2O3 in optical and electrical devices application is fascinating.In order to obtain better performance,Ge and F elements with similar electronegativity and atomic size are selected as dopants.Based on density functional theory(DFT),we systematically research the electronic structure and optical properties of dopedα-Ga2O3 by GGA+U calculation method.The results show that Ge atoms and F atoms are effective n-type dopants.For Ge-dopedα-Ga2O3,it is probably obtained under O-poor conditions.However,for F-dopedα-Ga2O3,it is probably obtained under O-rich conditions.The doping system of F element is more stable due to the lower formation energy.In this investigation,it is found that two kinds of doping can reduce theα-Ga2O3 band gap and improve the conductivity.What is more,it is observed that the absorption edge after doping has a blue shift and causes certain absorption effect on the visible region.Through the whole scale of comparison,Ge doping is more suitable for the application of transmittance materials,yet F doping is more appropriate for the application of deep ultraviolet devices.We expect that our research can provide guidance and reference for preparation ofα-Ga2O3 thin films and photoelectric devices.展开更多
Based on the region model of lambda bipolar transistor ( LBT), a dividing region theory model of PLBT is set up,simulated and verified. Firstly, the principal operations of different kinds of photoelectronic lambda bi...Based on the region model of lambda bipolar transistor ( LBT), a dividing region theory model of PLBT is set up,simulated and verified. Firstly, the principal operations of different kinds of photoelectronic lambda bipolar transistor ( PLBT) are characterized by a simple circuit model. Through mathematical analysis of the equivalent circuit, the typical characteristics curve is divided into positive resistance, peak, negative resistance and cutoff regions. Secondly, by analyzing and simulating this model, the ratio of MOSFET width to channel length, threshold voltage and common emitter gain are discovered as the main structure parameters that determine the characteristic curves of PLBT. And peak region width, peak current value, negative resistance value and valley voltage value of PLBT can be changed conveniently according to the actual demands by modifying these parameters. Finally comparisons of the characteristics of the fabricated devices and the simu- lation results are made, which show that the analytical results are in agreement with the observed devices characteristics.展开更多
Two-dimensional(2D)materials offer novel platforms to meet the increasing demands of next-generation miniaturized electronics.Among them,the recently emerged 2D Bi_(2)O_(2)Se with unique non-van der Waals interlayer i...Two-dimensional(2D)materials offer novel platforms to meet the increasing demands of next-generation miniaturized electronics.Among them,the recently emerged 2D Bi_(2)O_(2)Se with unique non-van der Waals interlayer interaction,high mobility,sizeable bandgap,and capability to fabricate homologous heterojunction,is of particular interest.In this Review,we introduce recent progress in preparation,transfer,mechanical and electrical properties,and electronic applications of 2D Bi_(2)O_(2)Se.First,we summarize methodologies to synthesize and massively produce 2D Bi_(2)O_(2)Se,as well as recent advances in transferring them from growth substrate to arbitrary substrates.Then,we review current understandings on the intrinsic mechanical properties of Bi_(2)O_(2)Se at 2D thickness limit,and its in-plane and out-of-plane electrical properties.Electronic devices including field-effect transistors,memristors,and sensors based on 2D Bi_(2)O_(2)Se for neuromorphic computing,memory,logic,and integrated circuits are discussed.Finally,challenges and prospects for the development of 2D Bi_(2)O_(2)Se are proposed.展开更多
With the rapid development of communication and computer,the individual identification technology of communication equipment has been brought to many application scenarios.The identification of the same type of electr...With the rapid development of communication and computer,the individual identification technology of communication equipment has been brought to many application scenarios.The identification of the same type of electronic equipment is of considerable significance,whether it is the identification of friend or foe in military applications,identity determination,radio spectrum management in civil applications,equipment fault diagnosis,and so on.Because of the limited-expression ability of the traditional electromagnetic signal representation methods in the face of complex signals,a new method of individual identification of the same equipment of communication equipment based on deep learning is proposed.The contents of this paper include the following aspects:(1)Considering the shortcomings of deep learning in processing small sample data,this paper provides a universal and robust feature template for signal data.This paper constructs a relatively complete signal template library from multiple perspectives,such as time domain and transform domain features,combined with high-order statistical analysis.Based on the inspiration of the image texture feature,characteristics of amplitude histogram of signal and the signal amplitude co-occurrence matrix(SACM)are proposed in this paper.These signal features can be used as a signal fingerprint template for individual identification.(2)Considering the limitation of the recognition rate of a single classifier,using the integrated classifier has achieved better generalization ability.The final average accuracy of 5 NRF24LE1 modules is up to 98%and solved the problem of individual identification of the same equipment of communication equipment under the condition of the small sample,low signal-to-noise ratio.展开更多
The rapid evolution of Fifth-Generation(5G)networks and the strategic development of Sixth-Generation(6G)technologies have significantly advanced the implementation of air-ground integrated networks with seamless cove...The rapid evolution of Fifth-Generation(5G)networks and the strategic development of Sixth-Generation(6G)technologies have significantly advanced the implementation of air-ground integrated networks with seamless coverage.Unmanned Aerial Vehicles(UAVs),serving as high-mobility aerial platforms,are extensively utilized to enhance coverage in long-distance emergency communication scenarios.The resource-constrained communication environments in emergencies by classifying UAVs into swarm UAVs and relay UAVs as aerial communication nodes is inversitgated.A horizontal deployment strategy for swarm UAVs is formulated through K-means clustering algorithm optimization,while a vertical deployment scheme is established using convex optimization methods.The minimum-path trajectory planning for relay UAVs is optimized via the Particle Swarm Optimization(PSO)algorithm,enhancing communication reliability between UAV swarms and terrestrial base stations.A three-dimensional heterogeneous network architecture is realized by modeling spatial multi-hop relay links.Experimental results demonstrate that the proposed joint UAV relay optimization framework outperforms conventional algorithms in both coverage performance and relay capability during video stream transmission,achieving significant improvements in coverage enhancement and relay efficiency.This work provides technical foundations for constructing high-reliability air-ground cooperative systems in emergency communications.展开更多
Fiber optic temperature sensors stand out in a variety of applications due to their small size,chemical resistance,and resistance to electromagnetic interference.The traditional optical fiber temperature sensor direct...Fiber optic temperature sensors stand out in a variety of applications due to their small size,chemical resistance,and resistance to electromagnetic interference.The traditional optical fiber temperature sensor directly places the sensing structure in the temperature to be measured,and uses the thermo-optical effect and thermal expansion effect of the SiO_(2)material that constitutes the sensing structure to achieve measurement,while the thermo-optical coefficient and thermal expansion coefficient of SiO_(2) are very small,which limits the high sensitivity response characteristics of the optical fiber temperature sensing structure.In order to solve the problem of low sensitivity of traditional optical fiber temperature sensors,a Mach-Zehnder interferometric temperature sensor with a liquid-encapsulated tapered microfiber is developed.The sensor converts the temperature change into a change in the refractive index of the liquid material and thus realizes the measurement of temperature.In the range of 25~50℃,as the temperature increases,the wavelength of the transmission spectrum shifts towards shorter wavelengths.Experimental results show that the sensitivity of the liquid encapsulated microfiber interferometric temperature sensor can reach-57.91 nk·nm^(-1).This sensor has great potential for applications in marine environmental monitoring,biomedical diagnosis,and aerospace.展开更多
Developing highly efficient and recyclable photocatalysts has been regarded as an attractive strategy to solve antibiotic contaminants.Herein,we designed and fabricated Cy-C_(3) N_(4)/TiO_(2) S-scheme heterojunction f...Developing highly efficient and recyclable photocatalysts has been regarded as an attractive strategy to solve antibiotic contaminants.Herein,we designed and fabricated Cy-C_(3) N_(4)/TiO_(2) S-scheme heterojunction film with boosted charge transfer and a highly hydrophilic surface.The as-prepared heterojunction exhibited outstanding removal efficiency on tetracyclines and fluoroquinolone antibiotics(more than 80% within 90 min).The removal rate of 300-Cy-C_(3) N_(4)/TiO_(2) on norfloxacin(NOR)was 2.12,and 1.59 times higher than that of pristine TiO_(2),C_(3) N_(4)/TiO_(2),respectively.The excellent photocatalytic performance of 300-Cy-C_(3) N_(4)/TiO_(2) was attributed to the highly hydrophilic surface and effective transfer and separation of carriers.Moreover,the NOR degradation pathways were proposed based on the results of density functional theory(DFT),and liquid chromatography-mass spectrometry.The toxicity assessment indicated the toxicity of intermediates can be remarkably alleviated.The DFT calculation and selective photo-deposition experiment demonstrated that an internal electric field was formed at the heterojunction interface,and the charge carriers migrated between Cy-C_(3) N_(4) and TiO_(2) following an S-scheme transfer pathway.This research not only provides a promising method for tracking charge distribution on thin-film heterojunction photocatalysts but also helps us to design high-efficiency,and recyclable heterojunctions to solve antibiotic contaminants.展开更多
The fiber optic sensing technology provides data support in structural health monitoring of the macro facilities,including design,construction,and maintenance of bridges,tunnels,ports and other infrastructures.In this...The fiber optic sensing technology provides data support in structural health monitoring of the macro facilities,including design,construction,and maintenance of bridges,tunnels,ports and other infrastructures.In this paper,a distributed vibration sensing system is proved to be responsive to a single touch over a 1.8-m-long equivalent fiber segment,covering a vibration frequency from 5 Hz to 25 kHz.The sensing fiber was arranged as an S type layout on the bridge to recognize the standing state,windblown disturbance,and walking vibration.Moreover,the knocking and climbing events are recognized fiber laying spinning lines and hanging on the fences,respectively.The demonstration shows an accurate positioning and sensitive vibration monitoring applied on the automated three-dimensional(3D) printed bridge,which is applicable to all kinds of 3D printed facilities as intelligent sensory neuro-networks.展开更多
A wideband low-profile aperture-coupled antenna based on a novel dual-mode-composite scheme is presented.The mode-composite scheme where the TM10 cavity mode and the TE121 dielec-tric resonator(DR)mode are combined of...A wideband low-profile aperture-coupled antenna based on a novel dual-mode-composite scheme is presented.The mode-composite scheme where the TM10 cavity mode and the TE121 dielec-tric resonator(DR)mode are combined offers an ap-proach to obtain a wide bandwidth accompanied by stable unidirectional radiation and high efficiency.The use of a lengthened coupling aperture that supports the one-wavelength resonance in the band of interest is an effective feed method of simultaneously excit-ing the two composite modes without compromising the increased complexity of the antenna geometry.An impedance bandwidth of 49%for|S_(11)|of less than-10 dB,a maximum gain of 10.8 dBi,and stable radiation patterns with low cross-polarization are realized ex-perimentally by a fabricated prototype.Considering the simplicity of the geometry,the wide bandwidth that can cover n77,n78,and n79 bands for the fifth generation(5G)mobile communications and the sat-isfying radiation performance,the proposed antenna would be a promising candidate for advanced wireless applications.展开更多
Four-level pulse amplitude modulation(PAM4)signals,recognized for enhanced energy efficiency and spectral utilization compared with non-return-to-zero(NRZ)counterparts,have been adopted in multiple high-speed serializ...Four-level pulse amplitude modulation(PAM4)signals,recognized for enhanced energy efficiency and spectral utilization compared with non-return-to-zero(NRZ)counterparts,have been adopted in multiple high-speed serializer/deserializer(SerDes)standards,but NRZ modulation remains predominant in industrial applications.This paper introduces a UMC 28 nm CMOS-based parallel configurable forward feedback equalization(FFE)dual-mode high-speed SerDes transmitter supporting 7-bit resolution with data rates of 56 Gb∙s^(-1)NRZ and 112 Gb∙s^(-1)PAM4,utilizing a hybrid architecture that integrates digital signal processing(DSP)with digital-to-analog conversion(DAC).The design processes parallel input signals and eight stored 8-bit tap coefficients through a configurable FFE multiplier module and parallel carry adder module,while achieving low-power serialization via low-speed 16∶4 multiplexers(MUXs)with two different 2∶1 MUXs and high-speed 4∶1 MUXs.A source series termination(SST)output network structure enhances lower power dissipation and higher output swing.Simulation results show that,under a 1.05 V supply voltage and a channel loss of 19.21 dB at 28 GHz,the output 56 Gb∙s^(-1)NRZ eye diagram has an eye height of 70.11 mV and an eye width of 12.16 ps(0.68 UI).The output 112 Gb∙s^(-1)PAM4 eye diagram has an eye height of 20.07 mV and an eye width of 7.49 ps(0.42 UI).The layout area of the dual-mode transmitter is 0.079 mm^(2),and the total circuit power consumption is 74.48 mW(energy efficiency is 1.33/0.67 pJ∙bit-1).展开更多
The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect...The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.展开更多
This paper proposes a novel method for the automatic diagnosis of keratitis using feature vector quantization and self-attention mechanisms(ADK_FVQSAM).First,high-level features are extracted using the DenseNet121 bac...This paper proposes a novel method for the automatic diagnosis of keratitis using feature vector quantization and self-attention mechanisms(ADK_FVQSAM).First,high-level features are extracted using the DenseNet121 backbone network,followed by adaptive average pooling to scale the features to a fixed length.Subsequently,product quantization with residuals(PQR)is applied to convert continuous feature vectors into discrete features representations,preserving essential information insensitive to image quality variations.The quantized and original features are concatenated and fed into a self-attention mechanism to capture keratitis-related features.Finally,these enhanced features are classified through a fully connected layer.Experiments on clinical low-quality(LQ)images show that ADK_FVQSAM achieves accuracies of 87.7%,81.9%,and 89.3% for keratitis,other corneal abnormalities,and normal corneas,respectively.Compared to DenseNet121,Swin transformer,and InceptionResNet,ADK_FVQSAM improves average accuracy by 3.1%,11.3%,and 15.3%,respectively.These results demonstrate that ADK_FVQSAM significantly enhances the recognition performance of keratitis based on LQ slit-lamp images,offering a practical approach for clinical application.展开更多
Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and...Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and effective management of traffic conditions.Traditional forecasting models often struggle to adequately capture these complexities,leading to suboptimal predictive performance.While neural networks excel at modeling intricate and nonlinear data structures,they are also highly susceptible to overfitting,resulting in inefficient use of computational resources and decreased model generalization.This paper introduces a novel heuristic feature extraction method that synergistically combines the strengths of non-neural network algorithms with neural networks to enhance the identification and representation of relevant features from traffic data.We begin by evaluating the significance of various temporal characteristics using three distinct assessment strategies grounded in non-neural methodologies.These evaluated features are then aggregated through a weighted fusion mechanism to create heuristic features,which are subsequently integrated into neural network models for more accurate and robust traffic prediction.Experimental results derived from four real-world datasets,collected from diverse urban environments,show that the proposed method significantly improves the accuracy of long-term traffic forecasting without compromising performance.Additionally,the approach helps streamline neural network architectures,leading to a considerable reduction in computational overhead.By addressing both prediction accuracy and computational efficiency,this study not only presents an innovative and effective method for traffic condition forecasting but also offers valuable insights that can inform the future development of data-driven traffic management systems and transportation strategies.展开更多
基金Quality Engineering Project of Higher Education Institutions in Anhui Province(2023aqnujyxm26,2023sx060,2023zyxwjxalk124)Natural Science Key Research Project for Higher Education Institutions of Anhui Province(2024AH051117,2024AH051126)+1 种基金Excellent Young Backbone Teachers’Domestic and Foreign Visiting and Training Program in Universities(gxgnfx20220262022)Research and Industrialization Project of High Precision Positioning System for Intelligent Connected Vehicles。
文摘This study focuses on the teaching reform of the communication application development course based on the core requirements of engineering education accreditation.To address key challenges such as the disconnection between software and hardware teaching and insufficient practical skills among students,a project-driven“learning-practiceapplication”teaching model is proposed.By optimizing course content,innovating teaching methods,and introducing university-industry collaboration mechanisms,the reform aligns the curriculum more closely with engineering education standards and industry demands.The approach significantly enhances students’comprehensive skills,practical abilities,and employability.This study provides theoretical foundations and practical strategies for the teaching reform of courses in communication engineering.
基金Project supported by the Science Challenge Project(Grant No.TZ2018004)the Natural Science Basic Research Program of Shaanxi Province,China(Grant No.2021JQ-697)+2 种基金the National Natural Science Foundation of China(Grant Nos.11874097,91961204,and 12004303)XinLiaoYingCai Project of Liaoning Province,China(Grant No.XLYC1905014)Key Research and Development Project of Liaoning Province,China(Grant No.2020JH2/10500003)。
文摘The GaN-based heterostructures are widely used in optoelectronic devices,but the complex surface reconstructions and lattice mismatch greatly limit the applications.The stacking of two-dimensional transition metal dichalcogenide(TMD=MoS_(2),MoSSe and MoSe_(2))monolayers on reconstructed GaN surface not only effectively overcomes the larger mismatch,but also brings about novel electronic and optical properties.By adopting the reconstructed GaN(0001)surface with adatoms(N-ter GaN and Ga-ter GaN),the influences of complicated surface conditions on the electronic properties of heterostructures have been investigated.The passivated N-ter and Ga-ter GaN surfaces push the mid-gap states to the valence bands,giving rise to small bandgaps in heterostructures.The charge transfer between Ga-ter GaN surface and TMD monolayers occurs much easier than that across the TMD/N-ter GaN interfaces,which induces stronger interfacial interaction and larger valence band offset(VBO).The band alignment can be switched between type-I and type-II by assembling different TMD monolayers,that is,MoS_(2)/N-ter GaN and MoS_(2)/Ga-ter GaN are type-II,and the others are type-I.The absorption of visible light is enhanced in all considered TMD/reconstructed GaN heterostructures.Additionally,MoSe_(2)/Ga-ter GaN and MoSSe/N-ter GaN have larger conductor band offset(CBO)of 1.32 eV and 1.29 eV,respectively,extending the range from deep ultraviolet to infrared regime.Our results revel that the TMD/reconstructed GaN heterostructures may be used for high-performance broadband photoelectronic devices.
基金supported by the National Key Research&Development Program of China(No.2024YFB3910102).
文摘The rapid expansion of the low-altitude economy is driving strong demand for highly accurate and reliable positioning technologies to support diverse aerial operations.This review examines core positioning methodologies within the low-altitude intelligent network(LAIN)framework,beginning with an analysis of positioning requirements and performance metrics for low-altitude flight scenarios.It systematically assesses the principles,strengths,and limitations of mainstream positioning systems,including Global Navigation Satellite Systems(GNSS),terrestrial wireless positioning,and autonomous navigation,and it surveys prevalent integrated and cooperative positioning schemes.Our analysis demonstrates that standalone positioning technologies are inadequate in complex low-altitude settings,underscoring the pivotal role of multi-source fusion and unmanned aerial vehicle(UAV)swarm cooperative positioning as future trends.To address infrastructure gaps and high deployment costs in current LAIN systems,we propose a“space−air−ground”integrated and cooperative positioning architecture centered on GNSS and the 5th generation mobile communication technology(5G).The ground layer integrates 5G and GNSS for wide-area enhanced positioning.The aerial layer uses 5G aircraft-to-everything(A2X)and sidelink(SL)communications to build self-organizing networks for cooperative UAV localization.The space layer leverages low Earth orbit(LEO)satellites to overcome coverage limitations in communication and positioning.This hierarchical architecture reduces deployment costs through infrastructure reuse and enables deep integration of communication and navigation capabilities.By supporting collaborative enhancement across all three domains,the framework improves positioning robustness and delivers cost-effective,ubiquitous,and highly reliable positioning services.Finally,we outline promising research directions.This review aims to provide a systematic reference and a novel architectural perspective for the ongoing development of LAIN.
基金supported by Natural Science Basic Research Program of Shaanxi Province of China (No. 2023-JC-YB-574)National Natural Science Foundation of China (No. 62304178)。
文摘β-Ga_(2)O_(3) MOS inverter should play a crucial role in β-Ga_(2)O_(3) electronic circuits. Enhancement-mode(E-mode) MOSFET was fabricated based on β-Ga_(2)O_(3) film grown by atomic layer deposition technology, and the β-Ga_(2)O_(3) inverter was further monolithically integrated on this basis. The β-Ga_(2)O_(3) n MOSFET exhibits excellent electrical characteristics with an on/off current ratio reaching 10^(5). The logic inverter shows outstanding voltage inversion characteristics under low-frequency from 1 to 400 Hz operation. As the frequency continues to increase to 10 K, the reverse characteristic becomes worse due to parasitic capacitance induced by processes, and the difference between the highest and lowest values of VOUT has an exponential decay relationship with the frequency. This paper provides the practice for the development of β-Ga_(2)O_(3)-based circuits.
基金the financial support from the National Key R&D Program of China(2021YFF0500500)the National Natural Science Foundation of China(62474131,62274132,and 62204189)。
文摘Ambient-air,moisture-assisted annealing is widely used in fabricating perovskite solar cells(PSCs).However,the inherent sensitivity of perovskite intermediate-phase to moisture—due to fast and spontaneous intermolecular exchange reaction—requires strict control of ambient humidity and immediate thermal annealing treatment,raising manufacturing costs and causing fast nucleation of perovskite films.We report herein a self-buffered molecular migration strategy to slow down the intermolecular exchange reaction by introducing a n-butylammonium bromide shielding layer,which limits moisture diffusion into intermediate-phase film.This further endows the notably wide nucleation time and humidity windows for perovskite crystallization in ambient air.Consequently,the optimized 1.68 e V-bandgap n-i-p structured PSC reaches a record-high reverse-scan(RS)PCE of 22.09%.Furthermore,the versatility and applicability of as-proposed self-buffered molecular migration strategy are certified by employing various shielding materials and 1.53 eV-/1.77 eV-bandgap perovskite materials.The n-i-p structured PSCs based on 1.53 eV-and 1.77 eV-bandgap perovskite films achieve outstanding RS PCEs of 25.23%and 19.09%,respectively,both of which are beyond of the state-of-the-art ambient-air processed PSCs.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.62341305 and 22269002)the Natural Science Foundation of Guangxi Zhuang Autonomous Region,China(Grant No.2024GXNSFFA010007)+2 种基金the Science and Technology Project of Guangxi Zhuang Autonomous Region,China(Grant No.AD19110038)the Key Laboratory of AI and Information Processing,Education Department of Guangxi Zhuang Autonomous Region(Grant No.2024GXZDSY015)the Innovation Project of Guangxi University of Science and Technology Graduate Education(Grant No.GKYC202408)。
文摘A facile one-step hydrothermal method has been reported to synthesize theα-Fe_(2)O_(3)nanosheet arrays with the preferred orientation along the[104]direction on the ITO substrate.Theα-Fe_(2)O_(3)nanosheet arrays-based W/α-Fe_(2)O_(3)/ITO memristor has been achieved by depositing the circular W top electrodes on theα-Fe_(2)O_(3)nanosheet arrays.The as-prepared W/α-Fe_(2)O_(3)/ITO memristor shows a reliable nonvolatile bipolar resistive switching behavior with the high resistance ratio of about 103at the reading voltage of 0.1 V,good resistance retention over 10~3s,ultralow set voltage of-0.6 V and reset voltage of 0.7 V,and good durability.In addition,the tunneling conduction mechanism modified by the oxygen vacancies has been proposed and suggested to be responsible for the nonvolatile resistive switching behavior of the as-prepared W/α-Fe_(2)O_(3)/ITO memristor.This work demonstrates that the as-preparedα-Fe_(2)O_(3)nanosheet arrays-based W/α-Fe_(2)O_(3)/ITO memristor would be a promising candidate for further ultralow power nonvolatile memory applications.
基金financially supported by the National Science and Technology Major Project(No.2024ZD1404705)。
文摘An advanced AlCrSiN/AlCrN/CrN/Cr multilayer coating was developed via hybrid multiarc ion plating and high-power impulse magnetron sputtering.The multilayer design enhanced the substrate-coating compatibility,achieving a critical load of 87.8 N.Silicon doping induced nanocrystallization and amorphization,increasing the hardness to 26 GPa.At high temperatures,a nanoscale Cr-rich(Cr,Al)_(2)O_(3) layer was formed,effectively inhibiting oxygen diffusion.The coating underwent unique phase transformations,during which Cr_(2)N and amorphous Si3N4 were converted into dispersed SiCr_(3) nanoparticles,which stabilized Cr atoms and suppressed their outward diffusion.Ab initio molecular dynamics simulations revealed that Cr atoms exhibited higher chemical activity and oxygen-capture capability than Al atoms and Si atoms served as diffusion barriers by pinning onto the oxidized surface,considerably improving the oxidation resistance of the coating.
基金Project supported by the National Natural Science Foundation of China(Grant No.51302215)the Natural Science Basic Research Program of Shaanxi Province,China(Grant Nos.2018JQ6084 and 2019JQ-860).
文摘The prospect ofα-Ga2O3 in optical and electrical devices application is fascinating.In order to obtain better performance,Ge and F elements with similar electronegativity and atomic size are selected as dopants.Based on density functional theory(DFT),we systematically research the electronic structure and optical properties of dopedα-Ga2O3 by GGA+U calculation method.The results show that Ge atoms and F atoms are effective n-type dopants.For Ge-dopedα-Ga2O3,it is probably obtained under O-poor conditions.However,for F-dopedα-Ga2O3,it is probably obtained under O-rich conditions.The doping system of F element is more stable due to the lower formation energy.In this investigation,it is found that two kinds of doping can reduce theα-Ga2O3 band gap and improve the conductivity.What is more,it is observed that the absorption edge after doping has a blue shift and causes certain absorption effect on the visible region.Through the whole scale of comparison,Ge doping is more suitable for the application of transmittance materials,yet F doping is more appropriate for the application of deep ultraviolet devices.We expect that our research can provide guidance and reference for preparation ofα-Ga2O3 thin films and photoelectric devices.
基金Supported by "973" National Key Basic Research Program ( No. 2002CB311905).
文摘Based on the region model of lambda bipolar transistor ( LBT), a dividing region theory model of PLBT is set up,simulated and verified. Firstly, the principal operations of different kinds of photoelectronic lambda bipolar transistor ( PLBT) are characterized by a simple circuit model. Through mathematical analysis of the equivalent circuit, the typical characteristics curve is divided into positive resistance, peak, negative resistance and cutoff regions. Secondly, by analyzing and simulating this model, the ratio of MOSFET width to channel length, threshold voltage and common emitter gain are discovered as the main structure parameters that determine the characteristic curves of PLBT. And peak region width, peak current value, negative resistance value and valley voltage value of PLBT can be changed conveniently according to the actual demands by modifying these parameters. Finally comparisons of the characteristics of the fabricated devices and the simu- lation results are made, which show that the analytical results are in agreement with the observed devices characteristics.
基金We acknowledge support from the Guangdong Basic and Applied Basic Research Foundation(No.2021A1515110980)the National Natural Science Foundation of China(Nos.51991343,51991340,52188101,and 52102179)+3 种基金the National Science Fund for Distinguished Young Scholars(No.52125309)Guangdong Innovative and Entrepreneurial Research Team Program(No.2017ZT07C341)the Bureau of Industry and Information Technology of Shenzhen for the“2017 Graphene Manufacturing Innovation Center Project”(No.201901171523)the Shenzhen Basic Research Program(No.JCYJ20200109144616617)。
文摘Two-dimensional(2D)materials offer novel platforms to meet the increasing demands of next-generation miniaturized electronics.Among them,the recently emerged 2D Bi_(2)O_(2)Se with unique non-van der Waals interlayer interaction,high mobility,sizeable bandgap,and capability to fabricate homologous heterojunction,is of particular interest.In this Review,we introduce recent progress in preparation,transfer,mechanical and electrical properties,and electronic applications of 2D Bi_(2)O_(2)Se.First,we summarize methodologies to synthesize and massively produce 2D Bi_(2)O_(2)Se,as well as recent advances in transferring them from growth substrate to arbitrary substrates.Then,we review current understandings on the intrinsic mechanical properties of Bi_(2)O_(2)Se at 2D thickness limit,and its in-plane and out-of-plane electrical properties.Electronic devices including field-effect transistors,memristors,and sensors based on 2D Bi_(2)O_(2)Se for neuromorphic computing,memory,logic,and integrated circuits are discussed.Finally,challenges and prospects for the development of 2D Bi_(2)O_(2)Se are proposed.
基金This work was supported by the National natural science foundation of China(No:62071057)Beijing nature fund(No:3182028).The support is gratefully acknowledged.
文摘With the rapid development of communication and computer,the individual identification technology of communication equipment has been brought to many application scenarios.The identification of the same type of electronic equipment is of considerable significance,whether it is the identification of friend or foe in military applications,identity determination,radio spectrum management in civil applications,equipment fault diagnosis,and so on.Because of the limited-expression ability of the traditional electromagnetic signal representation methods in the face of complex signals,a new method of individual identification of the same equipment of communication equipment based on deep learning is proposed.The contents of this paper include the following aspects:(1)Considering the shortcomings of deep learning in processing small sample data,this paper provides a universal and robust feature template for signal data.This paper constructs a relatively complete signal template library from multiple perspectives,such as time domain and transform domain features,combined with high-order statistical analysis.Based on the inspiration of the image texture feature,characteristics of amplitude histogram of signal and the signal amplitude co-occurrence matrix(SACM)are proposed in this paper.These signal features can be used as a signal fingerprint template for individual identification.(2)Considering the limitation of the recognition rate of a single classifier,using the integrated classifier has achieved better generalization ability.The final average accuracy of 5 NRF24LE1 modules is up to 98%and solved the problem of individual identification of the same equipment of communication equipment under the condition of the small sample,low signal-to-noise ratio.
文摘The rapid evolution of Fifth-Generation(5G)networks and the strategic development of Sixth-Generation(6G)technologies have significantly advanced the implementation of air-ground integrated networks with seamless coverage.Unmanned Aerial Vehicles(UAVs),serving as high-mobility aerial platforms,are extensively utilized to enhance coverage in long-distance emergency communication scenarios.The resource-constrained communication environments in emergencies by classifying UAVs into swarm UAVs and relay UAVs as aerial communication nodes is inversitgated.A horizontal deployment strategy for swarm UAVs is formulated through K-means clustering algorithm optimization,while a vertical deployment scheme is established using convex optimization methods.The minimum-path trajectory planning for relay UAVs is optimized via the Particle Swarm Optimization(PSO)algorithm,enhancing communication reliability between UAV swarms and terrestrial base stations.A three-dimensional heterogeneous network architecture is realized by modeling spatial multi-hop relay links.Experimental results demonstrate that the proposed joint UAV relay optimization framework outperforms conventional algorithms in both coverage performance and relay capability during video stream transmission,achieving significant improvements in coverage enhancement and relay efficiency.This work provides technical foundations for constructing high-reliability air-ground cooperative systems in emergency communications.
文摘Fiber optic temperature sensors stand out in a variety of applications due to their small size,chemical resistance,and resistance to electromagnetic interference.The traditional optical fiber temperature sensor directly places the sensing structure in the temperature to be measured,and uses the thermo-optical effect and thermal expansion effect of the SiO_(2)material that constitutes the sensing structure to achieve measurement,while the thermo-optical coefficient and thermal expansion coefficient of SiO_(2) are very small,which limits the high sensitivity response characteristics of the optical fiber temperature sensing structure.In order to solve the problem of low sensitivity of traditional optical fiber temperature sensors,a Mach-Zehnder interferometric temperature sensor with a liquid-encapsulated tapered microfiber is developed.The sensor converts the temperature change into a change in the refractive index of the liquid material and thus realizes the measurement of temperature.In the range of 25~50℃,as the temperature increases,the wavelength of the transmission spectrum shifts towards shorter wavelengths.Experimental results show that the sensitivity of the liquid encapsulated microfiber interferometric temperature sensor can reach-57.91 nk·nm^(-1).This sensor has great potential for applications in marine environmental monitoring,biomedical diagnosis,and aerospace.
基金funded by the National Natural Science Foundation of China(Nos.51772003 and 51701001)the Excellent Research and Innovation Team Project of Anhui Province(No.2023AH010077)the Key Research and Development Projects in Anhui Province(No.202004b11020021).
文摘Developing highly efficient and recyclable photocatalysts has been regarded as an attractive strategy to solve antibiotic contaminants.Herein,we designed and fabricated Cy-C_(3) N_(4)/TiO_(2) S-scheme heterojunction film with boosted charge transfer and a highly hydrophilic surface.The as-prepared heterojunction exhibited outstanding removal efficiency on tetracyclines and fluoroquinolone antibiotics(more than 80% within 90 min).The removal rate of 300-Cy-C_(3) N_(4)/TiO_(2) on norfloxacin(NOR)was 2.12,and 1.59 times higher than that of pristine TiO_(2),C_(3) N_(4)/TiO_(2),respectively.The excellent photocatalytic performance of 300-Cy-C_(3) N_(4)/TiO_(2) was attributed to the highly hydrophilic surface and effective transfer and separation of carriers.Moreover,the NOR degradation pathways were proposed based on the results of density functional theory(DFT),and liquid chromatography-mass spectrometry.The toxicity assessment indicated the toxicity of intermediates can be remarkably alleviated.The DFT calculation and selective photo-deposition experiment demonstrated that an internal electric field was formed at the heterojunction interface,and the charge carriers migrated between Cy-C_(3) N_(4) and TiO_(2) following an S-scheme transfer pathway.This research not only provides a promising method for tracking charge distribution on thin-film heterojunction photocatalysts but also helps us to design high-efficiency,and recyclable heterojunctions to solve antibiotic contaminants.
基金supported by the National Natural Science Foundation of China (No.6210031560)the Natural Science Foundation of Hebei Province (No.A2020202013)the Natural Science Foundation of Tianjin City (No.21JCQNJC00780)。
文摘The fiber optic sensing technology provides data support in structural health monitoring of the macro facilities,including design,construction,and maintenance of bridges,tunnels,ports and other infrastructures.In this paper,a distributed vibration sensing system is proved to be responsive to a single touch over a 1.8-m-long equivalent fiber segment,covering a vibration frequency from 5 Hz to 25 kHz.The sensing fiber was arranged as an S type layout on the bridge to recognize the standing state,windblown disturbance,and walking vibration.Moreover,the knocking and climbing events are recognized fiber laying spinning lines and hanging on the fences,respectively.The demonstration shows an accurate positioning and sensitive vibration monitoring applied on the automated three-dimensional(3D) printed bridge,which is applicable to all kinds of 3D printed facilities as intelligent sensory neuro-networks.
基金supported in part by the Beijing Natural Science Foundation No.JQ22011the National Science Foundation of China for Distinguished Young Scholars under Grant No.62325102+1 种基金the National Natural Science Foundation of China under Grant No.62031004the Fundamental Research Funds for the Central Universities No.2023YJS160.
文摘A wideband low-profile aperture-coupled antenna based on a novel dual-mode-composite scheme is presented.The mode-composite scheme where the TM10 cavity mode and the TE121 dielec-tric resonator(DR)mode are combined offers an ap-proach to obtain a wide bandwidth accompanied by stable unidirectional radiation and high efficiency.The use of a lengthened coupling aperture that supports the one-wavelength resonance in the band of interest is an effective feed method of simultaneously excit-ing the two composite modes without compromising the increased complexity of the antenna geometry.An impedance bandwidth of 49%for|S_(11)|of less than-10 dB,a maximum gain of 10.8 dBi,and stable radiation patterns with low cross-polarization are realized ex-perimentally by a fabricated prototype.Considering the simplicity of the geometry,the wide bandwidth that can cover n77,n78,and n79 bands for the fifth generation(5G)mobile communications and the sat-isfying radiation performance,the proposed antenna would be a promising candidate for advanced wireless applications.
基金Supported by the National Key R&D Program Broadband Communications and New Network Key Special Project(No.2019YFB1803600).
文摘Four-level pulse amplitude modulation(PAM4)signals,recognized for enhanced energy efficiency and spectral utilization compared with non-return-to-zero(NRZ)counterparts,have been adopted in multiple high-speed serializer/deserializer(SerDes)standards,but NRZ modulation remains predominant in industrial applications.This paper introduces a UMC 28 nm CMOS-based parallel configurable forward feedback equalization(FFE)dual-mode high-speed SerDes transmitter supporting 7-bit resolution with data rates of 56 Gb∙s^(-1)NRZ and 112 Gb∙s^(-1)PAM4,utilizing a hybrid architecture that integrates digital signal processing(DSP)with digital-to-analog conversion(DAC).The design processes parallel input signals and eight stored 8-bit tap coefficients through a configurable FFE multiplier module and parallel carry adder module,while achieving low-power serialization via low-speed 16∶4 multiplexers(MUXs)with two different 2∶1 MUXs and high-speed 4∶1 MUXs.A source series termination(SST)output network structure enhances lower power dissipation and higher output swing.Simulation results show that,under a 1.05 V supply voltage and a channel loss of 19.21 dB at 28 GHz,the output 56 Gb∙s^(-1)NRZ eye diagram has an eye height of 70.11 mV and an eye width of 12.16 ps(0.68 UI).The output 112 Gb∙s^(-1)PAM4 eye diagram has an eye height of 20.07 mV and an eye width of 7.49 ps(0.42 UI).The layout area of the dual-mode transmitter is 0.079 mm^(2),and the total circuit power consumption is 74.48 mW(energy efficiency is 1.33/0.67 pJ∙bit-1).
基金supported by the Key Research and Development Program of Shaanxi Province-International Science and Technology Cooperation Program Project (No.2020KW-001)the Contract for Xi'an Municipal Science and Technology Plan Project-Xi'an City Strong Foundation Innovation Plan (No.21XJZZ0074)the Key Project of Graduate Student Innovation Fund at Xi'an University of Posts and Telecommunications (No.CXJJZL2023013)。
文摘The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.
基金supported by the National Natural Science Foundation of China(Nos.62276210,82201148 and 62376215)the Key Research and Development Project of Shaanxi Province(No.2025CY-YBXM-044)+3 种基金the Natural Science Foundation of Zhejiang Province(No.LQ22H120002)the Medical Health Science and Technology Project of Zhejiang Province(Nos.2022RC069 and 2023KY1140)the Natural Science Foundation of Ningbo(No.2023J390)the Ningbo Top Medical and Health Research Program(No.2023030716).
文摘This paper proposes a novel method for the automatic diagnosis of keratitis using feature vector quantization and self-attention mechanisms(ADK_FVQSAM).First,high-level features are extracted using the DenseNet121 backbone network,followed by adaptive average pooling to scale the features to a fixed length.Subsequently,product quantization with residuals(PQR)is applied to convert continuous feature vectors into discrete features representations,preserving essential information insensitive to image quality variations.The quantized and original features are concatenated and fed into a self-attention mechanism to capture keratitis-related features.Finally,these enhanced features are classified through a fully connected layer.Experiments on clinical low-quality(LQ)images show that ADK_FVQSAM achieves accuracies of 87.7%,81.9%,and 89.3% for keratitis,other corneal abnormalities,and normal corneas,respectively.Compared to DenseNet121,Swin transformer,and InceptionResNet,ADK_FVQSAM improves average accuracy by 3.1%,11.3%,and 15.3%,respectively.These results demonstrate that ADK_FVQSAM significantly enhances the recognition performance of keratitis based on LQ slit-lamp images,offering a practical approach for clinical application.
基金supported by the Shandong Province Higher Education Young Innovative Talents Cultivation Programme Project:TJY2114Jinan City-School Integration Development Strategy Project:JNSX2023015the Natural Science Foundation of Shandong Province:ZR2021M F074.
文摘Traffic datasets exhibit complex spatiotemporal characteristics,including significant fluctuations in traffic volume and intricate periodical patterns,which pose substantial challenges for the accurate forecasting and effective management of traffic conditions.Traditional forecasting models often struggle to adequately capture these complexities,leading to suboptimal predictive performance.While neural networks excel at modeling intricate and nonlinear data structures,they are also highly susceptible to overfitting,resulting in inefficient use of computational resources and decreased model generalization.This paper introduces a novel heuristic feature extraction method that synergistically combines the strengths of non-neural network algorithms with neural networks to enhance the identification and representation of relevant features from traffic data.We begin by evaluating the significance of various temporal characteristics using three distinct assessment strategies grounded in non-neural methodologies.These evaluated features are then aggregated through a weighted fusion mechanism to create heuristic features,which are subsequently integrated into neural network models for more accurate and robust traffic prediction.Experimental results derived from four real-world datasets,collected from diverse urban environments,show that the proposed method significantly improves the accuracy of long-term traffic forecasting without compromising performance.Additionally,the approach helps streamline neural network architectures,leading to a considerable reduction in computational overhead.By addressing both prediction accuracy and computational efficiency,this study not only presents an innovative and effective method for traffic condition forecasting but also offers valuable insights that can inform the future development of data-driven traffic management systems and transportation strategies.
