Although eye problems can occur at any age, they are often common from the age of 40. Eye diseases with a prevalence associated with age and aging will continue to increase in the coming years. Most studies conducted ...Although eye problems can occur at any age, they are often common from the age of 40. Eye diseases with a prevalence associated with age and aging will continue to increase in the coming years. Most studies conducted on problems in middle-aged people have focused on visual disorders without taking into account all the ocular morbidities that may affect this segment of the population, hence the present study, the aim of which is to determine the proportions of different eye diseases in people aged 40 and over. Materials and Methods: This was a descriptive cross-sectional study carried out in the ophthalmology department covering the period from January 1 to December 31, 2020. Results: In total, we collected 828 patients aged 40 and over out of 1811 patients who received ophthalmological consultation during the study period, representing 45.72%. The most represented age group was 40 - 50 years, with an average age of 58.84 years and a maximum of 93 years. There were slightly more women (62.3%) than men (37.7%). The main reasons for consultation were decreased visual acuity (26.4%) and pruritus (19.9%). The main eye diseases diagnosed were cataracts (23%), allergic conjunctivitis (21.1%), and bacterial conjunctivitis (14.2%). Discussions: The predominance of cataracts in the diagnosed diseases confirms the literature data, according to which the main eye morbidities in middle-aged and elderly people are cataracts, glaucoma, and age-related macular degeneration. Conclusion: It is crucial to have a mastery of these epidemiological data of eye diseases in order to adapt the technical platforms of eye care structures to the needs of different segments of the population.展开更多
To explore the electrostatic discharge behavior of charged powders in industrial silos,discharge experiments are conducted based on a full-size industrial silo discharge platform.Electrostatic discharge mode,frequency...To explore the electrostatic discharge behavior of charged powders in industrial silos,discharge experiments are conducted based on a full-size industrial silo discharge platform.Electrostatic discharge mode,frequency,and energy are investigated for powders of different polarities.Although the powders have low charge-to-mass ratios(+0.087μC/kg for the positively charged powders and−0.26μC/kg for the negatively charged ones),electrostatic discharges occur approximately every 10 s,with the maximum discharge energy being 800 mJ.Powder polarity considerably influences discharge energy.The positive powders exhibit higher discharge energy than the negative ones,although discharge frequency remains similar for both.Effects of powder charge,humidity,and mass flow on discharge frequency and discharge energy are quantitatively analyzed,providing important insights for the improvement of safety in industrial powder handling.展开更多
Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency informatio...Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency information,which makes them ineffective in balancing users’long-and short-term preferences.At the same time,manymethods overlook the potential of frequency domainmethods,ignoring their efficiency in processing frequency information.To overcome this limitation,we shift the focus to the combination of time and frequency domains and propose a novel Hybrid Time-Frequency Dual-Branch Transformer for Sequential Recommendation,namely HyTiFRec.Specifically,we design two hybrid filter modules:the learnable hybrid filter(LHF)and the window hybrid filter(WHF).We combine these with the Efficient Attention(EA)module to form the dual-branch structure to replace the self-attention components in Transformers.The EAmodule is used to extract sequential and global information.The LHF andWHF modules balance the proportion of different frequency bands,with LHF globally modulating the spectrum in the frequency domain and WHF retaining frequency components within specific local frequency bands.Furthermore,we use a time domain residual information addition operation in the hybrid filter module,which reduces information loss and further facilitates the hybrid of time-frequency methods.Extensive experiments on five widely-used real-world datasets show that our proposed method surpasses state-of-the-art methods.展开更多
Semantic segmentation of novel object categories with limited labeled data remains a challenging problem in computer vision.Few-shot segmentation methods aim to address this problem by recognizing objects from specifi...Semantic segmentation of novel object categories with limited labeled data remains a challenging problem in computer vision.Few-shot segmentation methods aim to address this problem by recognizing objects from specific target classes with a few provided examples.Previous approaches for few-shot semantic segmentation typically represent target classes using class prototypes.These prototypes are matched with the features of the query set to get segmentation results.However,class prototypes are usually obtained by applying global average pooling on masked support images.Global pooling discards much structural information,which may reduce the accuracy of model predictions.To address this issue,we propose a Category-Guided Frequency Modulation(CGFM)method.CGFM is designed to learn category-specific information in the frequency space and leverage it to provide a twostage guidance for the segmentation process.First,to self-adaptively activate class-relevant frequency bands while suppressing irrelevant ones,we leverage the Dual-Perception Gaussian Band Pre-activation(DPGBP)module to generate Gaussian filters using class embedding vectors.Second,to further enhance category-relevant frequency components in activated bands,we design a Support-Guided Category Response Enhancement(SGCRE)module to effectively introduce support frequency components into the modulation of query frequency features.Experiments on the PASCAL-5^(i) and COCO-20^(i) datasets demonstrate the promising performance of our model.展开更多
Acoustic frequency combs(AFCs)contain equidistant coherent signals with unconventional possibilities on metrology.Previously,implementation of AFCs on mechanical microresonators with large air damping loss is difficul...Acoustic frequency combs(AFCs)contain equidistant coherent signals with unconventional possibilities on metrology.Previously,implementation of AFCs on mechanical microresonators with large air damping loss is difficult,which restricted their atmospheric applications.In this work,we explore the potentials of a composite diamond/silicon microcantilever for parametric generation of AFCs in ambient air.We discover that the diamond layer provides a viable route to reduce the linewidth of the primary flexural mode,yielding a 7.1-times increase of the quality factor.We develop a parametric driving scheme that enables generation of AFCs through injection locking and sequential nonlinear dynamic transitions involving subharmonic synchronization(Arnold tongue),and chaotic dynamics.Ultimately,we realize AFCs with a frequency range extending 800 kHz in the air.This work advances the understanding of AFCs and provides a viable route towards their applications in ambient air for high precision metrology.展开更多
Urban environments have challenging characteristics for bird acoustic communication.High levels of anthropogenic noise,as well as vegetation structure(e.g.,in urban parks),can potentially affect the song frequency cha...Urban environments have challenging characteristics for bird acoustic communication.High levels of anthropogenic noise,as well as vegetation structure(e.g.,in urban parks),can potentially affect the song frequency characteristics of several bird species.An additional factor such as the abundance of conspecific and heterospecific vocalizing birds may play an important role in determining the structure of bird songs.In this study,we analyzed whether noise levels,vegetation percentage,and abundance of conspecifics and heterospecifics influence the song characteristics of three syntopic songbird species:House Finch(Haemorhous mexicanus),Rufouscollared Sparrow(Zonotrichia capensis),and House Sparrow(Passer domesticus)living in urban sites.We recorded songs of these species and measured the peak frequency and entropy of their songs at 14 sites in the city of San Cristobal de Las Casas,Chiapas,Mexico.We found that the song peak frequency of House Finch and House Sparrow's songs was negatively related to the vegetation.The peak frequency of neither of the three species correlated with the average noise level.However,the abundances of conspecific and heterospecific were related to the peak frequency of the three species'songs.The entropy of the House Finch and House Sparrow songs was positively and negatively related,respectively,to noise levels.House Sparrow song entropy was negatively related to the percentage of vegetation.Song entropy of House Finches was negatively associated to conspecific and House Sparrow abundance.Song entropy of Rufous-collared Sparrows was positively related to conspecific abundance.In conclusion,the song peak frequency and song entropy of the three songbird species were differentially related to urban noise,vegetation,and conspecific and heterospecific abundance,suggesting these factors influence bird song characteristics.展开更多
Low-frequency(LF)electromagnetic waves have high penetration and low attenuation characteristics in media,making them essential for cross-media communications.In LF communication systems,the loop antenna commonly func...Low-frequency(LF)electromagnetic waves have high penetration and low attenuation characteristics in media,making them essential for cross-media communications.In LF communication systems,the loop antenna commonly functions as a receiver for detecting weak signals.However,traditional LF loop antennas typically require large structures to achieve high radiation efficiency,which poses challenges for portability and long-distance transmission.Here,a magnetic resonant coupling metamaterial(MRCM)antenna with high radiation capacity,frequency tunability,direction adjustability,and compact form is demonstrated.To elucidate its radiation mechanism and frequency modulation capabilities,the equivalent circuit model and electromagnetic simulations are carried out.Compared with conventional loop antennas,the MRCM antennas can realize the radiation magnetic flux density seven times and extend the effective magnetic transmission distance by three times.Besides,the MRCM antennas allow for adjustable radiation direction and operating frequency,enhancing its versatility in different application scenarios.This metamaterial antenna design allows a pocket-sized antenna to achieve an effective communication range of 180 m,presenting a promising solution for improving communication capabilities in changing environments such as underwater and underground settings.展开更多
Low-frequency signals have been proven valuable in the fields of target detection and geological exploration.Nevertheless,the practical implementation of these signals is hindered by large antenna diameters,limiting t...Low-frequency signals have been proven valuable in the fields of target detection and geological exploration.Nevertheless,the practical implementation of these signals is hindered by large antenna diameters,limiting their potential applications.Therefore,it is imperative to study the creation of lowfrequency signals using antennas with suitable dimensions.In contrast to conventional mechanical antenna techniques,our study generates low-frequency signals in the spatial domain utilizing the principle of the Doppler effect.We also defines the antenna array architecture,the timing sequency,and the radiating element signal waveform,and provides experimental prototypes including 8/64 antennas based on earlier research.In the conducted experiments,121 MHz,40 MHz,and 10 kHz composite signals are generated by 156 MHz radiating element signals.The composite signal spectrum matches the simulations,proving our low-frequency signal generating method works.This holds significant implications for research on generating low-frequency signals with small-sized antennas.展开更多
To tackle the issue of notch frequency and center frequency drift of the L(0,1)mode guided wave in ultra⁃sonic guided wave⁃based stress monitoring of prestressed steel strands,a method using higher⁃order mode plateau ...To tackle the issue of notch frequency and center frequency drift of the L(0,1)mode guided wave in ultra⁃sonic guided wave⁃based stress monitoring of prestressed steel strands,a method using higher⁃order mode plateau fre⁃quencies is adopted.First,the correlation between group velocity peaks and phase velocities at these plateau frequen⁃cies is analyzed.This analysis establishes a quantitative rela⁃tionship between phase velocity and stress in the steel strand,providing a theoretical foundation for stress monitor⁃ing.Then the two⁃dimensional Fourier transform is em⁃ployed to separate wave modes.Dynamic programming techniques are applied in the frequency⁃velocity domain to extract higher⁃order modes.By identifying the group veloc⁃ity peaks of these separated higher⁃order modes,the plateau frequencies of guided waves are determined,enabling indi⁃rect measurement of stress in the steel strand.To validate this method,finite element simulations are conducted under three scenarios.Results show that the higher⁃order modes of transient signals from three different positions can be ac⁃curately extracted,leading to successful cable stress moni⁃toring.This approach effectively circumvents the issue of guided wave frequency drift and improves stress monitoring accuracy.Consequently,it significantly improves the appli⁃cation of ultrasonic guided wave technology in structural health monitoring.展开更多
Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vi...Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vision, attracting the attention of many researchers. However, most HSI SR methods focus on the tradeoff between spatial resolution and spectral information, and cannot guarantee the efficient extraction of image information. In this paper, a multidimensional features network(MFNet) for HSI SR is proposed, which simultaneously learns and fuses the spatial,spectral, and frequency multidimensional features of HSI. Spatial features contain rich local details,spectral features contain the information and correlation between spectral bands, and frequency feature can reflect the global information of the image and can be used to obtain the global context of HSI. The fusion of the three features can better guide image super-resolution, to obtain higher-quality high-resolution hyperspectral images. In MFNet, we use the frequency feature extraction module(FFEM) to extract the frequency feature. On this basis, a multidimensional features extraction module(MFEM) is designed to learn and fuse multidimensional features. In addition, experimental results on two public datasets demonstrate that MFNet achieves state-of-the-art performance.展开更多
Albeit with the notable benefits associated with Dirichlet crash frequency models and spatial ones,there is little research dedicated to exploring their combined advantages.Such ensemble approach could be a viable alt...Albeit with the notable benefits associated with Dirichlet crash frequency models and spatial ones,there is little research dedicated to exploring their combined advantages.Such ensemble approach could be a viable alternative to existing models as it accounts for the unobserved heterogeneity by relaxing the constraints of specific distribution placed on the intercept while addressing the spatial correlations among roadway entities.To fill this gap,the authors aimed to develop Dirichlet semi-parametric models over the overdispersed generalized linear model framework while also incorporating spatially structured random effects using a distance-based weight matrix.Five models were developed which include four semi-parametric with flexible intercept and one parametric base model for comparison purposes.The four semi-parametric models entailed two models with a popular specification of stick-breaking Dirichlet process(DP)and two models with an alternative approach of Dirichlet distribution(DD),which are first applied in the field of traffic safety.All four models were estimated for mixture of points(discrete)and mixture of normals(continuous).The posterior density plots for the precision parameter justified the employment of the flexible Dirichlet approach to fit the crash data and supported the assumed prior for the precision parameter.All four Dirichlet models demonstrated the presence of distinct subpopulations suggesting that the intercepts of the models were not generated from a common distribution.The DP model based on mixture of normals illustrated better performance indicating its potential superiority to fit both insample and out-of-sample crash data.This finding indicated that the approach of continuous densities,unlike discrete points,may lend more flexibility to fit the data.展开更多
A wide passband frequency selective surface(FSS)is proposed using a five-layer stacked structure.The proposed structure applies four layers of dielectric plates and five layers of metal patches to provide a passband a...A wide passband frequency selective surface(FSS)is proposed using a five-layer stacked structure.The proposed structure applies four layers of dielectric plates and five layers of metal patches to provide a passband and exhibits more stable frequency responses and lower insertion loss under wide-angle oblique incidence compared with the typical three-layer metal-dielectric structure.According to the simulation results,the proposed FSS can achieve a passband range of 1.7-2.7 GHz with an insertion loss of less than 0.5 d B and a relative bandwidth of 44.1%,and it can preserve stable transmission characteristics with the incident angle ranging from 0°to 45°.展开更多
Hypersonic Glide Vehicles(HGVs)are advanced aircraft that can achieve extremely high speeds(generally over 5 Mach)and maneuverability within the Earth's atmosphere.HGV trajectory prediction is crucial for effectiv...Hypersonic Glide Vehicles(HGVs)are advanced aircraft that can achieve extremely high speeds(generally over 5 Mach)and maneuverability within the Earth's atmosphere.HGV trajectory prediction is crucial for effective defense planning and interception strategies.In recent years,HGV trajectory prediction methods based on deep learning have the great potential to significantly enhance prediction accuracy and efficiency.However,it's still challenging to strike a balance between improving prediction performance and reducing computation costs of the deep learning trajectory prediction models.To solve this problem,we propose a new deep learning framework(FECA-LSMN)for efficient HGV trajectory prediction.The model first uses a Frequency Enhanced Channel Attention(FECA)module to facilitate the fusion of different HGV trajectory features,and then subsequently employs a Light Sampling-oriented Multi-Layer Perceptron Network(LSMN)based on simple MLP-based structures to extract long/shortterm HGV trajectory features for accurate trajectory prediction.Also,we employ a new data normalization method called reversible instance normalization(RevIN)to enhance the prediction accuracy and training stability of the network.Compared to other popular trajectory prediction models based on LSTM,GRU and Transformer,our FECA-LSMN model achieves leading or comparable performance in terms of RMSE,MAE and MAPE metrics while demonstrating notably faster computation time.The ablation experiments show that the incorporation of the FECA module significantly improves the prediction performance of the network.The RevIN data normalization technique outperforms traditional min-max normalization as well.展开更多
This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the traini...This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the training dataset,and one solution is applied to improve the distribution of the training data by augmenting minority class samples using a deep convolutional generative adversarial network.Experi.mental results demonstrate that retraining the deep learning model with the newly generated dataset leads to a new fast radio burst classifier,which effectively reduces false positives caused by periodic wide-band impulsive radio frequency interference,thereby enhancing the performance of the search pipeline.展开更多
Large-scale quantum networks require dynamic and resource-efficient solutions to reduce system complexity with maintained security and performance to support growing number of users over large distances.Current encodi...Large-scale quantum networks require dynamic and resource-efficient solutions to reduce system complexity with maintained security and performance to support growing number of users over large distances.Current encoding schemes including time-bin,polarization,and orbital angular momentum,suffer from the lack of reconfigurability and thus scalability issues.Here,we demonstrate the first-time implementation of frequency-bin-encoded entanglement-based quantum key distribution and a reconfigurable distribution of entanglement using frequency-bin encoding.Specifically,we demonstrate a novel scalable frequency-bin basis analyzer module that allows for a passive random basis selection as a crucial step in quantum protocols,and importantly equips each user with a single detector rather than four detectors.This minimizes massively the resource overhead,reduces the dark count contribution,vulnerability to detector side-channel attacks,and the detector imbalance,hence providing an enhanced security.Our approach offers an adaptive frequency-multiplexing capability to increase the number of channels without hardware overhead,enabling increased secret key rate and reconfigurable multi-user operations.In perspective,our approach enables dynamic resource-minimized quantum key distribution among multiple users across diverse network topologies,and facilitates scalability to large-scale quantum networks.展开更多
This paper presents a novel technique for low-power generation of frequency combs(FC)over a wide frequency range.It leverages modal interactions between electrical and mechanical resonators in electrostatic NEMS opera...This paper presents a novel technique for low-power generation of frequency combs(FC)over a wide frequency range.It leverages modal interactions between electrical and mechanical resonators in electrostatic NEMS operating in air to provide a simple architecture for FC generators.A biased voltage signal drives the electrical resonator at resonance which is set to match an integer submultiple of twice the mechanical resonator’s resonance.Experimental results demonstrate that the NEMS displacement exhibit more than 150 equidistant peaks in the case of a 2:1 modal interaction and more than 60 equidistant peaks in the case of a 1:1 modal interaction.In both cases,the Free Spectral Range(FSR)was equal to the mechanical resonance frequency.Comparison between the FCs generated by the 2:1 and 1:1 modal interactions demonstrate the superiority of the former in terms of bandwidth and stability.The superior phase coherence of the FC generated via the 2:1 modal interaction was demonstrated via time-domain analysis.Our technique has the flexibility to generate multiple frequency combs and to fine-tune their FSR depending on the number of mechanical modes accessible to and the order of the activated modal interaction.It can be integrated into portable devices and is well aligned with modern miniaturization technology.展开更多
伴随着光纤技术的快速发展,光纤网络已部署于航空航天、舰船、数据中心和工业物联网中。传统的光时域反射仪(optical time domain reflectometer,OTDR)因原理限制,难以实现高分辨率测试,在上述复杂场景中应用受限。基于瑞利散射的光频...伴随着光纤技术的快速发展,光纤网络已部署于航空航天、舰船、数据中心和工业物联网中。传统的光时域反射仪(optical time domain reflectometer,OTDR)因原理限制,难以实现高分辨率测试,在上述复杂场景中应用受限。基于瑞利散射的光频域反射(optical frequency domain reflection,OFDR)技术可实现极高的空间分辨率、高传感灵敏度和快速的测试速率,该系列产品适用于光器件、光模块、短距离光网络的测试和故障排除,可实现从器件到光学链路全范围的插损、回损和长度测量。文中基于光频域反射法原理设计实现了一套光纤链路检测系统,针对偏振衰落效应和激光器非线性扫频等难题进行了研究,在112 m的测试链路上实现了20μm空间分辨率。展开更多
The FDR automatic soil moisture sensor must determine reference frequency in the air and water. Experimental studies show that the water reference frequency is influenced by water temperature. The variation of the ref...The FDR automatic soil moisture sensor must determine reference frequency in the air and water. Experimental studies show that the water reference frequency is influenced by water temperature. The variation of the reference frequency of the sensor is measured with the change of the water temperature,then analysis the influence of the volume water content measurement of the sensor,analysis found that the error is not more than 3% for the measurement of the volumetric water content of the temperature.展开更多
Orthogonal Time Frequency and Space(OTFS) modulation is expected to provide high-speed and ultra-reliable communications for emerging mobile applications, including low-orbit satellite communications. Using the Dopple...Orthogonal Time Frequency and Space(OTFS) modulation is expected to provide high-speed and ultra-reliable communications for emerging mobile applications, including low-orbit satellite communications. Using the Doppler frequency for positioning is a promising research direction on communication and navigation integration. To tackle the high Doppler frequency and low signal-to-noise ratio(SNR) in satellite communication, this paper proposes a Red and Blue Frequency Shift Discriminator(RBFSD) based on the pseudo-noise(PN) sequence.The paper derives that the cross-correlation function on the Doppler domain exhibits the characteristic of a Sinc function. Therefore, it applies modulation onto the Delay-Doppler domain using PN sequence and adjusts Doppler frequency estimation by red-shifting or blue-shifting. Simulation results show that the performance of Doppler frequency estimation is close to the Cramér-Rao Lower Bound when the SNR is greater than -15dB. The proposed algorithm is about 1/D times less complex than the existing PN pilot sequence algorithm, where D is the resolution of the fractional Doppler.展开更多
文摘Although eye problems can occur at any age, they are often common from the age of 40. Eye diseases with a prevalence associated with age and aging will continue to increase in the coming years. Most studies conducted on problems in middle-aged people have focused on visual disorders without taking into account all the ocular morbidities that may affect this segment of the population, hence the present study, the aim of which is to determine the proportions of different eye diseases in people aged 40 and over. Materials and Methods: This was a descriptive cross-sectional study carried out in the ophthalmology department covering the period from January 1 to December 31, 2020. Results: In total, we collected 828 patients aged 40 and over out of 1811 patients who received ophthalmological consultation during the study period, representing 45.72%. The most represented age group was 40 - 50 years, with an average age of 58.84 years and a maximum of 93 years. There were slightly more women (62.3%) than men (37.7%). The main reasons for consultation were decreased visual acuity (26.4%) and pruritus (19.9%). The main eye diseases diagnosed were cataracts (23%), allergic conjunctivitis (21.1%), and bacterial conjunctivitis (14.2%). Discussions: The predominance of cataracts in the diagnosed diseases confirms the literature data, according to which the main eye morbidities in middle-aged and elderly people are cataracts, glaucoma, and age-related macular degeneration. Conclusion: It is crucial to have a mastery of these epidemiological data of eye diseases in order to adapt the technical platforms of eye care structures to the needs of different segments of the population.
基金The National Natural Science Foundation of China(No.51976039)。
文摘To explore the electrostatic discharge behavior of charged powders in industrial silos,discharge experiments are conducted based on a full-size industrial silo discharge platform.Electrostatic discharge mode,frequency,and energy are investigated for powders of different polarities.Although the powders have low charge-to-mass ratios(+0.087μC/kg for the positively charged powders and−0.26μC/kg for the negatively charged ones),electrostatic discharges occur approximately every 10 s,with the maximum discharge energy being 800 mJ.Powder polarity considerably influences discharge energy.The positive powders exhibit higher discharge energy than the negative ones,although discharge frequency remains similar for both.Effects of powder charge,humidity,and mass flow on discharge frequency and discharge energy are quantitatively analyzed,providing important insights for the improvement of safety in industrial powder handling.
基金supported by a grant from the Natural Science Foundation of Zhejiang Province under Grant LY21F010016.
文摘Recently,many Sequential Recommendation methods adopt self-attention mechanisms to model user preferences.However,these methods tend to focus more on low-frequency information while neglecting highfrequency information,which makes them ineffective in balancing users’long-and short-term preferences.At the same time,manymethods overlook the potential of frequency domainmethods,ignoring their efficiency in processing frequency information.To overcome this limitation,we shift the focus to the combination of time and frequency domains and propose a novel Hybrid Time-Frequency Dual-Branch Transformer for Sequential Recommendation,namely HyTiFRec.Specifically,we design two hybrid filter modules:the learnable hybrid filter(LHF)and the window hybrid filter(WHF).We combine these with the Efficient Attention(EA)module to form the dual-branch structure to replace the self-attention components in Transformers.The EAmodule is used to extract sequential and global information.The LHF andWHF modules balance the proportion of different frequency bands,with LHF globally modulating the spectrum in the frequency domain and WHF retaining frequency components within specific local frequency bands.Furthermore,we use a time domain residual information addition operation in the hybrid filter module,which reduces information loss and further facilitates the hybrid of time-frequency methods.Extensive experiments on five widely-used real-world datasets show that our proposed method surpasses state-of-the-art methods.
文摘Semantic segmentation of novel object categories with limited labeled data remains a challenging problem in computer vision.Few-shot segmentation methods aim to address this problem by recognizing objects from specific target classes with a few provided examples.Previous approaches for few-shot semantic segmentation typically represent target classes using class prototypes.These prototypes are matched with the features of the query set to get segmentation results.However,class prototypes are usually obtained by applying global average pooling on masked support images.Global pooling discards much structural information,which may reduce the accuracy of model predictions.To address this issue,we propose a Category-Guided Frequency Modulation(CGFM)method.CGFM is designed to learn category-specific information in the frequency space and leverage it to provide a twostage guidance for the segmentation process.First,to self-adaptively activate class-relevant frequency bands while suppressing irrelevant ones,we leverage the Dual-Perception Gaussian Band Pre-activation(DPGBP)module to generate Gaussian filters using class embedding vectors.Second,to further enhance category-relevant frequency components in activated bands,we design a Support-Guided Category Response Enhancement(SGCRE)module to effectively introduce support frequency components into the modulation of query frequency features.Experiments on the PASCAL-5^(i) and COCO-20^(i) datasets demonstrate the promising performance of our model.
基金the National Science Foundation of China(NSFC)(Grant No.52172296,51472143)the National Key R&D Program of China(NKRD)(2017YFB0405403)the Initiative in Quantum Science of Shandong Provincial Natural Science Foundation(ZR2020LLZ005)for financial support.
文摘Acoustic frequency combs(AFCs)contain equidistant coherent signals with unconventional possibilities on metrology.Previously,implementation of AFCs on mechanical microresonators with large air damping loss is difficult,which restricted their atmospheric applications.In this work,we explore the potentials of a composite diamond/silicon microcantilever for parametric generation of AFCs in ambient air.We discover that the diamond layer provides a viable route to reduce the linewidth of the primary flexural mode,yielding a 7.1-times increase of the quality factor.We develop a parametric driving scheme that enables generation of AFCs through injection locking and sequential nonlinear dynamic transitions involving subharmonic synchronization(Arnold tongue),and chaotic dynamics.Ultimately,we realize AFCs with a frequency range extending 800 kHz in the air.This work advances the understanding of AFCs and provides a viable route towards their applications in ambient air for high precision metrology.
基金the Consejo Nacional de Humanidades,Ciencias y Tecnologías (CONAHCYT)of Mexico for providing funding for graduate studies of X.D.L. (No.001283)El Colegio de la Frontera Sur for PATM graduate fellowship for fieldwork。
文摘Urban environments have challenging characteristics for bird acoustic communication.High levels of anthropogenic noise,as well as vegetation structure(e.g.,in urban parks),can potentially affect the song frequency characteristics of several bird species.An additional factor such as the abundance of conspecific and heterospecific vocalizing birds may play an important role in determining the structure of bird songs.In this study,we analyzed whether noise levels,vegetation percentage,and abundance of conspecifics and heterospecifics influence the song characteristics of three syntopic songbird species:House Finch(Haemorhous mexicanus),Rufouscollared Sparrow(Zonotrichia capensis),and House Sparrow(Passer domesticus)living in urban sites.We recorded songs of these species and measured the peak frequency and entropy of their songs at 14 sites in the city of San Cristobal de Las Casas,Chiapas,Mexico.We found that the song peak frequency of House Finch and House Sparrow's songs was negatively related to the vegetation.The peak frequency of neither of the three species correlated with the average noise level.However,the abundances of conspecific and heterospecific were related to the peak frequency of the three species'songs.The entropy of the House Finch and House Sparrow songs was positively and negatively related,respectively,to noise levels.House Sparrow song entropy was negatively related to the percentage of vegetation.Song entropy of House Finches was negatively associated to conspecific and House Sparrow abundance.Song entropy of Rufous-collared Sparrows was positively related to conspecific abundance.In conclusion,the song peak frequency and song entropy of the three songbird species were differentially related to urban noise,vegetation,and conspecific and heterospecific abundance,suggesting these factors influence bird song characteristics.
基金supported by the National Natural Science Foundation of China(Grant Nos.U2241243,52102061,52372101)Beijing Natural Science Foundation(Grant No.JQ22010)+2 种基金the Fundamental Research Funds for the Central Universities(Grant No.2023ZCJH03)the Teaching Reform Projects at BUPT(Grant No.2024Y010)the Fund of State Key Laboratory of IPOC(BUPT)(Grant No.IPOC2024ZT13)。
文摘Low-frequency(LF)electromagnetic waves have high penetration and low attenuation characteristics in media,making them essential for cross-media communications.In LF communication systems,the loop antenna commonly functions as a receiver for detecting weak signals.However,traditional LF loop antennas typically require large structures to achieve high radiation efficiency,which poses challenges for portability and long-distance transmission.Here,a magnetic resonant coupling metamaterial(MRCM)antenna with high radiation capacity,frequency tunability,direction adjustability,and compact form is demonstrated.To elucidate its radiation mechanism and frequency modulation capabilities,the equivalent circuit model and electromagnetic simulations are carried out.Compared with conventional loop antennas,the MRCM antennas can realize the radiation magnetic flux density seven times and extend the effective magnetic transmission distance by three times.Besides,the MRCM antennas allow for adjustable radiation direction and operating frequency,enhancing its versatility in different application scenarios.This metamaterial antenna design allows a pocket-sized antenna to achieve an effective communication range of 180 m,presenting a promising solution for improving communication capabilities in changing environments such as underwater and underground settings.
基金Science and Technology Project of Aerospace Information Research Institute,Chinese Academy of Sciences(Y910340Z2F)Science and Technology Project of BBEF(E3E2010201)。
文摘Low-frequency signals have been proven valuable in the fields of target detection and geological exploration.Nevertheless,the practical implementation of these signals is hindered by large antenna diameters,limiting their potential applications.Therefore,it is imperative to study the creation of lowfrequency signals using antennas with suitable dimensions.In contrast to conventional mechanical antenna techniques,our study generates low-frequency signals in the spatial domain utilizing the principle of the Doppler effect.We also defines the antenna array architecture,the timing sequency,and the radiating element signal waveform,and provides experimental prototypes including 8/64 antennas based on earlier research.In the conducted experiments,121 MHz,40 MHz,and 10 kHz composite signals are generated by 156 MHz radiating element signals.The composite signal spectrum matches the simulations,proving our low-frequency signal generating method works.This holds significant implications for research on generating low-frequency signals with small-sized antennas.
基金The National Natural Science Foundation of China(No.52278303).
文摘To tackle the issue of notch frequency and center frequency drift of the L(0,1)mode guided wave in ultra⁃sonic guided wave⁃based stress monitoring of prestressed steel strands,a method using higher⁃order mode plateau fre⁃quencies is adopted.First,the correlation between group velocity peaks and phase velocities at these plateau frequen⁃cies is analyzed.This analysis establishes a quantitative rela⁃tionship between phase velocity and stress in the steel strand,providing a theoretical foundation for stress monitor⁃ing.Then the two⁃dimensional Fourier transform is em⁃ployed to separate wave modes.Dynamic programming techniques are applied in the frequency⁃velocity domain to extract higher⁃order modes.By identifying the group veloc⁃ity peaks of these separated higher⁃order modes,the plateau frequencies of guided waves are determined,enabling indi⁃rect measurement of stress in the steel strand.To validate this method,finite element simulations are conducted under three scenarios.Results show that the higher⁃order modes of transient signals from three different positions can be ac⁃curately extracted,leading to successful cable stress moni⁃toring.This approach effectively circumvents the issue of guided wave frequency drift and improves stress monitoring accuracy.Consequently,it significantly improves the appli⁃cation of ultrasonic guided wave technology in structural health monitoring.
基金supported by the Fundamental Research Funds for the Provincial Universities of Zhejiang (No.GK249909299001-036)National Key Research and Development Program of China (No. 2023YFB4502803)Zhejiang Provincial Natural Science Foundation of China (No.LDT23F01014F01)。
文摘Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vision, attracting the attention of many researchers. However, most HSI SR methods focus on the tradeoff between spatial resolution and spectral information, and cannot guarantee the efficient extraction of image information. In this paper, a multidimensional features network(MFNet) for HSI SR is proposed, which simultaneously learns and fuses the spatial,spectral, and frequency multidimensional features of HSI. Spatial features contain rich local details,spectral features contain the information and correlation between spectral bands, and frequency feature can reflect the global information of the image and can be used to obtain the global context of HSI. The fusion of the three features can better guide image super-resolution, to obtain higher-quality high-resolution hyperspectral images. In MFNet, we use the frequency feature extraction module(FFEM) to extract the frequency feature. On this basis, a multidimensional features extraction module(MFEM) is designed to learn and fuse multidimensional features. In addition, experimental results on two public datasets demonstrate that MFNet achieves state-of-the-art performance.
文摘Albeit with the notable benefits associated with Dirichlet crash frequency models and spatial ones,there is little research dedicated to exploring their combined advantages.Such ensemble approach could be a viable alternative to existing models as it accounts for the unobserved heterogeneity by relaxing the constraints of specific distribution placed on the intercept while addressing the spatial correlations among roadway entities.To fill this gap,the authors aimed to develop Dirichlet semi-parametric models over the overdispersed generalized linear model framework while also incorporating spatially structured random effects using a distance-based weight matrix.Five models were developed which include four semi-parametric with flexible intercept and one parametric base model for comparison purposes.The four semi-parametric models entailed two models with a popular specification of stick-breaking Dirichlet process(DP)and two models with an alternative approach of Dirichlet distribution(DD),which are first applied in the field of traffic safety.All four models were estimated for mixture of points(discrete)and mixture of normals(continuous).The posterior density plots for the precision parameter justified the employment of the flexible Dirichlet approach to fit the crash data and supported the assumed prior for the precision parameter.All four Dirichlet models demonstrated the presence of distinct subpopulations suggesting that the intercepts of the models were not generated from a common distribution.The DP model based on mixture of normals illustrated better performance indicating its potential superiority to fit both insample and out-of-sample crash data.This finding indicated that the approach of continuous densities,unlike discrete points,may lend more flexibility to fit the data.
基金supported by ZTE Industry-University-Institute Cooperation Funds under Grant No.IA20220800001。
文摘A wide passband frequency selective surface(FSS)is proposed using a five-layer stacked structure.The proposed structure applies four layers of dielectric plates and five layers of metal patches to provide a passband and exhibits more stable frequency responses and lower insertion loss under wide-angle oblique incidence compared with the typical three-layer metal-dielectric structure.According to the simulation results,the proposed FSS can achieve a passband range of 1.7-2.7 GHz with an insertion loss of less than 0.5 d B and a relative bandwidth of 44.1%,and it can preserve stable transmission characteristics with the incident angle ranging from 0°to 45°.
文摘Hypersonic Glide Vehicles(HGVs)are advanced aircraft that can achieve extremely high speeds(generally over 5 Mach)and maneuverability within the Earth's atmosphere.HGV trajectory prediction is crucial for effective defense planning and interception strategies.In recent years,HGV trajectory prediction methods based on deep learning have the great potential to significantly enhance prediction accuracy and efficiency.However,it's still challenging to strike a balance between improving prediction performance and reducing computation costs of the deep learning trajectory prediction models.To solve this problem,we propose a new deep learning framework(FECA-LSMN)for efficient HGV trajectory prediction.The model first uses a Frequency Enhanced Channel Attention(FECA)module to facilitate the fusion of different HGV trajectory features,and then subsequently employs a Light Sampling-oriented Multi-Layer Perceptron Network(LSMN)based on simple MLP-based structures to extract long/shortterm HGV trajectory features for accurate trajectory prediction.Also,we employ a new data normalization method called reversible instance normalization(RevIN)to enhance the prediction accuracy and training stability of the network.Compared to other popular trajectory prediction models based on LSTM,GRU and Transformer,our FECA-LSMN model achieves leading or comparable performance in terms of RMSE,MAE and MAPE metrics while demonstrating notably faster computation time.The ablation experiments show that the incorporation of the FECA module significantly improves the prediction performance of the network.The RevIN data normalization technique outperforms traditional min-max normalization as well.
基金supported by the Chinese Academy of Science"Light of West China"Program(2022-XBQNXZ-015)the National Natural Science Foundation of China(11903071)the Operation,Maintenance and Upgrading Fund for Astronomical Telescopes and Facility Instruments,budgeted from the Ministry of Finance of China and administered by the Chinese Academy of Sciences。
文摘This paper addresses the performance degradation issue in a fast radio burst search pipeline based on deep learning.This issue is caused by the class imbalance of the radio frequency interference samples in the training dataset,and one solution is applied to improve the distribution of the training data by augmenting minority class samples using a deep convolutional generative adversarial network.Experi.mental results demonstrate that retraining the deep learning model with the newly generated dataset leads to a new fast radio burst classifier,which effectively reduces false positives caused by periodic wide-band impulsive radio frequency interference,thereby enhancing the performance of the search pipeline.
基金Open Access funding enabled and organized by Projekt DEAL.
文摘Large-scale quantum networks require dynamic and resource-efficient solutions to reduce system complexity with maintained security and performance to support growing number of users over large distances.Current encoding schemes including time-bin,polarization,and orbital angular momentum,suffer from the lack of reconfigurability and thus scalability issues.Here,we demonstrate the first-time implementation of frequency-bin-encoded entanglement-based quantum key distribution and a reconfigurable distribution of entanglement using frequency-bin encoding.Specifically,we demonstrate a novel scalable frequency-bin basis analyzer module that allows for a passive random basis selection as a crucial step in quantum protocols,and importantly equips each user with a single detector rather than four detectors.This minimizes massively the resource overhead,reduces the dark count contribution,vulnerability to detector side-channel attacks,and the detector imbalance,hence providing an enhanced security.Our approach offers an adaptive frequency-multiplexing capability to increase the number of channels without hardware overhead,enabling increased secret key rate and reconfigurable multi-user operations.In perspective,our approach enables dynamic resource-minimized quantum key distribution among multiple users across diverse network topologies,and facilitates scalability to large-scale quantum networks.
基金K.M.acknowledges funding from the Canada Foundation for Innovation John R.Evans Leaders Fund(Project 35552)Ontario Research Fund—Research Infrastructure(Project 35552),the Waterloo Institute for Nanotechnology(WIN-NRC seed grant),and a Mitacs Globalink Research Award.
文摘This paper presents a novel technique for low-power generation of frequency combs(FC)over a wide frequency range.It leverages modal interactions between electrical and mechanical resonators in electrostatic NEMS operating in air to provide a simple architecture for FC generators.A biased voltage signal drives the electrical resonator at resonance which is set to match an integer submultiple of twice the mechanical resonator’s resonance.Experimental results demonstrate that the NEMS displacement exhibit more than 150 equidistant peaks in the case of a 2:1 modal interaction and more than 60 equidistant peaks in the case of a 1:1 modal interaction.In both cases,the Free Spectral Range(FSR)was equal to the mechanical resonance frequency.Comparison between the FCs generated by the 2:1 and 1:1 modal interactions demonstrate the superiority of the former in terms of bandwidth and stability.The superior phase coherence of the FC generated via the 2:1 modal interaction was demonstrated via time-domain analysis.Our technique has the flexibility to generate multiple frequency combs and to fine-tune their FSR depending on the number of mechanical modes accessible to and the order of the activated modal interaction.It can be integrated into portable devices and is well aligned with modern miniaturization technology.
文摘伴随着光纤技术的快速发展,光纤网络已部署于航空航天、舰船、数据中心和工业物联网中。传统的光时域反射仪(optical time domain reflectometer,OTDR)因原理限制,难以实现高分辨率测试,在上述复杂场景中应用受限。基于瑞利散射的光频域反射(optical frequency domain reflection,OFDR)技术可实现极高的空间分辨率、高传感灵敏度和快速的测试速率,该系列产品适用于光器件、光模块、短距离光网络的测试和故障排除,可实现从器件到光学链路全范围的插损、回损和长度测量。文中基于光频域反射法原理设计实现了一套光纤链路检测系统,针对偏振衰落效应和激光器非线性扫频等难题进行了研究,在112 m的测试链路上实现了20μm空间分辨率。
文摘The FDR automatic soil moisture sensor must determine reference frequency in the air and water. Experimental studies show that the water reference frequency is influenced by water temperature. The variation of the reference frequency of the sensor is measured with the change of the water temperature,then analysis the influence of the volume water content measurement of the sensor,analysis found that the error is not more than 3% for the measurement of the volumetric water content of the temperature.
文摘Orthogonal Time Frequency and Space(OTFS) modulation is expected to provide high-speed and ultra-reliable communications for emerging mobile applications, including low-orbit satellite communications. Using the Doppler frequency for positioning is a promising research direction on communication and navigation integration. To tackle the high Doppler frequency and low signal-to-noise ratio(SNR) in satellite communication, this paper proposes a Red and Blue Frequency Shift Discriminator(RBFSD) based on the pseudo-noise(PN) sequence.The paper derives that the cross-correlation function on the Doppler domain exhibits the characteristic of a Sinc function. Therefore, it applies modulation onto the Delay-Doppler domain using PN sequence and adjusts Doppler frequency estimation by red-shifting or blue-shifting. Simulation results show that the performance of Doppler frequency estimation is close to the Cramér-Rao Lower Bound when the SNR is greater than -15dB. The proposed algorithm is about 1/D times less complex than the existing PN pilot sequence algorithm, where D is the resolution of the fractional Doppler.
