101 results about "Band spectrum" patented technology

The invention discloses a photoacoustic spectrum-confocal Raman spectrum multi-view-field high-dimensional imaging method and a photoacoustic spectrum-confocal Raman spectrum multi-view-field high-dimensional imaging device. Nanosecond pulse laser with adjustable wavelength is generated through a continuous spectrum nanosecond pulser and a narrow-band spectrum tuning module and cooperates with a microscopic detection module, a depth imaging module, a galvanometer system and an electric displacement table to obtain spatial three-dimensional data, a photoacoustic spectrum and a Raman spectrum under multiple view fields; shallow photoacoustic spectrum-Raman spectrum tomography and deep three-dimensional photoacoustic imaging are realized; correcting a photoacoustic spectrum quantitative error by using a spectrum decoupling algorithm, and fusing the multi-view-field data to generate multi-view-field high-dimensional photoacoustic spectrum-Raman spectrum data; high-dimensional data features are extracted based on a 3D convolution instance segmentation network, and quantitative analysis and spatial distribution visualization of blood oxygen, inflammatory factors and the like are achieved. According to the method, the limitation that substances are difficult to analyze in photoacoustic imaging is overcome, the contradiction between the imaging resolution and the penetration depth is solved, and an accurate analysis means is provided for diagnosis of diseases such as knee osteoarthritis and sepsis.

The invention relates to the technical field of soil detection, and particularly discloses a rapid soil heavy metal pollution detection method based on a spectrum technology, which comprises the following steps: obtaining dual-band spectral reflectivity data and temperature and humidity parameters of to-be-detected soil, carrying out spectral characteristic compensation, generating a spectrum curve, and obtaining environment compensation parameters; extracting spectral characteristic parameters from the spectral curve, combining the spectral characteristic parameters to construct a multi-dimensional spectral characteristic matrix, and inputting the multi-dimensional spectral characteristic matrix into a CNN-RF hybrid model for soil heavy metal pollution detection to obtain a heavy metal pollution detection result. According to the method, in-situ lossless rapid detection of the soil heavy metals is achieved through the dual-band spectrum and the environment dynamic compensation technology, interference of environmental factors on spectral measurement is effectively reduced, the detection efficiency is improved, the pollution condition of the heavy metals in the soil is accurately recognized through the CNN-RF mixed model, the reliability and precision of soil heavy metal pollution are improved, and the method is suitable for popularization and application. Corresponding treatment measures can be taken in time, and the safety of the soil environment is guaranteed.

The invention discloses a visual detection optimization control method, device and equipment based on digital twinning and a storage medium, and relates to the technical field of visual detection, and the method comprises the steps: obtaining an analog image and a real image in a digital twinning environment, and carrying out the frequency domain feature transformation of the analog image and the real image, thereby obtaining an image spectrum feature; a frequency domain alignment model is established based on multi-band spectrum envelope guide residual mapping, and the structure of virtual and real image spectrum features is kept aligned; further extracting features through multi-scale convolution and channel dependence mapping to obtain virtual-real fusion features; quantifying channel similarity and establishing a covariance regularization constraint, performing channel correction on the cross-domain features, and eliminating feature drift to obtain second virtual-real fusion features; and finally, performing visual detection and micro defect identification based on the features. The problem that a virtual sample and a real sample are different in local texture structure and channel distribution is solved.

The invention discloses an unmanned aerial vehicle data link communication system for beyond-visual-range cluster measurement and control, and relates to the field of communication transmission. The system comprises a dynamic topology construction module which constructs a dynamic network topology according to unmanned aerial vehicle node parameters and selects an optimal communication path; the spectrum resource management module is used for optimizing spectrum allocation of multiple frequency bands and dynamically allocating spectrum resources; the relay node selection module is used for selecting an optimal relay node and optimizing the energy consumption and the signal quality of the relay node; the clock synchronization module is used for performing high-precision clock synchronization through clock skew modeling and correcting frequency drift; the channel distribution module is used for dynamically distributing communication channels; and the fault detection module monitors the fault state of the communication link of the unmanned aerial vehicle in real time through a fault monitoring model, and performs anomaly detection. Through dynamic topology construction, intelligent relay node selection and multi-dimensional utility function optimization, the flexibility, stability and resource allocation efficiency of a communication network are effectively improved, and efficient and reliable communication of an unmanned aerial vehicle cluster is ensured.

The invention discloses a multi-band spectrum resource allocation method based on A2-MADDPG, and belongs to the technical field of MAB attention mechanisms. According to the technical scheme, a communication system framework is constructed, and a multi-band resource allocation environment is defined; modeling is carried out for frequency band characteristics, and cooperative work is realized; extracting multi-scale features of a system state by adopting self-adaptive multi-scale convolution and a dynamic attention mechanism; the hyper-parameters are optimized through an evolutionary strategy, and optimal resource allocation is achieved; a dynamic reward mechanism is established, weights are adjusted in real time according to actions and environments, and short-term and long-term targets are considered. The beneficial effects are that the method combines multi-agent deep reinforcement learning, adaptive multi-scale convolution, an attention mechanism and an evolutionary strategy to construct a multi-band collaborative resource allocation scheme, and can reduce inter-band loss to improve the spectrum utilization rate; the self-adaptive scale convolution and the dynamic attention mechanism are fused, the self-adaptive capability of the algorithm is improved, the system can keep high performance under different network environments and demand changes, interference is remarkably reduced, and the spectrum utilization rate is improved.

The invention relates to a spectrum occupancy rate calculation method based on full-band spectrum data, and belongs to the field of radio signal spectrum analysis. According to the invention, based on full-band spectrum data at the same sampling time, frequency band division is carried out, and a full-band signal is divided into a plurality of sub-bands; calculating level thresholds of all sub-bands based on the full-band spectrum data at the primary sampling moment; and estimating the occupancy rate based on the frequency band spectrum data at different sampling moments within the measurement time. According to the method, a plurality of interval thresholds in each sub-band are obtained by means of the histogram, the between-class variance and the like, the method can be used for an abnormal electromagnetic signal monitoring scene, the abnormal signal alarm false alarm rate is further reduced, and the electromagnetic space safety supervision capability is improved.

The invention relates to the technical field of unmanned aerial vehicle radio detection, in particular to an unmanned aerial vehicle detection and judgment method based on cooperation of an air-ground system and radio detection, and the method comprises the steps: finding a suspected target through scanning of a multi-frequency-band radar, triggering an infrared thermal imager carried by an aerial unmanned aerial vehicle to approach an ultra-wide-frequency-band spectrum sensor for verification; accurate positioning is realized by combining ground photoelectric equipment multi-view triangulation positioning and a signal arrival time difference algorithm, a command and control center calculates threat levels through a multi-source data fusion engine, and navigation decoy of dynamically adjusting intensity is adopted for targets smaller than or equal to two levels. Hard killing of ground laser locking rotor center and air electromagnetic pulse synchronous coverage is executed on a target of more than or equal to three levels, the equipment state is synchronized in real time through an optical fiber link, high-energy-consumption equipment is intelligently closed after threat is relieved, a platform starting strategy is optimized, a detection closed loop of'three-dimensional perception-graded treatment-intelligent energy saving 'is constructed, and the intelligent energy saving effect is achieved. And an efficient and accurate low-altitude safety protection scheme is provided for a sensitive area.

The invention discloses a speech enhancement method and device based on full-band and sub-band fusion, a terminal and a medium. The method comprises the following steps: determining an initial noisy speech full-band spectrum according to a noisy speech signal; decomposing an amplitude spectrum corresponding to the full-band spectrum of the initial noisy speech into three sub-bands including a low-frequency band, a middle-frequency band and a high-frequency band through a sub-band enhancement module, performing speech enhancement based on each sub-band, and determining an initial sub-band enhanced amplitude spectrum; determining an initial full-band enhancement spectrum through a full-band enhancement module according to the initial noisy speech full-band spectrum; fusing the preliminary full-band enhancement spectrum and the preliminary sub-band enhancement amplitude spectrum through a spectrum fusion module, and determining a full-band-sub-band fusion spectrum; determining a fusion complex value mask based on the full-band-sub-band fusion spectrum; and determining a target speech enhancement spectrum according to the fused complex value mask and the initial noisy speech full-band spectrum. The problems of insufficient information interaction and insufficient feature fusion in the full-band and sub-band fusion process in the prior art are effectively solved.

The invention provides an intelligent veterinary drug residue detection system and method based on multi-dimensional spectral characteristics, and relates to the technical field of veterinary drug residue detection, spectral response data of a to-be-detected sample solution in a visible light band and a near-infrared band is collected, and the residue discrimination degree of veterinary drug residues under any wavelength when the concentration dependence change exists is determined; abnormal wave band combinations with quantitative response to veterinary drug residues are screened out from the visible light wave band and the near-infrared wave band respectively, the residual peak deviation of the veterinary drug residues in the region of interest of the spectrum is determined according to the abnormal wave band combinations, and trace residue marks in the multiple spectrum intensity distribution modes are determined; and performing in-situ nondestructive screening on veterinary drug residues in the to-be-detected animal-derived food sample according to the residue discrimination and the trace residue mark. According to the method, on the premise that the dual-band spectrum collaborative response is enhanced, the veterinary drug residue characteristics can be directionally distinguished and detected, so that the veterinary drug residue screening accuracy is improved.

The invention relates to the technical field of tea garden insect pests, and discloses a tea garden insect pest early warning system based on machine learning, which comprises a multi-mode sensing device, an edge intelligent terminal and a cloud decision platform. According to the tea garden insect pest early warning system based on machine learning, a narrow-band spectrum is dynamically selected through an adjustable spectrum imaging unit, image and sound wave features are synchronously acquired in combination with a vibration wave acquisition module, a heterogeneous calculation module of an edge intelligent terminal processes multi-source data in parallel, and a dynamic feature library management module updates a feature template in real time; the incremental learning mechanism triggers high-resolution imaging when unknown pests are detected, the meta-learning framework supports rapid migration training of a new model, the framework breaks through the limitation of a traditional single spectrum and a fixed model, and through multi-dimensional data fusion and adaptive learning, the pest recognition efficiency and adaptability are remarkably improved. The problems of low manual inspection efficiency and poor adaptability of traditional equipment are solved.

The invention discloses a method and a device for analyzing and correcting influence of phase errors between sub-bands of step-frequency ultra-wideband SAR (Synthetic Aperture Radar) data, and belongs to the technical field of synthetic aperture radar signal processing. The method comprises the following steps: carrying out amplitude error estimation and compensation on each sub-band echo signal; preliminarily estimating a phase error by using the partial overlapping characteristic of the sub-band spectrum; performing distance compression on each sub-band signal; establishing an optimization model based on a maximum norm, and solving an accurate phase error by taking the preliminarily estimated phase error as an iteration initial value; carrying out sub-band splicing and distance resolution compression on the sub-band signals after phase error compensation; and finally, a corrected SAR image is obtained through a CS imaging algorithm. According to the method, extra calibration signals are not needed, high-precision phase error correction can be achieved only based on echo data, the calculation efficiency is high, and the method is suitable for a high-resolution SF-SAR system.

The invention discloses a low-complexity cascade predistortion device and method for out-of-band spectrum suppression. The device comprises a predistorter unit, a model identification and trimming unit, a subtracter, a DAC unit, a power amplifier PA, a transmitting antenna, an ADC unit, a power normalization unit and a filter. The predistorter unit comprises a plurality of stages of predistortion modules; and the model identification and trimming unit is used for carrying out modeling, screening and parameter extraction on each stage of pre-distortion module, outputting a screened coefficient model coefficient and feeding back the coefficient model coefficient to the pre-distorter unit. The method is oriented to strong distortion conditions and specified frequency band out-of-band suppression, is low in training sample requirement, is low in implementation complexity, and is convenient to deploy a digital pre-distortion technical route on general programmable hardware, so that the supervision and standard indexes are met, the energy consumption and the engineering cost are reduced, and the adaptability to complex and changeable scenes is improved.

The invention discloses a wildfire monitoring method based on a multi-source time sequence remote sensing image, and relates to the technical field of remote sensing geoscience application, and the method comprises the steps: obtaining multi-source remote sensing data, constructing a time sequence waveband data set, and constructing a high-temperature thermal anomaly index through the atmospheric top reflectivity of a three-waveband spectrum; based on the high-temperature thermal anomaly index, extracting a first-level fire point pixel, a saturated pixel and a high-radiance pixel from the multi-source remote sensing data, and converting the saturated pixel and the high-radiance pixel into the first-level fire point pixel; extracting saturated pixels and high-radiance pixels from missed fire cluster cores in the multi-source remote sensing data; based on limiting conditions, building coverage area data and a high-frequency ground object mask, noise pixels in the first-level fire point pixels are removed, and second-level fire point pixels are obtained; transmitting to a local client and a cloud geographic space analysis platform, and creating a fire point list. Multi-source long-time sequence optical remote sensing data are combined, and the spatial and temporal distribution of the large-area wildfire is quickly and automatically identified.

The invention provides a real-time detection system and method for fluorescence signals of bacteria in air, and the system comprises a narrow-band spectrum light source which is used for radiating narrow-band exciting light matched with fluorescent substances in the bacteria, and exciting the fluorescent substances in the bacteria to generate fluorescence; the collimation system is used for collecting light energy output by the narrow-band spectrum light source and converting divergent light beams into parallel light beams; the light conversion aggregator is used for distinguishing bacterial fluorescence and Mie scattered light, converting the effective bacterial fluorescence into secondary fluorescence and converging the secondary fluorescence, and the narrow-band exciting light passes through the light conversion aggregator after being collimated by the collimation system; the detection system is used for converting the secondary fluorescence signal into an electric signal; the control system is respectively connected with the detection system and the narrowband spectrum light source and is used for controlling the detection system and the narrowband spectrum light source to work; and the flow field system is used for driving air flow to flow through the light conversion aggregator. The structure is simple, and optical signal interference generated by Mie scattering can be inhibited.

The invention provides a reflection wave band spectrum domain influence uncertainty quantification method and device for reference star radiation reference transmission, and relates to the technical field of data processing, and the method comprises the steps: inputting ground features, atmosphere and observation geometric parameters into a radiation transmission model, and simulating to generate high-resolution atmosphere top spectrum radiance data as a real radiation spectrum; carrying out convolution on the data and a reference instrument spectral response function to obtain a reference instrument measurement spectrum; the reference instrument measurement spectrum and the real radiation spectrum are respectively convolved with the target instrument spectral response function, the relative difference percentage is calculated, the root mean square is taken, and a first standard uncertainty is obtained; positive and negative deviation is applied to the central wavelength of the reference instrument, and a wavelength sensitivity coefficient is obtained through calculation and normalization; the variance is synthesized into related parameters to obtain the total standard uncertainty, then the coverage factors are fused, and the expansion uncertainty is output. The accuracy of the quantification result is improved.

The invention discloses a dual-band reflection optical film system and an optical lens. The film system comprises seven layers of all-dielectric optical films which are a trititanium pentoxide layer, a silicon dioxide layer, a trititanium pentoxide layer, a silicon dioxide layer, a zirconium dioxide layer, a trititanium pentoxide layer and a silicon dioxide layer in sequence, the reflectivity of the film system in the blue light band of 385-445 nm is 20-35%, the reflectivity in the near-infrared band of 780-1400 nm is larger than or equal to 40%, the reflectivity in the visible light band of 500-780 nm is smaller than or equal to 1%, and the film system is of an absorption-free structure. The film system is designed based on an all-dielectric pure interference structure, accurate regulation and control of multiband spectrums are achieved in a single film system, and the film system has the three characteristics of short-wave blue light blocking, near-infrared efficient heat insulation and extremely high visible light transmission. The invention fundamentally overcomes the technical bottlenecks of single function, heat accumulation caused by photo-thermal conversion, light transmittance damage and the like of the traditional absorption type protective lens, and is suitable for various visual protection products with strict requirements on color rendition degree and thermal protection.

The invention relates to the technical field of deep mineral detection, and discloses a deep mineral rapid detection method based on multiband spectrum fusion, which comprises the following steps: step 1, building a detection system; 2, carrying the detection system on a detection platform, and synchronously collecting three types of spectral data and course angle, pitch angle, roll angle and speed parameters; step 3, data preprocessing; and step 4, a cross-modal fusion module realizes alignment of three types of data through cross-modal feature matching, and then weighted fusion is carried out. According to the deep mineral rapid detection method based on multi-band spectrum fusion, an attitude sensing module collects a course angle, a pitch angle, a roll angle and speed parameters in real time, a data preprocessing module cooperates with a cross-modal fusion module, and an attitude correction matrix is constructed based on Euler angle transformation to carry out geometric correction on multi-band spectrum data; the detection platform advances according to a preset height and speed working condition, a high-precision inertial navigation system is not needed, and the cost and the detection precision are both considered.

The invention discloses an agricultural product pesticide residue rapid detection integrated system based on spectral imaging, which belongs to the technical field of agricultural product quality safety detection, and comprises a multi-band spectrum acquisition module, a differential spectrum analysis module, a feature fingerprint identification module, an automatic calibration module, a multi-pesticide synchronous detection module and a rapid output module, the surface of an agricultural product is scanned by adopting a multi-band near infrared spectrum technology, matrix interference of a sample is eliminated through differential spectrum analysis, synchronous detection of various pesticide residues is realized by utilizing characteristic fingerprint identification, long-term detection precision is kept through an automatic calibration mechanism, detection is completed within 90 seconds, and a result is output. The method is high in detection speed, high in accuracy and high in universality, and an efficient and practical technical scheme is provided for agricultural product safety monitoring.

The present application relates to a kind of spectrum occupation degree calculation method based on full-band spectrum data, belong to radio signal spectrum analysis field.The present application is based on the full-band spectrum data of same sampling time, and the full-band signal is divided into several sub-bands;Based on the full-band spectrum data of once sampling time, the level threshold of all sub-bands is calculated;Based on the spectrum data of different sampling time in measurement time, the occupation degree is estimated.The present application obtains multiple interval threshold values in each sub-band by histogram, interclass variance and other means, which can be used in abnormal electromagnetic signal monitoring scene, further reduces the false alarm rate of abnormal signal alarm, and improves electromagnetic space safety supervision ability.

The invention provides a multi-task multi-frequency component periodic vibration environment test condition determination method, which belongs to the field of equipment environment engineering, and comprises the following steps: S1, obtaining a vibration standardized data set X containing n task profiles; s2, obtaining a first type component and a second type component of the vibration actual measurement data; s3, taking the equivalent duration tL of the spectrum values corresponding to the rest L-1 groups of periodic frequencies fa as a reference, and adopting the following formula to convert the equivalent duration tL of the spectrum values corresponding to the rest L-1 groups of periodic frequencies fa; s4, obtaining a group of parameters with spectrum values and duration time; s5, taking the duration time TR of the random spectrum as a reference, converting the duration time of the periodic frequency fa into TR, and calculating a converted spectrum value; and S6, superposing the periodic spectrum or the narrow-band spectrum after the time unification with the random spectrum to form a vibration test condition for a vibration test. According to the invention, the frequency of the periodic / narrowband component can be directly determined, and the identification accuracy is higher.

The application provides a multi-band fusion industrial detection light source regulation method, system and medium, the method comprising: outputting a multi-band initial light source based on a quantum dot tuning wide spectrum emission mechanism; performing coordinated emission and time-sharing triggering on the multi-band initial light source based on a time sequence matrix to obtain a multi-band corresponding multi-path excitation light source; performing multi-band spectrum fusion and light mixing processing on the multi-band corresponding multi-path excitation light source based on an optical path transmission efficiency model to obtain a fusion light source; calculating a spectrum drift amount corresponding to a real-time temperature of a light source chip based on a spectrum drift function; calculating spectrum compensation information based on the spectrum drift amount, dynamically compensating the spectrum drift amount according to the spectrum compensation information, and performing spectrum drift dynamic correction; through multi-band light source fusion light mixing processing, and real-time analysis of the spectrum drift amount, spectrum dynamic correction is realized according to the thermal feedback of the real-time temperature of the light source chip, the emission center drift phenomenon caused by temperature drift is effectively eliminated, and the detection performance is improved.

A positive electrode active material includes a lithium composite metal oxide including nickel, cobalt, manganese, and aluminum. The positive electrode active material includes 85 mol % to 97 mol % of nickel, and 2 mol % to 5 mol % of cobalt, with respect to the total number of moles of metals other than lithium, and satisfies Expression (1): 0.25≤I550 / I700≤0.4. In Expression (1), I700 and I550 are respectively a maximum value of a peak intensity appearing in a range of 600 ppm to 800 ppm and a maximum value of a peak intensity appearing in a range of 450 ppm to 650 ppm when a spectral analysis (peak deconvolution) is performed on a 1D NMR center band spectrum extracted from a 2D 7Li Magic Angle Turning Phase Adjusted Spinning Sideband (MATPASS) NMR spectrum. A positive electrode and a lithium secondary battery are also included

The facility for detecting at least one characteristic parameter of a medium (M), has A DEVICE (3) illuminating the medium (M) with a visible polychromatic light, a device (4) exciting the medium (M) with a visible-light or infrared monochromatic radiation, a first device (5) collecting and conveying at least a first electromagnetic radiation emitted by the medium (M), a second device (6) collecting and conveying at least a second electromagnetic radiation emitted by the medium (M), at least one visible-band spectrometer (7) obtaining at least one visible-band spectrum of the medium (M) from the at least first electromagnetic radiation, at least one Raman spectrometer (8) obtaining at least one Raman spectrum of the medium (M) from the at least second electromagnetic radiation, and a device (9) detecting the at least one characteristic parameter from the at least one visible-band spectrum and the at least one Raman spectrum.

The application discloses a kind of metal material mechanics performance multi-parameter nondestructive evaluation method based on grating laser ultrasonic spectrum, the method utilizes high-energy pulsed laser beam to be irradiated on the surface of the metal material to be measured by micro-lens array and form periodic distribution transient grating, excite monochromatic ultrasonic Rayleigh wave of fixed wavelength in material interior, simultaneously, non-contact receive Rayleigh wave signal in the vicinity of the detection area by probe laser.The received Rayleigh wave signal is subjected to fast Fourier transform to obtain narrow-band spectrum, and the acoustic parameters such as sound velocity and acoustic nonlinear parameter can be calculated through the narrow-band spectrum.According to theoretical formula and experimental relationship calibration curve, the calibration relationship between the elastic modulus of metal material and sound velocity, hardness and acoustic nonlinear parameter, yield strength and acoustic nonlinear parameter, ultimate tensile strength and acoustic nonlinear parameter is established respectively, so as to realize the quantitative evaluation of the mechanical properties of metal material.

A method for performing in-band spectrum sharing in a mixed TN-NTN system is provided. The method is performed by a user equipment (UE). The UE receives reference signal (RS) configuration information from a satellite or TN base station and obtains measurement results indicative of coupling loss or interference based on the RS configuration. The UE transmits a measurement report to the satellite or an uplink RS to the TN base station for resource allocation decisions. The UE operates according to resource allocation results indicating a partition of shared radio resources across frequency, time, or polarization domains to mitigate interference. Finally, the UE communicates using the allocated resource portion of the shared radio resources.

The application discloses a kind of multi-band spectrum prediction method and terminal of electric power wireless network, the power spectrum density data of each frequency band of electric power wireless network in history different time collected is preprocessed, the feature of power spectrum density data is extracted layer by layer by stacked encoder, the abstract feature of data can be gradually understood;Long-term memory network is used to the feature matrix of encoder output Time series feature learning, long-term dependence in data can be learned, so as to improve the generalization ability and prediction effect of prediction model;By introducing attention mechanism, this mechanism makes the model can dynamically adjust weight according to the importance of input data, more accurately capture key feature information, can dynamically focus on the most relevant part in input sequence, improve the stability and reliability of prediction.In this way, by the model of stacked encoder-long-term memory network-attention mechanism, the reliability and accuracy of spectrum prediction can be effectively improved.

This invention discloses a rapeseed quality detection method based on feature wavelength optimization and multi-model fusion. The method involves collecting full-band near-infrared diffuse reflectance spectral data of rapeseed samples and simultaneously measuring the true values ​​of physicochemical indicators to construct an original dataset. Based on the data distribution characteristics of each physicochemical indicator, the original dataset is divided into a training set and a test set. Preprocessing algorithms are selected for each physicochemical indicator to process the full-band near-infrared diffuse reflectance spectral data. Dimensionality reduction algorithms are selected for each physicochemical indicator to extract feature wavelengths from the preprocessed spectral data. Predictive models for each physicochemical indicator are established based on the training set, and the performance of the predictive models is verified using the test set to determine the optimal algorithm combination for each physicochemical indicator. The variable importance projection algorithm is used to calculate the comprehensive contribution score of each wavelength point in the full-band spectrum to the four physicochemical indicators. A multi-indicator comprehensive threshold is set, and the core light source wavelength of the portable device is determined from the high-score region.

This invention discloses a modulation domain speech enhancement method based on group sparse representation, comprising the following stages: a training phase: calculating sub-band spectra of a mixed training signal, speech signal, and noise signal at different modulation frequencies; obtaining signals of different clustered frames of the sub-band spectra through frame clustering analysis; training a joint sub-dictionary using signals from different clustered frames of the sub-band spectra, and then concatenating the sub-dictionaries into a mixed, speech, and noise structured dictionary. A testing phase: calculating the sub-band spectrum of the mixed test signal, preserving its acoustic and modulation phase; calculating the projection coefficients of the mixed test signal's sub-band spectrum onto the mixed structured dictionary through group sparse coding; multiplying the structured dictionary and projection coefficients to recover the speech and noise modulation amplitude spectra, and combining this with a noisy modulation phase inverse transform to obtain the acoustic amplitude spectrum; calculating the weights of the two estimated sets based on the Gini coefficient, and optimizing the estimation using a Wiener filter. This invention utilizes the structured characteristics of modulation domain features as prior information, improving the performance of single-channel speech enhancement.

The present application relates to the technical field of unmanned aerial vehicle communication anti-jamming, and particularly relates to an unmanned aerial vehicle anti-jamming system based on spectrum sensing. A spectrum sensing data acquisition module acquires multi-band spectrum signal data to generate a spectrum interference characteristic spectrum; a jamming source positioning module calculates a jamming source azimuth and distance based on the spectrum interference characteristic spectrum to generate a spatial distribution data set; a three-dimensional interference field construction module combines real-time altitude and geographic coordinates of the unmanned aerial vehicle to construct a three-dimensional interference field model containing a jamming intensity gradient and a propagation direction; an anti-jamming strategy generation module analyzes communication link vulnerability to dynamically adjust band switching priority and transmission power allocation weight; and a beamforming control module calculates antenna array phase offset and beam pointing angle to generate an adaptive beamforming parameter set. The system can improve the anti-jamming capability and communication stability of the unmanned aerial vehicle in a complex electromagnetic environment.

The present application relates to a kind of wide frequency power grid harmonic fast extraction method based on multiscale collaborative measurement, belong to the field of power system harmonic analysis, including: construct multi-window parallel data buffer, utilize smooth transition function dynamically calculate the confidence weight of each window;Adopt adaptive window function strategy to load rectangular window and Blackman window respectively, and preliminary obtain full-band spectrum parameter in combination with zero padding FFT, introduce phase alignment algorithm based on time zero point to eliminate the phase drift caused by non-synchronous sampling;For random interference, use stability filtering mechanism to carry out effectiveness check.The present application realizes the unity of wide frequency harmonic transient fast locking and steady-state high-precision measurement by organically combining multi-window parallel processing with S-shaped smooth transition mechanism, significantly improves the dynamic adaptability and anti-noise reliability of measurement, and provides reliable technical foundation for new type power system wide frequency oscillation monitoring and control.