77 results about "Maximum eigenvalue" patented technology

The invention provides an automatic early warning method for sudden weather in a target area, which belongs to the technical field of weather early warning, and comprises the following steps of: establishing a primary dense matrix by adopting adaptive filtering processing and a frequency domain signal separation technology, and generating a secondary dense matrix by applying a marine meteorological recognition model of a spiral progressive network structure; a dynamic statistical equation is used to calculate the physical coupling relationship of each parameter to establish a multi-scale weather process balance matrix, a maximum flow and minimum cut algorithm is used to optimize a weather system coupling relationship network to calculate a coupling degree matrix, and a dynamic threshold adjustment mechanism is established according to coupling strength parameters to adjust the early warning detection frequency. And based on a comparison result of the coupling degree moment order maximum characteristic value and a preset risk threshold value, establishing a grading early warning system and outputting a corresponding early warning signal to control an offshore oil and gas platform emergency response system. The technical problem that a multi-time scale weather process coupling relationship cannot be effectively processed is solved.

The invention relates to the field of reactor data processing, and particularly discloses a numerical calculation method for solving a neutron transport equation. The method comprises the following steps: 1, describing the geometry of a reactor core of a nuclear reactor by utilizing a construction entity geometric modeling method, and subdividing a curved surface grid along a curved surface structure; 2, directly expanding a neutron flux distribution function in each curved surface grid by using a multivariate polynomial; 3, deriving a neutron transport equation and an integral residual equation of a boundary condition under each curved surface grid; 4, solving geometric integral parameters of each curved surface grid; step 5, establishing an adaptive linear system; and step 6, iteratively calculating and solving a maximum characteristic value and an unknown expansion coefficient.

The invention belongs to the technical field of geographic information systems and PS-InSAR, and discloses a stress-strain-considered InSAR building deformation risk assessment method, which comprises the following steps: carrying out PS-InSAR processing on time sequence SAR image data to obtain a time sequence deformation result and carrying out precision verification; adopting a PS point correction method based on a self-adaptive building contour to screen out PS points of the building; selecting five indexes of building strain, building cumulative inclination amount, building cumulative settlement amount, building age and building structure material type; building risk assessment modeling is carried out by applying an analytic hierarchy process, consistency check and weight calculation are carried out, and the maximum characteristic value of a judgment matrix established by the five risk assessment indexes is calculated; grading the indexes of the model, performing risk assessment on each index, and drawing; and verifying a building risk assessment result in combination with actual dangerous house data. According to the invention, accurate evaluation of the deformation risk level of the building is realized.

The invention discloses an image classification method based on multi-manifold joint metric learning, and the method comprises the steps: 1) data preprocessing: constructing Grassmann features and SPD features; 2) constructing a double-flow kernel matrix, including constructing a Grassmann manifold kernel matrix and an SPD manifold kernel matrix, and respectively calculating a Grassmann test kernel matrix and an SPD test kernel matrix; 3) implementing multi-manifold joint metric learning, and extracting kernel feature vectors from the kernel matrix to calculate an intra-class scatter matrix and an inter-class scatter matrix; solving the projection matrix, sorting the feature vectors according to the sizes of the feature values, and selecting the feature vectors corresponding to the first plurality of maximum feature values to form the projection matrix; the method comprises the steps of (1) extracting different ground features, (2) carrying out feature fusion, completing extraction and implementing sample mapping, and (3) constructing a sample-level fusion convolutional neural network and completing classification of different ground features of an original image.The method belongs to the technical field of image processing and machine learning, and the classification accuracy can be remarkably improved.

The invention provides a production line parameter real-time scheduling method based on reinforcement learning, and belongs to the technical field of production lines, and the method comprises the steps: collecting sensor data to form an original state vector, reducing the dimension of the original state vector into a low-dimensional feature state vector through a state compression encoder, inputting the low-dimensional feature state vector corresponding to a microcosmic scheduling unit into a parameter decision model, and carrying out the real-time scheduling of the microcosmic scheduling unit; the attention weight coefficient of the model is determined by the product of the historical scheduling success rate, the element value of the difference degree matrix and the maximum characteristic value of the inter-stage sensitivity matrix, after a scheduling instruction is output, simulation evaluation is performed in a digital twin platform, and after the scheduling instruction passes, a real production line is issued for execution and multi-time-scale deviation indexes are collected; and calculating a comprehensive reward value according to the indexes, and storing an experience sample to an experience playback buffer area for model online update training, thereby solving the technical problem that production line parameter scheduling is difficult to consider multi-level state feature recognition and dynamic decision weight optimization at the same time.

The invention provides a quantization method and a quantization device, which can perform singular value decomposition on a first matrix formed by word embedding vectors to obtain the maximum characteristic value of each block matrix in M block matrixes, and can determine the quantity of quantization bits of each block matrix according to the maximum characteristic value of each block matrix. The method comprises the following steps of: dividing each block matrix into a plurality of block matrixes, quantifying each block matrix to obtain each quantized block matrix, combining each quantized block matrix to obtain a quantized first matrix, and representing the amount of semantic information carried in each block matrix by a maximum characteristic value of each block matrix to a certain extent, according to the embodiment of the invention, the quantized bit number can be allocated to each block matrix according to the amount of semantic information carried by each block matrix, so that different bit numbers can be allocated to the block matrixes carrying different semantic information, differential bit number allocation can be realized, semantic information loss can be reduced, and user experience can be improved.

The invention discloses a rapid explosion suppression and flame prevention method for a transformer, and belongs to the technical field of power equipment safety, and the method specifically comprises the steps: S1, collecting multi-physical field data in real time through a sensor network disposed in the transformer; s2, based on the multi-physical field data, determining an electrothermal instability index, a fractal dimension of fire nucleus growth, a maximum characteristic value of an irradiance Hessian matrix and a spectral coherence length attenuation rate; s3, combining the electrothermal instability index, the fractal dimension of fire core growth, the maximum characteristic value and the spectral coherence length attenuation rate to construct a characteristic vector; and S4, identifying the initial fault state based on the feature vector, correcting the initial fault state by using a preset physical model to obtain a final fault state, and triggering a grading response strategy according to the final fault state. And a solid data basis is provided for subsequent accurate diagnosis.

The invention provides a muscle fiber disorder imaging method, a muscle fiber disorder detection method, a muscle fiber disorder detection system and a storage medium, and belongs to the technical field of biomedical detection software. The method is technically characterized by comprising the following steps: S100, acquiring a boundary voltage difference matrix of muscle to be detected; s200, carrying out normalization processing on the delta V; s300, the contour line of the muscle to be detected is extracted, the area in the contour line is an EIT simulation model area, and then a network node coordinate matrix of the muscle simulation model area to be detected is obtained; s400, normalizing the delta V and inputting the network node coordinate matrix into a pre-trained maximum eigenvalue matrix solving network to solve a maximum eigenvalue matrix delta lambda max; and S500, imaging is carried out. By means of the technical scheme, muscle changes can be visually displayed from the image in the mode of the three-dimensional image and the global muscle fiber disorder coefficient, and the muscle fiber disorder degree is quantified from the numerical value.

The invention discloses a neural network prediction-based LMS (Least Mean Square) power inversion adaptive step length adjustment method, which relates to the technical field of satellite navigation and signal processing, and comprises the following steps of: constructing and training a neural network model, inputting a current frame input signal power and a maximum characteristic value, and outputting a current frame prediction step length; compared with an LMS power inversion algorithm with a fixed step length, the method introduces a neural network to dynamically predict the optimal step length, solves the problem that traditional empirical step length adjustment is not adaptive, reflects interference intensity in combination with a covariance matrix and a maximum characteristic value of an input signal, realizes interference sensing adjustment, can increase the step length under the condition of relatively strong interference, and improves the accuracy of interference sensing adjustment. Convergence is accelerated and the tracking capability is enhanced; and under the condition of weak interference, the step length is reduced, so that steady state imbalance is reduced, and the anti-interference performance is improved. The method can solve the problem that in an existing LMS power inversion algorithm, step length parameters are difficult to adjust in a self-adaptive mode, and especially the problem that convergence is slow or system divergence is caused by improper step length setting in a multi-interference channel or dynamic environment.

The invention discloses a shared power bank battery safety early warning method and system based on big data, and relates to the technical field of battery safety management, and the method comprises the steps: constructing an information flow network, extracting the information propagation intensity of a directed edge, calculating the abnormal propagation rate of equipment, and generating a cumulative propagation effect in a time window. And calculating a fractional derivative of the cumulative propagation effect, constructing an adjacent matrix and calculating a maximum characteristic value, and calculating the spectral diffusion contribution and a comprehensive abnormal value of the equipment in combination with the cumulative propagation effect and the maximum characteristic value. According to the method, the dynamic strain field is combined with the flow field velocity analysis, the spatial distribution description capability of the internal structural risk of the battery is enhanced, the comprehensive coupling factor is combined with the information flow network propagation model, and the abnormal propagation mechanism modeling capability of the shared power bank in the large-scale mass use process is improved.

The invention relates to the technical field of energy storage system parallel control, and particularly discloses an optimization control method and system based on an energy storage inverter, and the method comprises the steps: building a phase cooperation matrix; constructing a weighted parallel network operator, analyzing the dynamic change of the maximum characteristic value based on a percolation phase change theory, and generating a percolation early warning signal in combination with an inversion probability coefficient; supercells are divided through a reformed group algorithm; solving a virtual impedance scale transformation coefficient fusing the maximum characteristic value of the super-cell and a power margin correction term, dynamically configuring the virtual impedance scale transformation coefficient, and injecting a chaotic disintegration signal to cut off long-range correlation; monitoring the percolation long-range correlation disintegration degree and the circulation entropy, and judging whether the system is stable or not; the system comprises a collaborative acquisition module, a phase change analysis module, an anchor point analysis module, a signal injection module and a stability analysis module. The method is beneficial for suppressing circulating current resonance and synchronous instability of multi-inverter parallel connection, and is suitable for a large-scale energy storage cluster grid-connected or off-grid operation scene.

This invention discloses a method, system, storage medium, and electronic device for suppressing mutual interference in spaceborne synthetic aperture radar (SAR) scattered waves based on matched filtering. The method includes the following steps: estimating relevant parameters of mutual interference based on prior orbit information and time-frequency relationships; performing range pulse compression on the interference signal in the echo data; detecting whether the range pulse contains interference using the maximum eigenvalue sequence detection method; calculating the relative signal-to-interference ratio (SIR) of the interference pulse; reconstructing and extracting the interference signal under high SIR and low SIR conditions using the eigenspace projection method and robust principal component analysis method, respectively; filtering it from the echo data; and then performing inverse range compression on the data to restore the interference-free normal echo. This invention effectively utilizes the compression gain of mutual interference, thereby more effectively extracting mutual interference signals from SAR echoes and achieving high-precision interference suppression while better protecting useful signals.

This invention provides a radar target detection method and apparatus based on dual-polarization maximum eigenvalues. The method includes: receiving first polarization echo data and second polarization echo data, and determining a detection unit and at least two reference units based on the first polarization echo data and the second polarization echo data; for any unit, determining a cross-covariance matrix based on the first polarization echo data and the second polarization echo data, and determining the maximum eigenvalue corresponding to the unit based on the cross-covariance matrix; determining an average maximum eigenvalue based on the maximum eigenvalues ​​corresponding to each reference unit; and determining a target detection result based on the maximum eigenvalue corresponding to the detection unit, the average maximum eigenvalue, and a preset threshold factor, wherein the target detection result includes whether the target exists or does not exist. This reduces the computational complexity of radar target detection, fully leverages the distinguishability of target echo signals and sea clutter, and improves the detection accuracy of radar target detection.

The embodiment of the invention provides a shear wave velocity inversion method and device, electronic equipment and a storage medium, and relates to the technical field of seismic wave inversion. The method comprises the following steps: acquiring a plurality of frequency dispersion energy diagrams from different seismological observation systems in the same observation area, preprocessing the plurality of frequency dispersion energy diagrams, and constructing an image matrix; constructing a target kernel matrix by using the image matrix and a sub-sample set obtained from the image matrix; performing eigenvalue decomposition on the target kernel matrix, and performing image reconstruction according to the obtained maximum eigenvalue of the target kernel matrix to obtain a fusion frequency dispersion energy diagram with higher resolution and higher quality; and performing shear wave velocity inversion by using the fused frequency dispersion energy diagram to obtain a shear wave velocity profile of the observation area, thereby remarkably improving the accuracy and reliability of the shear wave velocity inversion.

The invention discloses a potential safety hazard investigation and judgment management system and method, and relates to the technical field of potential safety hazard investigation. The system comprises a hidden danger checking module, a hidden danger identification and classification module, a hidden danger grade determination module, a hidden danger regulation and acceptance module and a data statistics and ledger management module which are operated in sequence. According to the technical scheme, an AHP analytic hierarchy process is introduced, a hidden danger grade judgment value G is calculated, a default weight and a user-defined mode can be set, and construction of a judgment matrix is supported, so that the maximum characteristic value, the consistency index CI and the proportion CR are calculated, and only CRlt is calculated; and the judgment result is valid when the judgment result is 0.1, so that quick response in a conventional scene is ensured, and decision support is provided for major hidden dangers, thereby effectively determining the allocation priority of the renovation resources, ensuring that subsequent high-risk hidden dangers can be preferentially handled, and realizing accurate grading of the potential safety hazards.

The invention relates to the technical field of new energy grid-connected stability control, in particular to a small-signal steady-state discrimination method and system for a double-fed wind power plant controlled by a virtual synchronous machine and a medium, and the method comprises the steps: building a linear state space model for a double-fed fan controlled by a single virtual synchronous machine; listing a characteristic polynomial, and deducing a single-machine stability criterion inequality by using a Routh criterion; the method comprises the following steps: constructing a full-order state space matrix for a doubly-fed wind power plant comprising M units; decomposing the matrix into M independent subsystems, and performing eigenvalue decomposition on the current collection network impedance matrix to obtain a maximum eigenvalue; the M independent subsystems are equivalent to M-1 isolated subsystems without network interaction and an equivalent single-machine subsystem; and carrying out Routh criterion derivation on the equivalent single machine subsystem to obtain an explicit stability criterion inequality of the double-fed wind power plant grid-connected system and carrying out steady state discrimination. According to the invention, the problem that the stability boundary cannot be explicitly displayed under the complex power grid topology in the prior art is effectively solved.

The present application provides a quantization method and a quantization apparatus. The method comprises: performing singular value decomposition on a first matrix composed of word embedding vectors, so as to obtain a maximum eigenvalue of each block matrix among M block matrices; on the basis of the maximum eigenvalue of each block matrix, determining the number of quantization bits for each block matrix; performing quantization on each block matrix to obtain quantized block matrices; and, merging the quantized block matrices to obtain a quantized first matrix. The maximum eigenvalue of each block matrix can to some extent characterize the amount of semantic information carried in each block matrix, and the number of quantization bits for each block matrix can be allocated on the basis of the amount of semantic information carried in each block matrix, thereby allowing different bit quantities to be allocated to block matrices carrying different semantic information, achieving differentiated bit quantity allocation and reducing loss of semantic information, and accordingly improving user experience.

The invention discloses a high-directivity phased array synthesis method based on circular polarization axial ratio control, and belongs to the technical field of satellite communication. The method comprises the following steps: firstly, establishing a one-to-one correspondence relationship between an axial ratio and an electric field; then, an array directivity coefficient maximization problem model with strict axial ratio constraint is established; a non-convex fractional programming problem is converted into a Rayleigh entropy form, axial ratio constraint is converted into linear constraint, the axial ratio constraint is replaced with first-dimension excitation through excitation dimension reduction, an original problem is converted into an unconstrained optimization problem, and the converted unconstrained optimization problem is further converted into the Rayleigh entropy form; and finally, carrying out eigendecomposition on the converted standard Rayleigh entropy form to obtain an eigenvector corresponding to the maximum eigenvalue, and recovering the excitation of the original problem according to the conversion form. According to the method, accurate axial ratio control can be achieved, meanwhile, the maximum directivity coefficient can be obtained under the condition that the current axial ratio is limited, and the iteration process is avoided.

The invention discloses a graph convolutional network confrontation defense method based on adaptive frequency spectrum filtering. Firstly, spectral analysis is performed on a graph structure, and spectral distribution of graph data is obtained and evaluated; then, an adaptive spectrum filter is constructed based on the distribution, and the filter can automatically adjust parameters according to the degree distribution and the maximum characteristic value of the graph, so that the most robust spectrum interval under the attack resistance is adaptively determined; on the basis, invalid or harmful frequency spectrum components introduced by disturbance are weakened by a filter and are embedded into the propagation process of the graph convolutional network, and finally, the resistance of the model to confrontation disturbance is improved at the frequency domain level, and a stable prediction result is output. According to the method, negative effects caused by disturbance can be inhibited without adding additional filtering plug-ins, so that the node classification precision and the stability of the model in an adversarial environment are improved.

The invention discloses a water turbine coating abrasion detection method and system, and belongs to the technical field of image processing. The method comprises the following steps: firstly, acquiring an original RGB image of a water turbine surface coating, extracting a relative color shift spectrum, a color disturbance gradient and a local texture disturbance index for each pixel point, and constructing a coupling tensor of the pixel point; the feature values of the coupling tensor are solved, the maximum feature value is screened to serve as the wear intensity of the corresponding pixel point, and a wear intensity distribution diagram is generated; the scale weight of each pixel point is calculated in the three-scale neighborhood range, and three-scale weight distribution diagrams are obtained; and finally, a multi-scale enhanced fusion network is adopted to process the wear strength distribution diagram and the three-scale weight distribution diagram, and a water turbine coating wear score is output. According to the invention, through collaborative design of multi-dimensional feature coupling and multi-scale weight fusion, refined characterization of the wear state of the coating is realized, and the detection precision is improved.

The present application relates to the field of intelligent unmanned aerial vehicle manufacturing and autonomous control, and discloses a kind of unmanned aerial vehicle position determination method based on dual-mode navigation, comprising: aligning satellite pseudo-range with the motion characteristics of relative inertial measurement unit output to construct data stream;According to the maximum eigenvalue of the difference projection matrix of visible satellite coordinates and the previous position prediction parameter, determine the geometric topology feedforward factor;When the factor breaks through threshold 3.5, the weight value converted by it is used to equal ratio decay initial gain matrix to reconstruct Kalman gain matrix;Solve the output space absolute coordinates and write back to the online correction register of relative inertial measurement unit in reverse, while inputting the space absolute coordinates into the position and heading adjustment unit of unmanned aerial vehicle flight control system in real time, the present application intercepts multipath jump noise in the prophase stage, blocks the penetration of noise to attitude estimation loop, eliminates the control overshoot induced by state divergence, and guarantees the flight path keeping accuracy.

This invention provides a method and apparatus for identifying wheat diseases, applied in the field of crop disease identification technology. The method includes: acquiring a training sample set; determining the mass-to-charge ratio (M / C ratio) signal intensity data of volatile organic compounds in wheat leaf samples from the training sample set, and determining the maximum eigenvalue of each M / C ratio channel in the M / C ratio signal intensity data; mapping the maximum eigenvalues ​​of all M / C ratio channels to pixels to obtain a two-dimensional pixel matrix of wheat leaf samples; extracting the parent ion and its corresponding fragment ions from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixels in the two-dimensional pixel matrix to obtain a reconstructed feature map; inputting the reconstructed feature map into a wheat disease identification model for network training, wherein the wheat disease identification model is used to identify the disease type and severity of wheat based on the M / C ratio signal features in the reconstructed feature map.

The invention relates to the technical field of quantum information and data processing, in particular to a practical k-entanglement measurement method and numerical calculation method independent of convex top continuation. The entanglement metric calculation method is based on a k-entanglement witness method and a power iterative numerical algorithm, and the core of the method is to estimate the maximum eigenvalue of a matrix and the eigenvector corresponding to the maximum eigenvalue through an iterative numerical method. According to the method, the maximum separable feature value and the corresponding separable feature vector are solved through power iteration, and then the calculation process is further optimized in combination with forward iteration and backward iteration, so that the accuracy and stability of the result are ensured. The whole solving process is intelligent and efficient, and a database of the maximum separable characteristic value and the corresponding quantum state can be established. The database is a basis for constructing a corresponding k-entanglement witness. According to the method, the calculation process is greatly optimized, the feasibility and reliability of numerical calculation are remarkably improved, and a practical, efficient and accurate entanglement measurement tool development method is provided for users.

This invention discloses a method and system for detecting wear on a turbine coating, belonging to the field of image processing technology. The invention first acquires the original RGB image of the turbine surface coating, extracts the relative color shift spectrum, color perturbation gradient, and local texture perturbation index for each pixel, and constructs a coupling tensor for each pixel. By solving the eigenvalues ​​of the coupling tensor and selecting the largest eigenvalue as the wear intensity of the corresponding pixel, a wear intensity distribution map is generated. Subsequently, the scale weight of each pixel is calculated within three scale neighborhoods, resulting in three scale weight distribution maps. Finally, a multi-scale enhanced fusion network is used to process the wear intensity distribution map and the three scale weight distribution maps to output a wear score for the turbine coating. This invention, through the collaborative design of multi-dimensional feature coupling and multi-scale weight fusion, achieves a refined characterization of the coating wear state and improves detection accuracy.

The present application belongs to the technical field of image recognition, and particularly relates to a diamond quality detection method and system based on image recognition, comprising the following steps: obtaining a diamond image to be detected, calculating the structure tensor in the local neighborhood of each pixel of the diamond image, constructing an initial structure guide map according to the eigenvalue distribution, and recording the difference between the maximum eigenvalue and the minimum eigenvalue of the structure tensor as an anisotropy degree map; performing multi-scale shear wave transformation on the diamond image, extracting high-frequency subband coefficients at each scale to calculate local energy, and fusing the local energy at each scale to obtain a multi-scale edge saliency map; and fusing the initial structure guide map and the multi-scale edge saliency map to generate a composite guide map. The present application avoids the edge blurring phenomenon caused by traditional filtering, and improves the completeness of diamond surface defect extraction and the accuracy of overall quality detection.

Certain aspects of the present disclosure provide a method for providing stacked multi-component seismic while drilling (SWD) seismic waveforms. The method includes obtaining multi-component SWD seismic pressure and shear waveforms associated with a planned well. The method includes pre-processing, at the at least one processor, the multi-component seismic SWD waveforms to generate an ordered subset of filtered SWD seismic waveforms and determining, based on a maximum eigenvalue associated with first and maximum correlation matrices of the pressure and shear waveform components, a number of the subset of filtered SWD seismic waveforms to stack. The method includes estimating a first arrival time (FAT) and sending the stacked number of filtered SWD seismic waveforms to a surface equipment, centered around the estimated FAT.

The invention discloses an army officer and soldier mental health classification method in the technical field of mental health, and the method comprises the following specific steps: 1, obtaining psychological values A of a plurality of samples and a plurality of sample feature value matrixes I, the psychological values A being 0 or 1; step 2, storing a plurality of psychological values A and a plurality of sample characteristic value matrixes I in a table form; step 3, carrying out standardization processing on the sample characteristic value matrix I after the label listing, extracting principal component characteristics based on principal component analysis, and then selecting characteristic vectors corresponding to the first k maximum characteristic values to form a transformation matrix; and step 4, performing model training for the dichotomy task, finally predicting principal component features of the test set, and outputting a probability value or a category label of the mental health state. According to the method, through combination of principal component analysis and machine learning, high-dimensional feature dimension reduction and automatic risk prediction are realized, and the method is suitable for mental health classification in army officer and soldier training, resident training and task deployment scenes.

This invention provides an intelligent data sharing method and system for food inspection and testing, relating to the field of data processing technology. The method includes: Step 2, based on a standardized inspection dataset, using a time-series feature learning model to simulate data change patterns under different conditions, constructing a covariance matrix of the inspection index data, calculating the eigenvalues ​​of the covariance matrix, sorting them in descending order by eigenvalue size, extracting the eigenvectors corresponding to the top few largest eigenvalues, linearly combining the original data and eigenvectors, calculating the projection values ​​of each sample along the principal component direction, forming a key feature matrix reflecting the main change patterns of the data, and simulating and expanding the multi-dimensional correlation features of the inspection indicators by combining linear transformation and spatial mapping to generate an extended inspection scenario dataset. This invention improves the efficiency and consistency of food inspection and testing data in standardized integration and secure sharing.

The application relates to a source number estimation method based on orthogonal matching pursuit and signal subspace matching, and relates to a source number estimation method. The application aims to solve the problem that in an underwater acoustic environment, the signal-to-noise ratio of array snapshot data is usually not high, and is easily disturbed by colored noise and coherent sources, and for a high-speed moving target, the effective snapshot number is small, thus leading to low accuracy of traditional source number estimation. The application uses an orthogonal matching pursuit method to iteratively solve a support vector, and uses a characteristic vector corresponding to a maximum characteristic value of a residual to construct two subspaces, and realizes source number estimation through a signal subspace matching criterion. Simulation results show that compared with existing source number estimation methods, the application method has a lower requirement for a signal-to-noise ratio, performs better under small snapshot conditions, and is not sensitive to coherent sources and colored noise. Lake test data processing results show that the method can effectively estimate the number of fixed and moving underwater acoustic targets.

The application discloses a neural network-based LMS power inversion adaptive step size adjustment method, and relates to the technical fields of satellite navigation and signal processing. By constructing and training a neural network model, the input is the current frame input signal power and the maximum eigenvalue, and the output is the current frame predicted step size. Compared with the fixed step size LMS power inversion algorithm, the application introduces a neural network to dynamically predict the optimal step size, solves the problem of non-adaptive step size adjustment in the traditional experience formula, and combines the covariance matrix and the maximum eigenvalue of the input signal to reflect the interference intensity, realize interference sensing adjustment, increase the step size in the case of strong interference, accelerate convergence and enhance tracking ability, reduce the step size in the case of weak interference, reduce the steady-state imbalance, and improve the anti-interference performance. The application can solve the problem that the step size parameter in the existing LMS power inversion algorithm is difficult to adaptively adjust, especially the slow convergence or system divergence problem caused by improper step size setting in a multi-interference channel or a dynamic environment.