15 results about "Fastica algorithm" patented technology

The invention belongs to the technical field of fault detection, and particularly relates to a wire drawing machine fault detection method based on the Internet of Things. Historical vibration and temperature and real-time data of operation of the wire drawing machine are collected through an Internet of Things sensor and are transmitted to a server through an Internet of Things protocol. After data are subjected to preprocessing denoising such as variational mode decomposition and a FastICA algorithm, a vibration anomaly judgment index (including effective anomaly time period judgment, vibration mode matching degree calculation and index synthesis) and a temperature anomaly judgment index (including correlation index calculation, heat conduction model establishment and index synthesis) are calculated, and then the two indexes are fused to obtain an anomaly index. A prediction model is constructed by using a neural network algorithm, after optimization of a genetic algorithm, a fault probability is output in combination with real-time data, and finally, a probability value and an abnormal index are normalized and averaged to obtain a fault score, so that fault detection and evaluation are realized. The method improves the fault detection accuracy and the equipment operation reliability, and reduces the maintenance cost.

The invention discloses a voiceprint feature extraction and positioning method for a pipeline leakage point, and the method specifically comprises the following steps: S1, a voiceprint feature extraction stage: collecting a pipeline sound signal through a distributed sensor, extracting and optimizing a multi-domain voiceprint feature after adaptive denoising, and relates to the technical field of pipeline leakage detection. According to the voiceprint feature extraction and positioning method for the pipeline leakage points, multiple leakage point signals of the same pipeline section can be effectively separated and positioned by improving a FastICA algorithm and a density peak value clustering algorithm, the positioning error can be controlled within 0.8 m, the positioning precision in a multi-leakage point scene is remarkably improved, and the positioning accuracy is improved. Based on an edge node-cooperative gateway-cloud center information physical system architecture, a distributed clock synchronization protocol is adopted, cooperative monitoring and simultaneous positioning of multiple branch pipelines are achieved, the response time is shortened to be within 3 seconds, and the problems of positioning conflict and response delay of the multiple branch pipelines in a traditional method are solved.

The application discloses a kind of star-based ADS-B overlap signal time of arrival estimation method and system, signal is received by uniform array antenna, based on FastICA algorithm decomposition obtains two ADS-B signals;Since there is uncertainty in signal decomposition by FastICA algorithm, therefore, two kinds of mixed signals are designed for each time delay, that is, any signal may be ahead of another signal;By cross-correlation coefficient signal processing method, the designed mixed signal is calculated with the original mixed signal;Compare two cross-correlation coefficients, find the larger value, which corresponds to the correct order of each signal in the mixed signal, the maximum cross-correlation coefficient corresponds to the relative time delay of two signals;Based on the relative time delay and the order of the mixed signal before separation obtained, the signal is effectively separated, the signal error rate is low, and the arrival time of ADS-B signal is accurately estimated.

The application discloses an elevator door fault detection method based on image enhancement, and relates to the technical field of image enhancement, and comprises the following steps: collecting running state image data in the running process of an elevator door, and generating an elevator door standard image sequence; performing image enhancement processing, and generating an elevator door enhanced image sequence; constructing a door body mixed visual state matrix; performing independent component separation by using a FastICA algorithm, and generating a door body independent fault source matrix; constructing an improved Hindmarsh Rose neural network, and generating a door body fault evolution state sequence; determining an elevator door fault evolution path, and generating an elevator door fault detection result and an elevator door fault early warning result. By introducing the FastICA algorithm and the improved Hindmarsh Rose neural network, the elevator door fault source decoupling analysis, the fault propagation path modeling, the fault type identification and the fault risk early warning are realized.

The invention relates to the technical field of hovercars, in particular to a hovercar noise testing method and system. The method comprises the following steps: measuring hovercar noise and environmental background noise; the method comprises the following steps: acquiring acceleration direction and motion state data of an aerocar, establishing a mapping relation database of flight attitude and noise features, and inputting the extracted features into a hybrid classification architecture for working condition classification; the method comprises the following steps: carrying out noise separation, converting a time domain signal into a time-frequency domain signal, carrying out initial separation of a noise source by adopting an improved FastICA algorithm, carrying out space optimization by adopting an improved algorithm based on IVA, and carrying out feature refinement and classification based on a CNN-RNN hybrid model; different compensation methods are matched and selected according to time domain and frequency domain analysis results of environment background noise and hovercar noise; and performing spectral analysis and sound source localization on the separated noise source, and verifying a compensation effect. According to the technical scheme, the noise source of the hovercar can be dynamically measured and comprehensively analyzed.

The application discloses a MIMO radar main lobe interference suppression method based on semi-blind source separation and relates to the technical field of radar anti-interference. The application converts a complex convolution mixing problem into a linear blind source separation problem by performing frequency domain conversion and phase space reconstruction on a single channel signal after beam forming, and realizes signal separation by adopting adaptive parameter estimation and a FastICA algorithm, so that main lobe deceptive jamming can be effectively suppressed under the condition of low signal-to-noise ratio and small sample, and the target detection and tracking performance of the MIMO radar in a complex electromagnetic environment is significantly improved.

The invention discloses a radar time domain signal processing method based on CEEMDAN-ICA, and the method comprises the steps: carrying out the decomposition of a received radar time domain signal through employing a CEEMDAN algorithm, obtaining IMF components with frequencies from high to low, calculating the fuzzy entropy coefficient of each order of IMF components, solving the mean value of the IMF components, selecting an IMF component larger than the mean value of fuzzy entropy according to the fuzzy entropy coefficient, determining the IMF component as a noise component layer, and carrying out the recognition of the noise component layer. The radar time domain signals and original signals are simultaneously used as input of a FastICA algorithm, and effective signals in the radar time domain signals are separated. According to the scheme, a modal decomposition method is utilized, priori knowledge and a fixed primary function are not needed, the original signal can be processed only by adjusting parameters, and noise can be effectively suppressed, so that the signal-to-noise ratio of the radar time domain signal is improved, the signal can be adaptively subjected to modal decomposition, and the robustness of the radar time domain signal is improved. Noise can be well suppressed while radar time-domain signal characteristics can be reserved as far as possible, and meanwhile, the accuracy of radar time-domain signal sorting is improved.

The invention relates to the technical field of hovercars, in particular to a composite noise source separation method. Comprising the following steps: performing short-time Fourier transform processing on an acquired noise signal to obtain a time-frequency domain signal; performing preliminary separation by adopting an improved FastICA algorithm, initializing a rotor noise component by utilizing a rotor rotating speed, and initializing an engine noise component by utilizing an engine working condition; spatial optimization is carried out by adopting a frequency domain joint diagonalization algorithm based on independent vector analysis (IVA), sound source position probability distribution is calculated by combining microphone array position information, a TDOA technology and a beam forming technology, and the number of noise sources is adaptively detected; and a CNN-RNN hybrid model is constructed to carry out fine trimming and classification on the separated noise features, the CNN-RNN hybrid model adopts a deep separable convolution and BiLSTM network structure, and flight attitude parameters are used as additional input. According to the technical scheme, the composite noise source of the hovercar can be separated more accurately and effectively.

The invention belongs to the field of magnetic particle imaging, particularly relates to a multicolor magnetic nanoparticle aliasing signal separation method based on adaptive signal processing, and aims to solve the problem that various SPIONs signals are difficult to accurately distinguish in the prior art. The method comprises the following steps: acquiring a to-be-separated initial signal, and preprocessing the initial signal to obtain a target signal; carrying out blind source separation on the target signal by adopting an improved FastICA algorithm to obtain various independent signals; and analyzing the various independent signals to obtain an analysis result, and iteratively adjusting the learning rate and the objective function of the algorithm according to the analysis result until the analysis result meets a preset threshold value. Based on the method, the separation precision of different signals is effectively improved by adaptively adjusting related parameters in the blind source separation process.

The application discloses a kind of electroencephalogram signal separation methods based on genetic algorithm and fast independent component analysis, including the preprocessing of electroencephalogram signal data, using particle swarm algorithm (PSO) to improve genetic algorithm to construct the initial matrix W of FASTICA algorithm, and the signal separation of the noisy electroencephalogram blind source signal is carried out by FASTICA algorithm.The present application makes full use of the fast convergence of genetic algorithm and particle swarm algorithm, combines the advantages of FASTICA for fast separation of blind source signal, carries out signal separation to electroencephalogram signal, solves the complexity and slow convergence speed of previous FASTICA using random initial matrix, further improves the work efficiency of the staff handling signal separation problem, reduces the working hours of staff, effectively extracts noise signal from blind source signal and improves the accuracy of electroencephalogram blind source signal separation.

The present application belongs to the field of magnetic particle imaging, and particularly relates to a multi-color magnetic nanoparticle aliasing signal separation method based on adaptive signal processing, aiming at solving the problem that the prior art cannot accurately distinguish various SPIONs signals. The method of the present application comprises: obtaining an initial signal to be separated for preprocessing to obtain a target signal; using an improved FastICA algorithm to perform blind source separation on the target signal to obtain various independent signals; analyzing the various independent signals to obtain an analysis result, and iteratively adjusting the learning rate and the target function of the algorithm according to the analysis result until the analysis result meets a preset threshold. Based on the method, the separation accuracy of different signals is effectively improved by adaptively adjusting the related parameters of the blind source separation process.

The application discloses a kind of intelligent dual-band microwave vital sign monitoring method and device, to solve the insufficient microwave monitoring of existing anti-interference, multi-target resolution etc..Dog walks according to planning route, rotates head and emits 2.36GHz and 2.45GHz+1310nm laser auxiliary dual-band microwave acquisition signal, when signal is poor, adjust route and head direction, pre-process, FastICA algorithm separates heartbeat and breathing signal;Combined with SVM classification motion state, establish reference signal library and adopt Kalman filtering, GRU network etc. mode optimization signal according to SNR grading;Through joint time-frequency analysis, Hilbert transform is calculated respectively heart rate and respiratory rate, mark monitoring position, can also judge vital sign anomaly and set data transmission priority according to three levels of emergency, warning, normal, realize high-precision, real-time multi-target vital sign monitoring in harsh environment.

The application is particularly a kind of intelligent extraction method and system for urban road collapse hidden danger sensitive elements, relating to the technical field of urban road underground hidden danger detection, comprising: obtaining the template signal corresponding to each sensitive element category in the sensitive element feature library, respectively calculating the waveform similarity and the frequency spectrum matching degree of each independent signal source component and each category template signal.In the application, the blind source separation processing mode of wavelet packet decomposition combined with FastICA algorithm is adopted to decompose the multi-source aliasing echo signal into statistically independent signal source components, solving the technical problem that the feature extraction on the aliasing signal directly leads to feature ambiguity due to the mutual interference of signal sources, making the feature extraction process of sensitive elements have traceability and physical interpretability.

The invention relates to the technical field of satellite communication, and particularly discloses a satellite communication anti-interference method and system based on a kernel method and FastICA, and the method comprises the steps: building a post-nonlinear hybrid model, so as to simulate the nonlinear signal distortion caused by an amplitude limiter in satellite communication, the post-nonlinear mixing model comprises a linear mixing stage and a nonlinear compression stage; a kernel method is adopted to map observation signals to a high-dimensional regeneration kernel Hilbert space, nonlinear features are represented through a quartic polynomial kernel function, and a nonlinear mixing problem is converted into a linear separable form; in a high-dimensional kernel space, based on a FastICA algorithm, a communication signal and an interference signal are separated through a non-Gaussian criterion of maximizing negentropy; in combination with regularization pre-whitening processing and a symmetric fixed point iterative optimization strategy, a separation matrix is updated; and outputting the separated communication signal to realize nonlinear interference suppression.

The application discloses a downlink data anti-interference processing method for a 5G terminal, characterized in that at least comprising the following steps: in the 5G terminal, a multi-antenna is used to receive a downlink PDSCH signal, and FFT processing is performed on each time slot signal of each antenna to obtain a frequency domain multi-path signal; a plurality of subbands are divided in the frequency domain, and each subband signal is spliced in rows according to time slots to form a first matrix; the first matrix is subjected to centralization and whitening preprocessing to obtain a second matrix; in step S13, the second matrix is subjected to initial spatial domain weighting and merging, a complex FastICA algorithm is used to iteratively solve the optimal value of the spatial domain weighting coefficient, and the spatial domain weighting and merging of the second matrix is performed by using the final solution of the iteration to obtain a final anti-interference output signal. The application further discloses a corresponding storage medium and device. By implementing the application, the calculation amount can be reduced, the convergence speed is improved, and the anti-interference effect is very good for both weak interference and strong interference.