105 results about "Noise immunity" patented technology

The invention relates to a microphone array sound source positioning method and system based on cross-correlation-beam forming closed-loop optimization, and belongs to the technical field of sound source positioning. The method comprises the following steps: collecting multichannel sound signals through a microphone array and preprocessing the multichannel sound signals to extract time-frequency features and suppress noise interference; time delay information among the microphones is estimated by adopting a generalized cross-correlation phase transformation algorithm, and an optimization strategy is introduced to improve estimation stability and anti-interference performance; enhancing the target sound source signal in combination with a minimum variance undistorted response beam forming algorithm and an adaptive Kalman filtering mechanism; constructing a closed-loop feedback optimization mechanism based on the beam output signal to realize feedback adjustment; and adopting a hybrid network architecture, taking the beam output signal amplitude spectrum as input, and outputting the frequency spectrum or mask of the obtained target sound source signal. The method has the advantages of high calculation efficiency, high positioning precision and strong anti-interference capability, and is suitable for real-time acoustic signal processing in a complex environment.

The invention discloses an HPLCamp (High Performance Liquid Chromatography) based on deep learning. The invention discloses an HRF dual-mode communication adaptive coding modulation and anti-noise method. The method comprises the following steps: acquiring an optical radio frequency signal amplitude-phase change rate and synchronously sampling and normalizing; calculating a node amplitude-phase residual error to generate a nonlinear mapping coefficient; monitoring coherent change to solve a drift trend, adjusting a modulation coding optimization scheme, compensating distortion and outputting an anti-noise result. According to the method, the instantaneous amplitude and phase of the optical radio frequency dual-mode signal are extracted, a multi-dimensional amplitude-phase characteristic matrix is formed in combination with time domain synchronization and a normalization template, differential residual modeling and nonlinear mapping coefficient calculation are carried out between impedance nodes, and dynamic compensation of amplitude-phase mismatch and envelope offset is achieved. A drift trend quantity is generated based on coherent offset parameter differentiation, feedback is provided for modulation format and coding strategy optimization, amplitude equalization and phase correction are completed, the signal synchronization degree and amplitude-phase consistency are improved, and the steady-state response and anti-disturbance performance of a transmission link are enhanced.

The invention belongs to the technical field of wireless communication and artificial intelligence, and discloses a multi-mode semantic communication method and device oriented to multiple tasks, which avoid low-layer signal coding transmission of original data through extraction and transmission of semantic features, realize end-to-end optimization and channel adaptive robust transmission of semantic level communication, and improve the communication efficiency. The bandwidth occupation is obviously reduced; and the anti-noise performance is improved. Semantic transmission and task execution of multi-source data such as images, texts and voices under a unified framework are achieved through a shared neural network encoder and a channel interface, the robustness of tasks is kept under the complex channel condition, and cross-domain semantic expression among modes such as vision, texts and voices is achieved through the framework. Radio frequency data are sent and received in a real hardware platform, physical combination of an algorithm and a radio frequency link is achieved, and implementation of deep semantic communication in an actual wireless link shows that the method has engineering deployability and expansibility.

The invention discloses a signal control system for a mining locomotive, and aims to solve the problems of weak interference resistance and low safety of a traditional system. The system comprises a signal acquisition module, a dynamic frequency coding processing module, an intelligent decision-making module, an execution module, a communication module and a power supply module. The signal acquisition module acquires the position, speed and environmental parameters of the locomotive through an infrared position, speed and temperature and humidity sensor; the dynamic frequency coding module monitors the signal-to-noise ratio of a sub-band every 10 ms, when the signal-to-noise ratio of the sub-band is lower than 15 dB, the sub-band is automatically switched to a high-signal-to-noise-ratio sub-band larger than or equal to 20 dB, an encryption mapping table is generated for synchronous updating, and the equipment deals with sudden interference through an emergency band. The intelligent decision-making module generates a control instruction, and speed adjustment and parking are achieved through the execution module. An explosion-proof cable and a low-frequency through-the-earth technology are combined for communication, and a power supply is of an explosion-proof type and has multiple protections. The system improves the anti-interference performance and safety of signal transmission, adapts to the mine environment, and supports the intelligent operation of mine transportation.

According to the intelligent frequency spectrum prediction system and method with the dynamic anti-interference function, through the lightweight AI prediction model deployed in the embedded terminal, prospective accurate prediction of the future frequency spectrum hole probability is achieved, and the time delay defect of traditional passive detection is overcome; through an interference identification and classification module, fine identification of an interference source is realized, and a basis is provided for a differential anti-interference strategy; dynamic global optimization of communication parameters is realized through fusion of a utility function decision mechanism of a prediction result, interference information and historical quality and GFSK / QPSK adaptive modulation switching, and optimal balance is obtained between noise immunity and spectrum efficiency; through maintenance of the pre-switching channel pool and a microsecond-level rapid switching mechanism, challenges brought by strong interference burst and high-speed moving scenes are effectively dealt with, and continuity and high reliability of communication are ensured.

The invention relates to the technical field of computers, and discloses a power battery SOH prediction method based on big data. The method comprises the following steps: acquiring voltage, current and temperature time sequence data in a battery charging process; intercepting a constant current charging segment and generating a two-dimensional time sequence spectrogram through continuous wavelet transform; constructing a multi-task attention deep convolutional network, performing decoupling quantization on three attenuation mechanisms of lithium ion precipitation, active substance loss and internal resistance increase, and outputting quantization vectors of the attenuation mechanisms; and finally, calculating an SOH value through a comprehensive evaluation model with physical significance. According to the method, the unification of end-to-end self-learning, strong noise immunity and high interpretability is realized, and the SOH prediction precision and practicability are improved.

The invention discloses an artificial intelligence recognition device based on big data in the technical field of outfield explosion parameter testing equipment, and the device comprises a multi-modal data collection module which is used for collecting explosion infrared light, images, heat, sound and vibration signals; the photoelectric clock synchronization module is used for realizing space-time alignment of multi-modal signals; the noise immunity module is used for performing transient feature enhancement and background noise suppression on the signal; the intelligent signal generation module is used for generating a trigger signal based on multi-modal combined logic and intelligently optimizing a threshold parameter; the intelligent signal triggering module outputs a standard triggering timestamp to the detection instrument; the above modules achieve synchronization through multi-mode fusion, an intelligent learning model and hardware acceleration, and the trigger delay and the false trigger rate are reduced. The intelligent triggering system which is ultrahigh in precision, strong in anti-interference and self-adaptive is achieved, and the strict requirements for reliability, real-time performance and environmental adaptability of a photoelectric parameter identification triggering mechanism in the explosion test and safety monitoring field are met.

The invention relates to the technical field of power quality identification, and discloses a power quality disturbance identification method and device, equipment and a storage medium, and the method comprises the steps: obtaining a power quality disturbance signal; performing feature extraction on the power quality disturbance signal by using a preset one-dimensional convolutional neural network and a one-dimensional residual neural network to obtain a target time feature vector of the power quality disturbance signal; performing feature fusion processing on the target time feature vector by using a preset multi-head attention mechanism to obtain a fused signal feature; and classifying the fused signal features by using a preset classifier to obtain a final classification result used for indicating the disturbance type of the power quality disturbance signal. According to the method, various feature information can be integrated, the influence of noise on a classification result is effectively reduced, the classification performance and generalization ability of a classifier are improved, better performance is achieved in the aspects of classification accuracy and noise immunity, disturbance recognition accuracy is improved, and effective support is provided for electric energy fault diagnosis in an electric power system.

The invention belongs to the field of signal processing, particularly relates to a signal processing method, system and equipment based on an MIMO radar, and aims to solve the problem of limited detection precision caused by insufficient utilization of multi-channel signals and simple fusion strategy in the prior art. The method comprises the following steps: obtaining a target multi-channel echo signal, carrying out distance and angle processing, extracting phase change by using phase unwrapping, carrying out two-stage weighted fusion based on slope entropies of a receiving channel and a transmitting channel in sequence, obtaining a micro-motion signal, carrying out variational mode decomposition and frequency band matching screening, and reconstructing a final output signal. According to the method, the problem of insufficient utilization of multi-channel information is effectively overcome, and the signal noise immunity and the micro-motion feature fidelity are remarkably improved through a double fusion strategy, so that the detection and recognition precision of a weak target is improved.

The invention discloses a mechanical equipment residual life prediction method based on multiple expert models, and belongs to the technical field of mechanical equipment. Comprising the steps of receiving an original mechanical vibration signal, performing preprocessing, constructing a residual life prediction model of the mechanical equipment, training the residual life prediction model of the mechanical equipment based on a preset data set, inputting a mechanical vibration signal to be detected into the trained residual life prediction model of the mechanical equipment, and performing fault category prediction of the mechanical equipment. According to the method, the accuracy, generalization and practicability of residual life prediction of mechanical equipment are remarkably improved, and through multi-scale time-frequency domain feature fusion and a double attention mechanism, the identification capability and noise immunity of the model to a complex fault mode are enhanced; the domain adaptive routing network realizes cross-device efficient knowledge migration, greatly improves the prediction performance on unseen devices, has high precision and high real-time performance, and provides reliable technical support for industrial predictive maintenance.

The invention discloses a model-free predictive current control method based on an improved observer, and belongs to the technical field of permanent magnet synchronous motor control. The control method comprises the following steps: firstly, by analyzing the frequency characteristic of ESO, selecting an extended state observer order for obtaining balance between the anti-interference capability and the noise suppression capability; secondly, designing a model-free predictive current loop controller, and calculating the observer gain through a bandwidth method; then, a judgment mechanism of a current loop controller is constructed, the mechanism judges whether the system is in a steady state or not at the moment according to the current deviation value in the current state, and therefore the observer order with the better control effect is selected. According to the model-free predictive current control method based on the improved observer, the anti-interference performance of the system and the noise suppression capability of the current are remarkably improved, it is guaranteed that the improved observer has rapid tracking response and excellent steady-state performance, and the robustness of the control system is enhanced.

The invention relates to the technical field of voltage regulation cabinet monitoring, and particularly provides a medium and low voltage voltage regulation cabinet intelligent monitoring method based on big data and Internet of Things, and the method comprises the following steps: installing a sensor in voltage regulation cabinet equipment, and collecting key parameters, such as pressure, temperature, flow, current and the like, in the operation process of the equipment in real time; transmitting the data to a cloud platform through a wireless network; and the cloud platform performs deep analysis on the acquired data based on big data storage, processing and analysis technologies, and performs equipment operation health scoring, state evaluation and fault prediction by utilizing innovative design and fusing a vector random forest (SRA) of a support vector machine and a random forest algorithm. According to the algorithm, fine classification capability, noise immunity and feature evaluation are fused, the equipment operation health state can be accurately scored, and faults can be effectively predicted. The innovation point lies in that the advantages of the two algorithms are complemented by using a vector fusion method, efficient feature extraction and decision fusion are realized in a multi-dimensional feature space, and the complex monitoring requirements of big data and the Internet of Things are met. According to the invention, a data acquisition module of a high-precision sensor and a data transmission module of wireless communication are deployed in the gas medium-pressure equipment, and a big data storage and processing technology, a machine learning algorithm and an intelligent early warning mechanism are combined, so that real-time monitoring, fault prediction and maintenance optimization of the operation state of the equipment are realized; therefore, the safety, the reliability and the operation and maintenance efficiency of the gas equipment are improved.

The invention relates to the technical field of audio signal processing. The invention provides a method, system and device for predicting pulmonary nodules through sound and a storage medium. The method comprises the following steps: in a quiet test environment, directionally collecting a sound signal of a patient through an audio sensor; wherein the sound signals comprise breath sounds, cough sounds, vowels and sentences; performing noise reduction processing on the collected sound signals to obtain a noise reduction data set; extracting multi-dimensional voiceprint features from the noise reduction data set, and constructing a sound feature matrix; and performing multi-modal fusion analysis on the sound feature matrix and clinical data of the patient, inputting the sound feature matrix and the clinical data into a sound prediction model optimized by a focus loss function, and outputting a pulmonary nodule risk prediction result and a visual diagnosis report. The problems that the misdiagnosis and missed diagnosis rate is high, the radiation risk is large and the equipment dependence is strong when the pulmonary nodules are examined by the existing iconography, and the characteristic analysis is single, the noise immunity is poor and the screening accuracy is insufficient in the sound detection technology are solved.

A high-noise-immunity and low-delay level shift circuit for a GaN half-bridge drive chip, relating to the technical field of power management in integrated circuits. By additionally providing a path from a floating power rail VDDH to the ground, the level shift circuit simulates the noise effect caused by a sudden change in the voltage of the floating power rail in the level shift circuit. The level shift circuit has a good noise immunity capability and a simple structure.

The application belongs to the field of signal processing, and particularly relates to a signal processing method, system and device based on MIMO radar, aiming to solve the problem of insufficient utilization of multi-channel signals and simple fusion strategy in the prior art, which limits the detection accuracy. The application comprises the following steps: obtaining target multi-channel echo signals and performing distance and angle processing, extracting phase changes by using phase unwrapping, performing two-stage weighted fusion based on the slope entropy of the receiving channel and the transmitting channel in sequence, obtaining micro-motion signals, performing variational mode decomposition and frequency band matching screening on the micro-motion signals, and reconstructing the final output signals. The application effectively overcomes the problem of insufficient utilization of multi-channel information, significantly improves the signal noise immunity and micro-motion feature fidelity through the double fusion strategy, and thus improves the detection and recognition accuracy of weak targets.

This invention belongs to the field of communication technology and discloses a voice communication system based on deep ultraviolet light. The system includes: a voice communication terminal, a transmitting information processing module, an electro-optical conversion module, a transmitting optical system, a receiving optical system, a photoelectric conversion module, and a receiving information processing module. Compared with traditional radio frequency communication, ultraviolet light has excellent anti-interference and anti-interception capabilities; compared with point-to-point communication in the deep infrared band, ultraviolet light has non-line-of-sight characteristics, enabling communication even in jungle warfare. This invention uniquely employs FMCW (Frequency Modulated Continuous Wave) modulation. FMCW encodes bit information as a continuous increase or decrease in frequency. The receiving end obtains the difference frequency signal through mixing, which can be demodulated using matched filtering or Fourier transform, significantly improving processing gain and noise immunity. Even under weak scattered light conditions, signals can be reliably extracted from strong backgrounds, greatly improving communication link margin.

This invention provides a method for constructing and implementing a memristor-based chaotic synchronization system using FPGA circuits, comprising the following steps: Step 1, constructing the memristor chaotic synchronization system; Step 2, discretizing the memristor chaotic synchronization system to obtain discretized equations; Step 3, designing the various modules of the FPGA chaotic signal generator and converting the discretized equations into a hardware design language; Step 4, designing based on a state machine to solve the chaotic sequence; Step 5, simulating and verifying the designed program at the board level, and analyzing the implementation effect. This invention incorporates a memristor into the chaotic synchronization system, giving it more complex dynamic behavior and noise immunity, enabling it to reach synchronization faster and achieve stable error convergence. This solution addresses the shortcomings of current chaotic systems using analog circuits, such as unstable signals and susceptibility to external environmental interference, and can resist temperature changes and electromagnetic interference, generating high-quality chaotic signals.

This application relates to a clock tree synthesis method and system for reducing the vertical coupling capacitance of the clock tree, belonging to the field of clock tree synthesis technology. The core of the method lies in the fact that, during the clock tree synthesis process in the physical design phase, by using a preset vertical design of the clock tree and a coupling-aware clock tree synthesis optimization cost function, the parasitic coupling capacitance between clock lines and adjacent signal lines in the vertical metal layer is systematically reduced, thereby improving the noise immunity of the clock network and reducing delay uncertainty and dynamic power consumption.

The invention discloses a servo motor disturbance suppression method based on variable gain control, and the method comprises the steps: adding a nonlinear branch for disturbance suppression on the basis of a conventional position closed-loop control structure; the nonlinear branch is formed by connecting a phase extractor and a gain regulator in series, the phase extractor extracts effective phase information according to a tracking error signal of the servo system, and the gain regulator outputs a corresponding nonlinear compensation signal according to the phase information. After the compensation signal is superposed to the position loop, the loop gain of the system at a low frequency band can be dynamically improved, so that the disturbance suppression capability is remarkably enhanced. The nonlinear branch is essentially a switching function, but is different from a traditional switching function or a fixed gain control mode, the gain regulator adopts a reset integral mechanism to generate nonlinear gain, and the gain regulator has excellent low-frequency disturbance suppression performance and anti-noise capability. The device has the advantages of simple structure, quick response and the like.

The utility model discloses a power device overcurrent short circuit protection circuit, which is connected between a power device driving signal circuit and a desaturation detection circuit, and comprises a first divider resistor, a second divider resistor, a first amplitude control resistor and a second amplitude control resistor which are electrically connected in sequence, the first divider resistor is connected with a power device driving signal output end of the chip; a branch of a connection point between the first amplitude control resistor and the second divider resistor is connected to a collector of the power device; a connection point between the first amplitude control resistor and the second amplitude control resistor is used as an input end of the desaturation detection circuit; according to the utility model, on the basis of realizing the desaturation detection circuit, the fine tuning effect of blanking time (namely, the anti-noise time delay can be controlled) is realized at the same time, and the driving structure cost is obviously reduced on the premise of ensuring the driving reliability of the power device without depending on the DESAT function pin of the driving chip.

The utility model relates to a broadband anti-burning anti-crosstalk annular low noise amplifier assembly, which comprises an input end, a low noise amplifier assembly and an output end, the input end is connected with a band-pass filter, and the output end is connected with a low-pass filter; the band-pass filter is connected with the low-noise amplifier assembly, and the low-pass filter is connected with the low-noise amplifier assembly through the down-conversion mixer; the low-noise amplifier assembly is of a multi-stage cascade system structure and comprises a front-stage broadband low-noise amplifier, a rear-stage SPDT radio frequency switch, a biasing circuit with temperature and voltage compensation effects and an ESD protection circuit. The front-stage broadband low-noise amplifier is cascaded with the rear-stage SPDT radio frequency switch; the front-stage broadband low-noise amplifier and the rear-stage SPDT radio frequency switch are respectively cascaded with a biasing circuit with temperature and voltage compensation effects; the utility model has the advantages of reasonable structure, strong anti-interference performance, safety and reliability, and can avoid burnout.

The useful solution refers to a pump control circuit in water tank systems or water filtration devices with water tanks, including a power supply block, a logic processing block, and an output block, arranged on a compact printed circuit board. It uses relay elements to switch the load, and also includes a power supply protection block, noise immunity, and processing logic to limit control errors when the water level signal or water request signal is fluctuating or constantly changing.

The invention discloses a signal identification and measurement method based on 5GNR SSB time-frequency feature matching, and aims to solve the problems that in the prior art, a traditional 5GNR module chip is high in complexity and high in cost, and a simple analog detection scheme is prone to noise interference and cannot effectively distinguish LTE and 5G signals. The method comprises the following steps: extracting an energy envelope contour of a digital signal sample, adaptively calculating a detection threshold based on a dynamic range of the energy envelope contour, identifying candidate signal envelopes, and carrying out LTE interference pre-judgment; after the candidate envelopes are paired, verifying an effective envelope group through matching of multi-dimensional time-frequency characteristics including time domain periodicity, frequency domain form, waveform correlation and the like, and executing periodic suspicion rejection to eliminate interference; and finally, calculating the 5GNR signal intensity based on the effective envelope group. According to the invention, high-accuracy identification, high-efficiency measurement and strong anti-interference performance of the 5GNR SSB signal can be realized, and the hardware cost is reduced.

The invention discloses a motor bearing fault diagnosis method, device and system and a medium, and relates to the technical field of rail transit trains, and the method comprises the steps: constructing a rotating speed value sequence of a motor according to a rotating speed pulse signal outputted by a pulse encoder; filtering the rotating speed numerical value sequence to filter components irrelevant to faults in the rotating speed numerical value sequence; a variational mode decomposition algorithm is adopted to decompose the rotating speed numerical value sequence after filtering processing, and an intrinsic mode component is obtained; and performing spectral analysis on the intrinsic mode component to judge whether the motor bearing has a fault or not and determine the fault type. The method has the advantages of being low in cost, good in anti-interference performance, high in self-adaption and high in signal-to-noise ratio.

A noise immunity of a detected capacitance is prevented or inhibited from lowering on a driving electrode different in width from the other driving electrodes, provided in an input device. A touch panel serving as an input device has a plurality of driving electrodes extending in an X-axis direction and arranged in a Y-axis direction intersecting with the X-axis direction, and a driving electrode arranged outside one side of an arrangement of the driving electrodes and extending in the X-axis direction. Further, the touch panel TP1 has a plurality of detecting electrodes extending in the Y-axis direction and arranged in the X-axis direction. The width of the driving electrode is smaller than the widths of the driving electrodes and the detecting electrode includes an expanding portion for expanding the area of the detecting electrode on the side opposite to the plurality of driving electrodes via the driving electrode.

The invention discloses a transformer fault detection method based on double-flow auditory feature fusion and a random forest, and the method comprises the steps: obtaining a real-time voiceprint signal during the operation of a transformer, carrying out the pre-emphasis, framing and windowing of the real-time voiceprint signal, and obtaining a voiceprint frame; extracting Mel frequency cepstrum coefficient characteristics of the voiceprint frame based on a Mel filter bank; extracting Gammatone frequency cepstrum coefficient characteristics of the voiceprint frame based on a Gammatone filter bank; calculating time domain statistical index characteristics of the voiceprint frames; constructing a joint feature, inputting the joint feature into a pre-trained random forest model, and calculating a Gini importance score so as to obtain an anti-noise robust feature subset; and inputting the anti-noise robust feature subset into a random forest classifier, and outputting a fault type identification result of the transformer by using an integrated voting mechanism of a multi-decision tree. The method can solve the problems that in the prior art, feature extraction is single, noise immunity is poor, attention to transient impact is lacked, and a feature dimension reduction and classification method has defects.

This invention belongs to the field of wireless optical communication technology, specifically a sub-Poisson multi-user photon counting communication system and its design method based on the WMSE criterion. The invention includes constructing a multi-user, multi-input, single-output photon counting communication model and introducing a sub-Poisson photon counting reception mechanism; deriving the mean square error expression for each user under sub-Poisson conditions, introducing the reciprocal of the noise variance as a weighting coefficient, and constructing a noise-aware weighted MSE optimization objective; designing a multi-user precoding scheme using a continuous convex approximation method, and minimizing the system's weighted error and optimizing performance by updating the weighting coefficients in the outer layer and optimizing the precoding matrix in the inner layer. This invention can effectively suppress the impact of heteroscedastic noise on system performance under sub-Poisson noise conditions, significantly reduce the system's average bit error rate, and improve the transmission performance and noise immunity of the multi-user photon counting communication system; it can be applied to scenarios such as UAV networks, satellite-to-ground optical communication, and secure optical transmission.

This invention provides a dynamic impedance calibration method for high-frequency magnetic ring inductors based on convolutional neural networks, relating to the fields of radio frequency power electronics, switching power supplies, and electromagnetic compatibility design. The method includes: acquiring amplitude-frequency response, phase shift data, and ambient temperature of a high-frequency magnetic ring within the 0.1MHz–1GHz frequency band; after adaptive filtering, extracting amplitude-frequency timing features and phase shift / phase lag features using a dual-branch convolutional neural network; dynamically fusing these features using a bidirectional attention mechanism to generate a fused feature map; and finally outputting a dynamic calibration coefficient matrix to correct impedance measurements in real time. This invention solves the calibration accuracy degradation problem caused by device parasitic parameters and temperature drift in the high-frequency band (>400MHz) of traditional methods, achieving microsecond-level dynamic response over a wide temperature range. Furthermore, through multi-modal data fusion and noise immunity design, it significantly improves calibration robustness under high-frequency nonlinear conditions.

This invention discloses a multi-element channel coding and decoding system based on iterative feedback, belonging to the field of communication technology. It includes a transmitter and a receiver; the transmitter includes a multi-element LDPC coding module, a serial-to-parallel conversion / mapping module, and a modulator module; the receiver includes a soft demodulator module, a serial-to-parallel conversion / demapping module, and a multi-element LDPC soft decoding module. Based on multi-element LDPC coding, this invention designs a special modulation constellation diagram, such that each 16QAM symbol corresponds to two mutually related GF4 multi-element LDPC coded symbols, and the average Euclidean distance of each symbol is maximized. Simultaneously, it introduces iterative feedback decoding, improving noise immunity and enhancing the performance of the optical communication system.