1480results about How to "Solve the noise" patented technology

The invention discloses a self-adaptive tracking loop, which comprises an unscented Kalman filter (UKF), an observation noise variance matrix detection module, a fuzzy inference system, an unscented transformation (UT) scale factor regulation module, a state compensator, a carrier wave numerical controlled oscillator (NCO), scale factors, a code NCO, an integration and zero-clearing module, a code loop phase discriminator and a second order code loop filter, and additionally discloses an implementation method for the self-adaptive tracking loop. The implementation method comprises a step 1 ofsignal correlation, integration and zero clearing; a step 2 of code phase tracking; a step 3 of UKF modeling; a step 4 of observation noise variance matrix estimation; a step 5 of process noise variance matrix estimation; a step 6 of UT scale factor regulation; a step 7 of state estimation deviation compensation; and a step 8 of assistance of the carrier wave NCO in the code NCO. According to theself-adaptive tracking loop, the UKF, the observation noise variance matrix detection module and the fuzzy inference system are designed in the carrier tracking loop, so not only can a contradiction between thermal noise vibration in the tracking loop and a dynamic stress error be solved, but a process noise variance matrix and an observation noise variance matrix can be regulated in a self-adaptive manner according to changes of the external environment, and thereby the self-adaptive ability of the tracking loop under complex changeable environments of high dynamic, strong interference, and the like is effectively improved.

The invention discloses an improved self-adaptive histogram equalization method. By finding a dividing gray scale which divides the total number of pixels of each sub-image block into a proximate half, and counting the the total number of the pixels in both high light areas and low light areas of the images, the adjusting parameters of the high light areas and low light areas of the images are calculated, and sub-image blocks are divided into high light areas and low light areas and then respectively subjected to histogram equalization, thereby effectively solving the problems of noise diffusion and texture merging.

The invention provides a micro-electromechanical system (MEMS) triaxial accelerometer and a manufacturing method thereof. The MEMS triaxial accelerometer comprises a sensitive device layer, an upper cover board layer and a lower supporting body layer, wherein clearances are reserved between the sensitive device layer and the upper cover board layer as well as between the sensitive device layer and the lower supporting body layer; the sensitive device layer comprises a supporting frame body, an elastic beam, three independent sensitive mass blocks, movable comb teeth, fixed comb teeth and an electrode; the three independent sensitive mass blocks in the sensitive device layer are used for detecting acceleration signals of three axes X, Y and Z respectively; an acceleration sensor in each direction is hung in the supporting frame body through the corresponding sensitive mass block by the elastic beam which is only sensitive to the detection direction; a plurality of pairs of movable comb teeth are formed on each sensitive mass block by using a body silicon processing technology; a plurality of pairs of fixed comb teeth are correspondingly formed on the supporting frame body to form a pair of differential capacitors serving as sensitive capacitors; and the differential comb tooth capacitors in different directions generate a differential capacitance change in response to the acceleration signals in the corresponding directions.

The invention relates to a bone-conduction headphone which comprises an upper cover (1) and a lower cover (2), wherein the upper cover (1) and the lower cover (2) can form a cavity (4) for containing a bone-conduction loudspeaker assembly (3) after closed; the bone-conduction loudspeaker assembly (3) is arranged in the cavity (4); the lower cover (2) is formed by one-step injection molding and is provided with a vibration plane (21); and the bone-conduction loudspeaker assembly (3) comprises a magnet (31), a coil (32) and a buffering ring (33), wherein the magnet (31) is fastened on the inner wall (211) of the vibration plane (21) of the lower cover (2), the buffering ring (33) is sheathed on the magnet (31) and is provided with the coil (32). The invention has the advantages of almost no noise, deep bass, simple structure and low cost.

The present invention relates to a control circuit and a control method of a synchronous rectifier. In a preferred embodiment according to the invention, the synchronous rectifier control circuit comprises the following components: (1) a first detection circuit which is used for detecting a voltage between a first power terminal and a second power terminal before the device is switched on and generating a first control signal for regulating the switching-on time of the synchronous rectifier; (2) a second detection circuit which is used for detecting a voltage between the first power terminal and the second power terminal after the device is switched off and generating a second control signal for regulating the switching-off time of the device; and (3) a driving control circuit which receives the first control signal and the second control signal and generating a grid control signal for driving a control terminal of the synchronous rectifier.

The invention discloses a non-contact type gesture recognition system and method based on a continuous-wave micro-Doppler radar.A transmitting chain comprises a baseband signal generation module, a transmitter and a transmitting antenna; a receiving chain comprises a signal collection and processing module, three receivers and receiving antennae connected with the receivers respectively, and the signal collection and processing module is connected with an upper computer and is in clock synchronization with the baseband signal generation module; patch antennas are used as the antennae which face the direction of a human hand in parallel, and the three receiving antennae are different in locating place and are not collinear; a transmitted signal is reflected by an object and then received by the receiving antennae, a motion trajectory of the whole human hand in a three-dimensional space is obtained through data resolving of different received and collected signals, and movements of fingers are obtained through characteristic matching recognition of the collected signals.According to the non-contact type gesture recognition system and method based on the continuous-wave micro-Doppler radar, locating, tracking and gesture recognition of the human hand in the three-dimensional space are achieved in a non-contact mode, the advantages of being high in anti-interference capacity, simple in architecture and low in cost are achieved, data processing is convenient, and computing resources are saved.

The invention discloses a collaborative filtering method based on socialized labels, which comprises the following steps that: (1) first, a tripartite graph is used to build models for three different node, i.e. a user, an article and the socialized label, and random walk algorithm is applied to each user to recommend top-N personalized articles; (2) in order to solve the problem of the sparsity of socialized labels (i.e. articles are always labeled by only a small number of labels) and noise brought by the subjective factors of the user, the invention provides a method for expanding the labeling of the articles with a lasso logistic regression model, i.e. labels related to the semantics are increased for each article, and the labels with noise are removed; and (3) the weight of the labels in the recommendation process is regulated. The collaborative filtering method based on socialized labels organically integrates the semantic information of the socialized labels to the description of the articles, uses the lasso logistic regression model to expand the labels, solves the sparsity and noise problems of the socialized labels so as to greatly improve the precision and the performance of a personalized recommendation system.

This invention is about the directly repair shape method, belongs to the manufacturing process of the gear wheel. This can make out the fixed-shape gear wheel straightly by numerical control machine, and no need to fix any more. The process is that: ensuring the fixed-magnitude, the max fix-length of the tooth profile and the longitudinal. It uses the three-dimensional shaping CAD software to build the four theoretical involutes. It has the tip relief by using conjugate gear three-dimensional model. It transforms the fixed-shape gear three-dimensional model into numerical code inputting, and then the numerical machine can directly fix the blank. This invention is easy to carry out, and can reduce the noise and shake, and at the same time it can reduce the edge-effect.

The invention discloses a fuel battery-based emergency power generation vehicle. The vehicle comprises a vehicle body and is characterized in that: a hydrogen supply module, a fuel battery system, a high-power boost DC/DC converter, a DC/AC inverter and a distribution box are arranged on the vehicle body and electrically connected in turn; a secondary battery pack is connected between the high-power boost DC/DC converter and the DC/AC inverter through a low-power bidirectional DC/DC converter which is connected in parallel with the high-power boost DC/DC converter and the DC/AC inverter; a start-stop switch module, a 24V DC/DC converter and the input end of a 12V DC/DC converter are also connected in parallel between the low-power bidirectional DC/DC converter and the secondary battery pack; the start-stop switch module is connected with a control panel; the 24V DC/DC converter and the output end of the 12V DC/DC converter are connected with a monitoring module respectively; and the monitoring module is also connected with the hydrogen supply module, the fuel battery system, the high-power boost DC/DC converter and the start-stop switch module respectively. The fuel battery-based emergency power generation vehicle has the advantages of simple structure, convenient use and capacity of effectively solving the problems of high noise, easy environmental pollution and the like in the prior art by using hydrogen energy sources.

The invention discloses a speed regulation method for a fan. Wherein, acquiring the lowest and highest rotation speed of fan as well as opposite chip temperature to determine the variation relation between speed and temperature; when need, obtaining current chip temperature to determine fan rotation speed according to said relation. It also discloses a fan combination, which comprises a single board with chip, a heat sink on chip, a fan fixed on said sink, and a temperature sensor on said board. This invention not only solves heat dissipation problem of board well, but also overcomes the noise and dust difficulty.

The embodiment of the invention discloses a voice noise reduction method and device, a storage medium, and a mobile terminal. The method includes the steps: receiving a voice signal collected by a single microphone of an earphone, extracting an initial amplitude spectrum and a phase spectrum of the voice signal, and transmitting the initial amplitude spectrum to a pre-trained noise reduction modelto obtain a first noise reduction amplitude spectrum, masking the first noise reduction amplitude spectrum, generating a second noise reduction amplitude spectrum, and recombining the second noise reduction amplitude spectrum and the phase spectrum to generate a first noise reduction speech. Through the above scheme, the method maintains the phase spectrum of the voice signal, and performs the noise reduction of the amplitude spectrum of the voice signal based on the noise reduction model, performs the masking of the processed first noise reduction amplitude spectrum, eliminates the distortion in the noise reduction process, solves a problem that the noise of the voice signal collected in a noisy environment is big, and achieves the quick and precise noise reduction of the voice signal collected by the earphone.

The invention discloses a convolutional neural network feature extraction method based on principal component analysis. The convolutional neural network feature extraction method based on principal component analysis includes the steps: selecting the convolutional neural network which has been trained on an Imagenet data set, taking the convolutional neural network as the feature extractor of an image, extracting the feature mapping graph from the output of each pooling layer of the convolutional neural network, taking all the extracted feature mapping graphs of each layer as the deep featuresof the image, utilizing principal component analysis to perform dimensionality reduction on the deep features, and utilizing bilinear interpolation to reset the final result feature mapping graph tothe original image size to obtain the efficient image deep features. The deep features obtained through convolutional neural network feature extraction method based on principal component analysis have rich semantic information of the image, are low in the feature dimension and small in data size, and can be used for various tasks of identification and classification of images.

The invention discloses a compressor high-frequency harmonic torque compensation method, a compressor controller and an air conditioner. The method comprises the steps of acquiring a vibration amplitude of the compressor in operation; when the vibration amplitude of the compressor is higher than a first preset vibration value, acquiring a load torque harmonic wave and an output electromagnetic torque harmonic wave of the compressor in the current operation working condition, determining a harmonic compensation order according to the load torque harmonic wave and the output electromagnetic torque harmonic wave; determining a harmonic frequency which corresponds with the harmonic compensation order, stacking a sine electromagnetic torque with a corresponding frequency into an output electromagnetic torque according to the harmonic frequency, adjusting a harmonic phase which corresponds with the sine electromagnetic torque according to a multiple-iteration method; and when a fact that the harmonic amplitude which corresponds with the output electromagnetic torque with the stacked sine electromagnetic torque reaches a highest value, recording the harmonic wave which corresponds with the current sine electromagnetic torque as an optimal compensation phase, and performing compensation on the high-frequency harmonic torque of the compressor according to an optimal compensation phase. The compressor high-frequency harmonic torque compensation method, the compressor controller and the air conditioner realize high-frequency harmonic torque compensation of the compressor.

The present invention provides a static measuring method of basic electric parameter of three-phase permanent magnet synchronous motor. In a state that a rotating shaft is fixed, the rated current value of machine to be tested and 150% of the rated current value are respectively input. The corresponding voltage value of V100and V150are recorded. The quotient obtained through dividing the voltage difference to the current difference is the resistance value of motor. The electric period is divided into six voltage vectors in the vector control state. Each electric vector is executed with a voltage circulation. One quarter of V150% is output with step mode. V150% is output when the current is stable. Then a stable current is waited. The steps are repeated for four cycles. Then voltage vector is alternated and the voltage circulation is repeated until all the voltage vectors are completed. Finally the rising times of each vector are compared and converted to an inductance. The greatest value is defined to q-axis inductance. The least value is defined to d-axis inductance.

The invention discloses a mixed induction type active electric power filter system. The mixed induction type active electric power filter system comprises a rectifier transformer, a filtering branch, a rectifier and the other filtering branch, wherein the other filtering branch comprises three passive filters and a voltage source inverter, the passive filters are connected to the voltage source inverter, the primary side of the rectifier transformer is connected to a three-phase power grid through connecting impedor, an epitaxy winding of the secondary side of the rectifier transformer is connected with the rectifier, and the passive filters are connected to three junction points between a delta winding and the epitaxy winding of the secondary side of the rectifier transformer in parallel respectively. According to the mixed induction type active electric power filter system, influence of frequency fluctuation of the power grid side and the high-capacity nonlinear load side on electric power filtering performance is eliminated; meanwhile, the problems that serious harmonic pollution, high additional loss of a master device, high vibration and noise, low system power factor and low system operation efficiency are caused by harmonic waves and idle work to an industrial power distribution and consumption system are effectively solved on the premise that electric power quality on the power grid side meets the national standard of China.

The invention discloses a hinged railway vehicle bogie and a railway vehicle. The hinged railway vehicle bogie comprises a non-integral framework which is formed mainly by hinging a first portion to a second portion. A first wheel axle is installed on the first portion of the framework, a second wheel axle is installed on the second portion of the framework, and the first wheel axle and the second wheel axle form an included angle when the railway vehicle passes a curve. According to the hinged railway vehicle bogie and the railway vehicle, the accessible curve radius is small, the eccentric wear of a walking wheel is less, and the maintenance cost is low.

The invention provides a double-column rotary corrosive-wear testing machine, comprising a testing stand, which is provided with two columns, wherein a crossbeam is arranged on the columns and a through hole is formed in the center of the crossbeam; a crossbeam elevating and driving mechanism, which is arranged on the test stand and connected with the crossbeam; a testing drum, which is arranged on the center of the test stand, wherein an annular temperature control/cooling pipeline is arranged in the lower part of the testing drum; a transmission shaft, which passes through and is arranged in the through hole in the crossbeam; a sample clamping disc, which is fixed at the lower end of the transmission shaft, wherein the upper end surface of the sample clamping disc is provided with an annular groove, and the outer sides of the annular groove are provided with installation through holes; working electrodes, i.e., to-be-tested samples, which are inserted into the installation through holes; an auxiliary electrode, i.e., an annular platinum wire, which is arranged in the annular groove of the sample clamping disc; reference electrodes, which are uniformly distributed on the sample clamping disc at positions corresponding to the working electrodes; and a dynamic connection pool, which comprises annular dovetail cups, wherein the annular dovetail cups sleeve the upper part of the transmission shaft, leading-out circular rings are arranged at the bottom of the inner sides of the annular dovetail cups, the outer sides of the leading-out circular rings are provided with wiring copper columns penetrating the bottoms of the cups, the interior of each annular dovetail cup is provided with an electric needle, the lower ends of the electric needles are certain distance away from the leading-out circular rings, and the working electrodes, the auxiliary electrode and the reference electrodes are connected with the dynamic connection pool through leads.

A model-based distortion compensating noise reduction apparatus for speech recognition, includes: a speech absence probability calculator for calculating the probability distribution for absence and existence of a speech using the sound absence and existence information for the frames; a noise estimation updater for estimating a more accurate noise component by updating the variance of the clean speech and noise for each frame; and a speech absence probability-based noise filter for outputting a first clean speech through the speech absence probability transmitted from the speech absence probability calculator and a first noise filter. Further, the model-based distortion compensating noise reduction apparatus includes a post probability calculator for calculating post probabilities for mixtures using a GMM containing a clean speech in the first clean speech; and a final filter designer for forming a second noise filter and outputting an improved final clean speech signal using the second noise filter.

The invention discloses an air public transportation method which comprises a plurality of air tower stations and supporting tower frames arranged among the air tower stations. Cables are connected between the air tower stations and the supporting tower frames; all the air tower stations form an air public transportation network; a plurality of air public transportation vehicles are hung on the cables; each air tower station comprises a tower base, a plurality of layers of boarding areas arranged at the upper part of the tower base, a waiting area arranged at the lower end of the tower base and electric lifts vertically operating between the lower end of the tower base and the boarding areas. Based on a unique and novel concept, the invention solves the problem of traffic congestion caused by excessive vehicles on roads at the height of low altitude, is safe to use and environment friendly, suitable for generalization in medium-sized and small cities, has less investment and can thoroughly solve the problems of traffic jam, traffic accidents, environmental pollution, noise pollution and the like in medium-sized and small cities.

The invention relates to a fan control system, which is used for detecting temperature and power of at least a work component to control the rotating speed of at least a fan used for cooling the work component. The fan control system comprises at least a work component temperature detecting unit and a work component power detecting unit to detect the temperature and the power of the work component to respectively obtain the temperature of the work component and the power of the work component. The fan control system also comprises a fan control circuit, and the fan control circuit resets lower limit reference temperature, upper limit reference temperature and reference power in advance, wherein, the upper limit reference temperature is higher than the lower limit reference temperature. After being compared with the temperature of the work component and the power of the work component, the invention outputs a control signal to control the rotating speed of the fan. When the fan control system of the invention is adopted, the effects of stabilizing rotate speed of the fan and reducing instant activation noise and protecting the work component can be received.

A novel means of provided a hybrid flexure mounted moving mirror component in an interferometer is introduced herein. In particular, a linear bearing in combination with a novel flexure mounting having novel tilt and velocity control of the moving optical component is provided. Such an arrangement enables correction of the errors at the mirror itself while also solving the problem of isolating vibration and noise caused by the imperfections in the bearing surfaces used in many conventional interferometers. Using such a coupled flexure mounting of the present invention, in addition to the above benefits, also enhances velocity control because the resultant low mass of the moving mirror assembly enables the systems disclosed herein to respond faster than conventional mirror velocity controlled interferometer instruments and with a lower velocity error so as to provide a more stable and lower noise spectra from the analytical instrument.

The invention discloses a system and a method for instantly releasing boiling water from a water dispenser. The system comprises a shell, a water tank, and a heater, a gate valve, a water pump, connecting pipelines, a temperature sensor and a control circuit which are arranged in the shell. The system is also provided with a flow stabilizing device, wherein one water outlet of the flow stabilizing device is connected with a water inlet of the heater through the pipeline; the other water outlet of the flow stabilizing device is connected with a water inlet of the water tank through a three-way device; a water inlet of the flow stabilizing device is connected with a water outlet of the water tank through the pipeline and the water pump; a water vapor outlet of the heater is connected with the water inlet of the water tank through the pipeline and the three-way device; and one end of a water outlet pipeline of the heater is arranged in the heater, and the other end of the water outlet pipeline of the heater extends out of the heater and then is connected with the gate valve. The system adopts the method of arranging the flow stabilizing device, a cheap centrifugal pump and a water supply system preheating water in the water tank to stably supply water, is insusceptible to environment, can enable the boiling water to be instantly released, and has the advantages of simple structure, instantaneity and energy saving.