236 results about "Adaptive filtering algorithm" patented technology

One embodiment of the present invention relates to an adaptive filtering apparatus comprising first and second real valued adaptive filters, respectively configured to receive an adaptive filter input signal based upon a transmission signal in a transmission path. The first real valued adaptive filter is configured to operate a real valued adaptive filter algorithm on the input signal to estimate a first intermodulation noise component (e.g., an in-phase component) in a desired signal and to cancel the estimated noise. The second real valued adaptive filter is configured to operate a real valued adaptive filter algorithm on the input signal to estimate a second intermodulation noise component (e.g., a quadrature phase component) in the desired signal and to cancel the estimated noise. Accordingly, each filter operates a real valued adaptive algorithm to cancel a noise component, thereby removing complex cross terms between the components from the adaptive filtering process.

The present invention provides an FMCW broadband life detection radar breathing and heartbeat signal extraction algorithm. Compared with the prior art, the algorithm has the advantages that breathing and heartbeat signals of multiple persons can be obtained simultaneously, the boundaries and energy of reconstructed breathing signals can be very well kept by using Coiflets wavelets compared with traditional frequency-domain filtering, especially processing conducted on short-time signals is obvious, higher harmonics of the breathing signals are filtered out by adopting a double-coefficient based LMS self-adaptive filtering algorithm, and the self-adaptive filtering algorithm does not need additional reference signals. Due to the fact that a degressive search step size coefficient is adopted in the algorithm, the heartbeat signals can be better retained while breathing signal harmonics are filtered out, the heartbeat signals can be rapidly converged by selecting the appropriate search step size coefficient and adopting the double-coefficient based LMS self-adaptive filtering algorithm, and smaller stable imbalance is kept in longer time, so that finally separated heartbeat signals are more accurate in frequency component within a long time.

The invention belongs to the technical field of wireless communication signal processing and the technical field of full-duplex communication systems and relates to a method for suppressing passive intermodulation interface in a full-duplex communication system, in particular to a technology for suppressing passive intermodulation interface based on adaptive filtering. A simplified model of a passive nonlinear device is established and used for conducting interference suppression according to a digital signal processing method. The self-adaptive filtering algorithm is adopted as the estimation algorithm which is effect and enables hardware implementation to be achieved easily, so that compensation for passive intermodulation interface on intermediate frequency signals is achieved. A simulation result shows that passive intermodulation interface can be effectively suppressed and certain signal-to-ratio gain is achieved. The method is simple, computation complexity is low, resource cost is low, hardware implementation is easily achieved, the method is stable in performance, the system is strong in adaptability, the real-time performance is good, the method can be used for effectively suppressing passive intermodulation interference, and the higher signal to interference ratio gain is achieved.

Vibrations at harmonic frequencies are reduced by injecting harmonic balancing signals into the armature of a linear motor / alternator coupled to a Stirling machine. The vibrations are sensed to provide a signal representing the mechanical vibrations. A harmonic balancing signal is generated for selected harmonics of the operating frequency by processing the sensed vibration signal with adaptive filter algorithms of adaptive filters for each harmonic. Reference inputs for each harmonic are applied to the adaptive filter algorithms at the frequency of the selected harmonic. The harmonic balancing signals for all of the harmonics are summed with a principal control signal. The harmonic balancing signals modify the principal electrical drive voltage and drive the motor / alternator with a drive voltage component in opposition to the vibration at each harmonic.

The invention discloses an amplitude phase error correction system and an amplitude phase error correction method for a transmitting channel of a phased-array antenna. The amplitude phase error correction system comprises a correction signal source, a rotary selector switch, a to-be-corrected transmitting channel, an auxiliary receiving channel and a correction coefficient calculating module. When the amplitude phase error correction is carried out on an n transmitting channel, a correction signal source transmits a correction signal y (k), and the auxiliary receiving channel receives a signal yn (k) at the same time. A correction coefficient calculating module calculates a correction weight signal wn through the signal yn (k) and a reference signal yref (k) in such a way of adopting a minimum mean square self-adaptive filtering algorithm. According to the method and system provided by the invention, the methods engineering can be realized simply, and the correction accuracy is higher, therefore the system and the method can be applied to the field of the amplitude phase error correction of a uniform circular array phased-array antenna.

The invention discloses a method for estimating a lithium battery charge state. The method comprises steps of estimating the charge state in accordance with the system state initial value and the state equation; calculating the residual error of the measured value and the estimated value and then calculating the fading factor; calculating the time-varying fading factor; calculating and obtaining the adjusted value of the fading factor in accordance with the residual error change and the current of the actual station; obtaining the new value of the fading factor, and obtaining the gain matrix; updating the charge estimation state; and estimating and measuring the noise covariance matrix through the self-adapting filtering algorithm. Compared with the prior art, by the aid of the strong tracking filter algorithm and the modification of the strong tracking filter algorithm, the noise and the fading factor can be measured and updated in accordance with actual condition, the estimation accuracy of the lithium battery charge state can be improved effectively compared with the traditional Kalman filter algorithm, and the tracking and the self adaption of the algorithm can be improved.

A method for eliminating interference to blood oxygen saturation monitoring includes steps of digitally filtering acquired red light and infrared light PPG (photoplethysmography) signals, removing low-frequency interference signals in direct current components, computing accurate direct-current component value, computing PPG signal envelope curves to extract alternating-current components, separating signals relevant with interference and serving the same as reference signals to be subjected to self-adaptive filtering, eliminating movement interference generated from aliasing of frequency and target signals in the alternating-current components, computing to obtain the accurate PPG alternating-current component value, and finally computing the blood oxygen saturation after the interference is eliminated according to the Lambert-Beer's law. The method is capable of suppressing movement interference well and computing to obtain the accurate blood oxygen saturation value. By extracting the reference signals directly from the red light and infrared light PPG signals to realize self-adaptive filtering, the method is simple in construction, simple in algorithm and small in computing number, has no need of an extra device for providing reference signals, is free of multiple self-adaptive filtering, and facilitates miniaturization and portability of monitoring equipment.

The invention discloses an active echo self-interference suppression scheme of a wireless broadband multi-antenna MIMO full-duplex system. A wave beam forming technology and self-adaption iteration filtering processing are applied to a full-duplex self-interference suppression method, pre-weighting processing is carried out all paths of emitting signals on a sending side through airspace wave beam forming, and echo self-interference signals are accurately estimated and eliminated on a receiving side through repeated iteration of a self-adaptive filtering algorithm. An FD-MIMO system frame is introduced through an airspace wave beam forming method, the broadband wireless FD-MIMO system echo self-interference suppression scheme combined with an active suppression mechanism is provided, the high self-interference signals of the full-duplex (FD) system are suppressed obviously, meanwhile, the bit error rate (BER) performance of the system is effectively controlled, the channel transmission capacity of the system is comprehensively improved, and the overall performance of the full-duplex system is improved. Under the condition of a fading channel, the bit error rate (BER) and system capacity of the full-duplex system under an existing ideal channel can be achieved, and the scheme can be used for various new-generation broadband wireless full-duplex communication systems.

The invention relates to a wearable human arm tremor detection device and a signal processing method thereof. The device comprises a tremor signal detection system and a tremor signal processing system; wherein, the tremor signal detection system comprises an elbow joint tremor detection system, a wrist joint tremor detection system and a hand tremor detection system; the detection systems adopt independent modular structures; DSP processor units in the detection systems are connected with own sensor modules, wireless communication units and power supply units; the tremor signal processing system comprises a DSP central processing unit and a wireless communication unit as well as a display, which are connected with the DSP central processing unit; the DSP processor unit in the detection system is connected with the DSP central processing unit in the tremor signal processing system by the wireless communication unit. The tremor signal processing method adopts the 'two-stage' adaptive filtering algorithm to obtain the normal movement information and the tremor movement information of the arms of the patients. The device adopts the elbow-wrist-hand type wearing structure and detects the arm tremor information of the tremor patients in real time.

The invention discloses an echo cancellation method, a voice activation method, an echo cancellation device, an audio device and a computer readable storage medium. The method comprises the followingsteps: a first audio signal and a second audio signal are acquired, the first audio signal is a signal input to a loudspeaker, and the second audio signal is a signal picked up by a microphone; estimating an echo signal caused by the first audio signal through a linear adaptive filtering algorithm to obtain an echo estimation signal; extracting a characteristic parameter of the echo estimation signal as a first characteristic; extracting a characteristic parameter of the error signal as a second characteristic; inputting the first feature and the second feature into a pre-trained neural network model, and outputting the gain of the user voice signal by the neural network; and calculating a product of the error signal and the gain to obtain a user voice signal.

The invention discloses a method for controlling the random vibration of an electrohydraulic servo system based on RLS filters, which comprises the following steps: converting an acceleration power spectral density reference signal into an acceleration time domain driving signal; taking an acceleration input signal of the electrohydraulic servo system as that of an RLS filter I; online regulating weights of the RLS filter I and an RLS filter II in real time by using an RLS adaptive filter algorithm and performing identification on a frequency response function of the electrohydraulic servo system and an impedance function of the RLS filter I respectively; and constructing a filter III based on the identification result of the impedance function of the RLS filter II, filtering the acceleration time domain driving signal and taking an output signal of the filtration as the acceleration input signal of the electrohydraulic system. The convergence property of the RLS adaptive filter algorithm adopted by the method guarantees the convergence property of the real-time online iteration of acceleration power spectral density when the characteristics of the electrohydraulic servo system and a test piece are changed.

Adaptive filtering algorithm training signals incorporate adaptation-enhancing signals having properties, such as spectral whiteness and rapidly decaying auto-correlation between samples, which allow the adaptive filtering algorithms to converge more quickly as compared to conventional algorithms. Exemplary adaptation-enhancing signals are incorporated such that the training signals are not perceptibly altered as compared to conventional training signals. For example, in a hands-free telephony application, an adaptation-enhancing signal according to the invention is dynamically tailored so that it is masked by a loudspeaker signal (the conventional training signal) and is thus inaudible to hands-free telephone users.

The invention relates to a self-adaptive three-sectional protection method which is based on three protection means of current quick-break protection, current quick-break protection with time limit, and overcurrent protection. The method comprises the following steps: 1). accurately acquiring voltage and current in real time through an adaptive filtering algorithm so as to obtain a fundamental wave current amplitude i1; 2). respectively performing setting on the three protection means so as to obtain corresponding setting values IDZ.I, IDZ.II, and IDZ.III; 3). when i1> IDZ.I, taking a current quick-break protection measure, otherwise, skipping to step four; 4). after a time period TII, if i1> IDZ.II, taking a measure of the current quick-break protection with time limit, otherwise, skipping to step five; and 5). after a time period TIII, if i1> IDZ.III, taking an overcurrent protection measure, otherwise, ending the three-sectional protection method. The method and device provided by the invention has the following advantages: the reliability and stability of a distribution network are improved, and the protection sensitivity and accuracy are substantially enhanced without changing the basic principle of three-sectional current protection.

The invention discloses a method for smoothing output power of a wind power station by utilizing an energy storage system. In the method, not only wind power is introduced to predict or rectify the output power of the wind power station, but an improved self-adaptive filtering algorithm is applied. Since the output power of the wind power station is rectified by predicting the wind power, the change rate of the output power of the wind power station is decreased, and the output power of the wind power station can meet standards to inject into a power grid. By employing the self-adaptive filtering algorithm, regarding to the input power in different time, time weight coefficient in an adaptive filtering structure can be automatically calculated according to the input, so that the defect that different input cannot correspond to different filtering time constant by utilizing a low pass filter smoothing method in the past can be overcome. After the smoothing method is employed to smooth the output power of the wind power station, the output voltage of the wind power station is kept stable, fluctuation and flicker of the power grid voltage is reduced, harmonic pollution is reduced and accidents are reduced.

The invention discloses an instantaneous frequency measurement method based on the adaptive filtration theory. The adaptive filtering algorithm is utilized to decompose a nonstationary multi-frequency component signal into a plurality of progressive single-frequency signals to be measured, thus solving the analyzing problem of the multi-frequency component signal. The Hilbert transformation is adopted to perform instantaneous frequency measurement to the progressive single-frequency signals, thus complete and accurate instantaneous physical quantity of the nonstationary multi-frequency component signal can be obtained, the time-frequency distribution of the nonstationary signal can be further obtained and the requirements of instantaneous frequency measurement in the fields of military communication and electronic countermeasure can be satisfied.

This invention provides a passive intermodulation interference cancellation method based on sparse fractional order Fourier transformation, and belongs to the technical field of anti-interference. The method comprises the following steps: 1, in a communication system with a transceiving antenna, transmitting a downlink signal and receiving an uplink signal by a duplexer; 2, performing sparse fractional order Fourier transformation on the received uplink signal and estimating parameter information of a power series model; 3, reestablishing a PIM (Passive Intermodulation) interference signal by using the parameter information estimated by the step 2 and the downlink signal; 4, judging whether the amplitude of the PIM interference signal reestablished in the step 3 is greater than a threshold or not, and deciding to jump to step 5 or step 6; step 5, performing PIM interference adaptive filtering by using an adaptive filtering algorithm to acquire an adaptively filtered PIM interference signal; and 6, correcting the uplink signal to complete PIM interference cancellation. Without considering the problem of convergence difficulty of parameter estimation, the method processes LFM (Linear Frequency Modulation) signals by using the sparse fractional order Fourier transformation, and realizes PIM interference suppression with low computation amount.

The invention relates to a gain compensation self-adaptive filtering-based SINS (strap-down inertial navigation) / DVL (Doppler velocity log) combined positioning method and belongs to the technical field of high-precision SINS / DVL combined positioning. The invention aims to solve the problem of low positioning precision due to the influence of the insufficient flexibility of the filtering algorithmof conventional SINS / DVL combined navigation. The method of the invention includes the following steps that: corresponding state initial values and observation values are acquired based on the sensorinformation of a strap-down inertial navigation system and a Doppler velocity log; a corresponding system equation and observation equation based on a combined navigation error model are constructed,a gain compensation self-adaptive filtering algorithm is adopted to correct errors, so that the post-correction speed and position error information of a target is obtained; and finally, the obtainederror information and the observation information of the strap-down inertial navigation system and the Doppler velocity log are fused, so that a high-precision positioning result is obtained.

A method for filtering power frequency digit by frequency-tracking includes switching power frequency reference signal into signal input end of phase locked loop, outputting eight frequency multiplier signal of power frequency by said loop when eight frequency division signal of power frequency outputted by central processor is switched into comparison input end of said loop, using said frequency multiplier signal as sampling signal sequence, using comb filtering algorithm to realize trop of power frequency and its frequency multiplier, finally controlling and outputting filtered signal by central processor.

The invention discloses a testing system for on-line self adaptive harmonic currents, which is characterized in that the system is composed of a signal preprocessing part, a referencing orthogonal signal producing part, a self-adaptive filtering part of computer software, and a computer analyzing and displaying part. The superiorities of the invention lie in that delays and phase shifts which are produced in signals and can't be compensated can be compensated by the system, the dynamic responsing characteristics and the testing accuracy of the testing system are increased, an improved type self-adaptive filtering algorism is presented, which overcomes fixed limitations which are existed in traditional self-adaptive testing methods. active power and reactive power of base waves can be tested out through the system, which is convenient for APF reactive power current compensation, a method for producing referencing voltage through phase-locked loop is adopted, which decreases sampling interference of voltage signals to the lowest level and produces sine wave type referencing voltage, and a method for connecting a low-pass filter in front of an integrator is adopted, which increases the testing accuracy and the dynamic response characteristics of the testing system. The system also applies for testing each component of the voltage.

The invention provides a self-adaptive filtering method for data of a pipeline magnetic flux leakage inner detector, and relates to the technical field of fault diagnosis techniques of pipelines. The method comprises the following steps that magnetic flux leakage signals are acquired by the pipeline inner detector, are amplified and stored, and are transformed into the actual digital voltage signals, the digital voltage signals are subjected to baseline correction, and then the abnormal data are searched, eliminated and compensated; the single-channel signal classification are conducted, and different types of signals are subjected to line filtering through corresponding single-channel self-adaptive filtering algorithms; the influence degree of noises between the adjacent channels is compared and analyzed to implement multi-channel equalization on the magnetic flux leakage signals; and the multi-channel signal area classification is conducted, and the different signal areas are subjected to surface filtering through corresponding multi-channel self-adaptive filtering algorithms. According to the method provided by the invention, the predication of the service life of the pipeline is more accurate, the feature is much easier to extract in the fault diagnosis of the pipeline, the filtering effect is more obvious, and the function of rapidly and accurately processing big data is achieved.

The invention relates to an inverse matrix-free self-adaptive filtering-based SINS (strap-down inertial navigation) / DVL (Doppler velocity log) combined positioning method and belongs to the technicalfield of high-precision SINS / DVL combined positioning. The invention aims to solve the problems of low calculation efficiency and poor stability of a traditional SINS / DVL combined navigation filteringalgorithm due to the complex calculation process of the algorithm. The inverse matrix-free self-adaptive filtering-based SINS (strap-down inertial navigation) / DVL (Doppler velocity log) combined positioning method of the invention includes the following steps that: corresponding state initial values and observation values are acquired based on the sensor information of a strap-down inertial navigation system and a Doppler velocity log; a corresponding system equation and observation equation based on a combined navigation error model are constructed, an inverse matrix-free self-adaptive filtering algorithm is adopted to correct errors, so that the post-correction speed and position error information of a target is obtained; and finally, the obtained error information and the observation information of the strap-down inertial navigation system and the Doppler velocity log are fused, so that a high-precision positioning result is obtained. With the method of the invention adopted, on the basis of decreasing a calculation amount and optimizing a calculation process, the reliability and stability of a system can be ensured, and the positioning precision of underwater autonomous navigation is improved.

The invention discloses a variable regularization proportional normalized sub-band adaptive filtering method, which is used for updating an adaptive filter based on a hearing aid feedback control system of a normalized subband adaptive filtering algorithm, based on proportional adaptive technology and variable regularization parameter variation and combined with a subband adaptive filtering algorithm. In order to further improve performance of the adaptive filtering algorithm, the method updates regularization parameters based on input signals, mean square error and estimated system noise energy and through the mean square deviation maximum reduction principle to form a variable regularization proportional parameter-update sub-band adaptive filtering method; and the method, applied to an acoustic feedback system of a hearing aid, can obtain a better acoustic feedback cancellation effect and can obtain lower steady-state misadjustment amount and faster convergence speed.

The invention discloses a variable step size normalized sub-band adaptive filter method applied to acoustic feedback inhibition. According to a hearing-aid feedback control system based on a normalized sub-band adaptive filter algorithm, on the basis of a system noise cancellation principle, estimation is carried out through adoption of an iterative shrinkage method, thereby obtaining noiseless sub-band prior error power; and each sub-band step size is updated, thereby achieving the purpose of eliminating an acoustic feedback signal. According to the method, through adoption of the variable step size adaptive filter method, a coefficient of a sub-band filter is updated; through application of the method in the hearing-aid feedback control system with a long echo path and a short echo path, the acoustic feedback signal can be eliminated well; moreover, relatively high convergence rate and relatively low maladjustment can be obtained; and the contradiction between the steady state maladjustment and the convergence rate in the adaptive filter method is solved.

The invention discloses a signal processing method for a borborygmus signal monitoring system. According to the method, after borborygmus data are quantized into digital signals with sampling rate of Fs by a microphone, the digital signals are treated based on the self-adaptive filtering algorithm and sectioned according to window length, the signal frequency spectrum of each signal section is solved by means of the fast Fourier transform algorithm, the signal frequency spectra are firstly applied to intensity detection, then subjected to correlation-based detection and analysis with a median frequency spectrum model within the frequency band to be tested, and finally sent to a comparator to be compared with a reference value, only the signals with correlation coefficients high enough can be shown in a data window as borborygmus signals and presented by the output result 1, and 0 is output otherwise. On this basis, RIM, IMF and AMT are also provided to be used for evaluating the intestinal tract activity condition of a testee. The method is used for the wearable human body borborygmus signal monitoring system and can be implanted in equipment hardware or a monitoring server in hospital, and the position of a borborygmus of the testee can be marked according to borborygmus signals input into the system, so that technical support is provided for a doctor to offer reasonable postoperative rehabilitation nursing to abdominal surgery patients.

The invention relates to a nuclear magnetic resonance signal real-time noise offsetting device for multiple near-end reference coils. A main receiving coil and one or more reference receiving coils pass through a multi-channel signal conditioning circuit of a ground nuclear magnetic resonance instrument and are connected with a digital analog converter to be converted into digital values, the digital values are input into a digital signal processing chip, the digital analog converter is connected with a main self-adaptation filter, and reference signal channel digital analog converters are connected with self-adaptation filters respectively. The multiple reference coils can offset the environment noise produced by a plurality of noise sources, the occupied space limit caused by pavement of the coils can be reduced through near-end reference, the application range of the reference coils is widened, the loss of the nuclear magnetic resonance signals collected in the main receiving coil can be avoided through the multiple sets of self-adaptation filters, the signal to noise ratio is improved, the real-time self-adaptation filtering algorithm is achieved through the digital signal processing chip, and data collecting and processing efficiency is improved.

Vibrations of a principal machine are reduced at the fundamental and harmonic frequencies by driving the drive motor of an active balancer with balancing signals at the fundamental and selected harmonics. Vibrations are sensed to provide a signal representing the mechanical vibrations. A balancing signal generator for the fundamental and for each selected harmonic processes the sensed vibration signal with adaptive filter algorithms of adaptive filters for each frequency to generate a balancing signal for each frequency. Reference inputs for each frequency are applied to the adaptive filter algorithms of each balancing signal generator at the frequency assigned to the generator. The harmonic balancing signals for all of the frequencies are summed and applied to drive the drive motor. The harmonic balancing signals drive the drive motor with a drive voltage component in opposition to the vibration at each frequency.

The invention belongs to the technical field of electrocardiogram monitoring, and provides a fetus electrocardiosignal extracting method and device. The method includes the steps that one-way abdominal wall electrocardiosignals obtained from parent body abdominal wall monitoring of an abdominal wall electrode are received; a parent body electrocardio R wave crest position in the abdominal wall electrocardiosignals is extracted; N left and right sections, with the preset length, of the R wave crest position are intercepted and averaged to obtain an average waveform template, the average waveform template and a current waveform module are superposed to generate a recursion waveform template, and N is a positive integer larger than or equal to 5; the recursion waveform template is used as reference signals, and fetus electrocardiosignals are obtained from the abdominal wall electrocardiosignals through a self-adaptive filtering algorithm. As the recursion waveform template superposing is conducted on the abdominal wall electrocardiosignals to generate reference signals of a self-adaptive filter, and self-adaptive fetus electrocardiosignal separation is achieved under a single lead system.

The invention discloses an adaptive filtering-based spacecraft guidance instrument on-orbit calibration method. According to the method, carrier position and attitude output by a strapdown inertial navigation device and an attitude matrix output by a star sensor are used for construction and measurement, and a measurement equation is built. A self-adaptive filtering algorithm is designed and estimated values of gyroscope random constant drift and star sensor installation error are acquired through filtering calculation so that combined system on-orbit self-calibration is realized. The method solves the problem that the existing guidance tool error separation technology causes misalignment of the spacecraft in the sky and ground coordinates, realizes real-time on-orbit calibration of an error coefficient of the guidance tool, utilizes a simple algorithm and is easy to project.

The invention relates to a variational Bayesian self-adaptive filtering method. For the target status estimation problem in a case of unknown process noise time-varying, inverse Wishart distribution is used for a priori modeling on process noise covariance through introducing a new hidden variable, and target status, the process noise covariance and the hidden variable are jointly optimized undera variational Bayesian framework through an iterative mechanism to improve tracking accuracy of the target status. Compared with the existing filtering algorithms, the method of the invention directlyselects the conjugate a priori distribution of the process noise covariance, carries out iterative optimization of a target status estimation result under the variational Bayesian framework, obtainsa new self-adaptive filtering algorithm, and has certain theoretical and practical engineering significance for the field of filters.