17 results about "Memory polynomial" patented technology

The application belongs to the technical field of digital pre-distortion, and particularly relates to a digital pre-distortion method of an improved memory polynomial model based on an FFT convolution correlation function. The digital pre-distortion method adopts an improved memory polynomial model, adds a direct current term to correct errors caused by partial direct current bias, and adds a conjugate term to correct distortion caused by an image frequency. In a delay alignment module, a convolution correlation function method of fast Fourier transform (FFT) is used as a loop delay estimation algorithm. Compared with other correlation function loop delay estimation algorithms, the efficiency is greatly improved, and oversampling is not needed to improve accuracy.

The application provides a signal predistortion processing method and device, electronic equipment and storage medium, and relates to the technical field of communication. The method comprises the following steps: firstly, calculating predistortion complex array by a memory polynomial model based on historical baseband data and historical baseband feedback data; then, applying the predistortion complex array to a baseband signal to generate an in-phase component signal and a quadrature component signal containing amplitude and phase predistortion information; then, respectively performing carrier modulation on the in-phase component signal and the quadrature component signal to obtain an in-phase branch signal and a quadrature branch signal; then, calculating a predistortion real signal based on the in-phase branch signal and the quadrature branch signal; and finally, performing carrier modulation on the predistortion real signal to obtain an up-converted signal. The technical scheme provided by the application solves the technical problem that phase predistortion information cannot be preserved in the signal predistortion process.

The invention provides a single-input digital pre-distortion design method, medium and system for an impedance mismatch power amplifier, and the method comprises the steps: obtaining a plurality of output signals and a double-input memory polynomial model of the power amplifier in a complex impedance mismatch environment, and enabling the output signals and input signals to form a data set; introducing a neural network model into the double-input memory polynomial model to obtain a single-input fusion model; and training the single-input fusion model by using each output signal in the data set so that the single-input fusion model learns towards the direction of the input signal to obtain solved overall model parameters and internal parameters, and then performing pre-distortion control. According to the invention, the single directional coupler is used for collecting the output signal for solving the unknown parameter, and then the single-input fusion model containing the known parameter is used for predistortion control, so that the hardware overhead can be reduced, and the control effect is improved as the output signal is most accurate.

The application discloses a wideband digital pre-distortion method based on a multi-dimensional amplitude switching memory polynomial, and proposes a switching behavior model, which divides the input signal sequence of the model into different classes by using the amplitudes of the current input signal and the amplitudes of the memory input signal, and uses a separate model to operate the input data of each class. Each class uses a memory polynomial model as a sub-model, combines the multi-dimensional amplitude classification algorithm with the MP model, proposes a memory polynomial model based on multi-dimensional amplitude switching, and adopts a direct learning structure to iteratively complete nonlinear system identification and model coefficient calculation. The method finally restores data closer to the real power amplifier output signal at the same sampling rate, achieves better pre-distortion effect, and has certain improvement in performance compared with the digital pre-distortion based on the MP model.

The application discloses a DPD implementation method and system in a hybrid beamforming system, wherein a PA inverse transfer function parameter is estimated according to a PA baseband input signal and an output signal through a generalized memory polynomial (GMP) model, and digital pre-distortion DPD is implemented through a cascade digital pre-distortion architecture comprising a phase shifter parameter matrix module, a PA inverse transfer function module and a phase compensation module. The application significantly simplifies the hardware configuration of a feedback link, and can maintain excellent linearization performance without recalibration when the beam direction changes, and has strong robustness.

The invention discloses a digital pre-distortion method and system for a gallium nitride power amplifier. The method comprises the following steps: calculating an instantaneous modulus value of an original signal, looking up a table to obtain a pre-distortion coefficient, carrying out pre-distortion compensation, inputting the pre-distortion compensation into the gallium nitride power amplifier, amplifying and outputting the pre-distortion compensation. The construction method of the predistortion coefficient lookup table comprises the following steps: acquiring an original signal and a feedback signal, and performing synchronous alignment processing; calculating an instantaneous module value corresponding to each sampling point in the synchronously aligned original signal, and averaging to obtain an average module value; constructing all items of the generalized memory polynomial model; adding a moving average generalized memory polynomial item in the generalized memory polynomial model to obtain a pre-distortion model; and estimating coefficients of all items in the pre-distortion model, and converting the coefficients into a pre-distortion coefficient lookup table which takes the instantaneous module value as an index. According to the invention, high-precision compensation for the signal distortion of the gallium nitride power amplifier is realized while high processing efficiency is achieved.

The application discloses a kind of based on two-stage attention neural network's down-sampling DPD method and device, method includes: using memory polynomial method determines rough estimate DPD parameter;Based on a plurality of input sample signals and each input sample signal corresponding rough estimate DPD signal constructs sample set;Wherein, each input sample signal corresponding rough estimate DPD signal is determined based on the input sample signal and rough estimate DPD parameter;Utilize sample set training DPD neural network model, and according to the model after training realizes down-sampling DPD signal estimation;Utilize sample set training model includes: input input sample signal into model, to make model output the input sample signal corresponding output DPD signal;According to the difference between the output DPD signal and rough estimate DPD signal corresponding to input sample signal, incremental training is carried out to DPD neural network model, reduce the calculation cost, improve the power amplifier nonlinear correction efficiency.

The application provides a visible light communication pre-distortion method based on twin support vector regression, which comprises the following steps: step S1, a visible light communication offline digital pre-distortion platform is built; step S2, based on the signal information obtained by the visible light communication offline digital pre-distortion platform in step S1, a training set of an amplitude enhancement structure is constructed; step S3, based on the training set of the amplitude enhancement structure constructed in step S2, a visible light communication pre-distortion nonlinear suppression method based on amplitude TSVR is established, and a test set is formed; step S4, the training set of the training set of the amplitude enhancement structure and the test set are used for model training; step S5, after model training, pre-distortion device parameters are obtained, and a visible light communication pre-distortion system is further designed and constructed. The amplitude enhancement-based TSVR time delay structure is obviously improved in modeling accuracy, compared with the existing memory polynomial modeling, the NSME is improved by more than 22 dB; compared with the generalized memory polynomial, the accuracy is improved by about 5.5 dB on average.

A distortion apparatus for a power amplifier comprises: pre-distortion circuitry that comprises a memory polynomial look-up-table (LUT) circuit; and post-distortion circuitry that updates the pre-distortion circuitry based on an output of the power amplifier. The pre-distortion circuitry comprises a plurality of LUTs. Each LUT corresponds to a different memory depth of a memory polynomial of a Volterra series. Each LUT indexed by an instantaneous power of an input sample delayed by an amount corresponds to a respective memory depth of each LUT. An output of the memory polynomial LUT circuit corresponds to a summation of an output of each LUT multiplied by the input sample delayed by the amount corresponding to the respective memory depth of each LUT, and an output of the pre-distortion circuitry is provided to the power amplifier.

The invention discloses a high-precision fractional memory polynomial digital predistortion method and device, and the method comprises the steps: carrying out the synchronous sampling of an input signal of a predistorter and an output signal of a power amplifier, and obtaining a sampling sequence; substituting the sampling sequence into a pre-constructed fractional memory polynomial power amplifier expression, deriving a fractional memory polynomial power amplifier coefficient by adopting a least square theory, and solving by adopting characteristic decomposition; the fractional memory polynomial power amplification coefficient is subjected to predistortion compensation through an adaptive learning device; according to the fractional memory polynomial power amplifier model provided by the invention, the characteristics of a linear region and a saturation region of the power amplifier are fitted at the same time, and a good linearization effect on the power amplifier is achieved; the fractional memory polynomial power amplifier model can directly solve a numerator and denominator combination coefficient by using a characteristic decomposition theory, and can update a fractional polynomial power amplifier coefficient in real time on orbit.

The application discloses a kind of multi-phase digital pre-distortion processing method, device, equipment, medium and program product, the method includes: according to digital pre-distortion memory polynomial model, setting polynomial calculation module and filter module;According to target processing clock frequency, determine the input road number of the polynomial calculation module;Then the polynomial calculation module is used to carry out polynomial calculation to multiple signal components, and multiple polynomial calculation results are obtained;Wherein, multiple signal components are obtained after the transmission signal to be input to power amplifier is decomposed according to the input road number;The filter module is used to combine the polynomial calculation result, and the transmission signal after pre-distortion processing is obtained, so that the processing clock frequency of digital pre-distortion DPD can be reduced, high-speed pre-distortion processing is completed using lower rate, and the burden of system is reduced.

Methods and systems for AI-based digital pre-distortion for digital envelope tracking power amplifiers. A method includes receiving a measurement of a digital envelope at a digital predistortion module having a neural network (NN)-based digital predistortion structure for a digital envelope tracking (DET) system, receiving a transmit signal at the digital predistortion module, inputting the measurement of the digital envelope and the transmit signal into a generalized memory polynomial (GMP) model of the AI-based digital predistortion structure to produce a pre-distorted transmit signal, processing the pre-distorted transmit signal using an AI model of the AI-based digital predistortion structure to produce an adjusted pre-distorted transmit signal to correct one or more non-linearities in a power amplifier operable coupled to the digital predistortion module, and generating an output signal using the power amplifier based on the adjusted pre-distorted transmit signal.

This invention relates to a low-complexity dual-cycle digital predistortion method for MIMO transmitters with crosstalk, belonging to the field of radio frequency communication technology. It solves the problem in existing technologies where predistortion parameter extraction modules cannot balance estimation accuracy and computational complexity. The specific steps include: establishing a power amplifier memory polynomial model based on the acquired input and output signals of the power amplifier in the MIMO transmitter; establishing a digital predistortion model and a post-distortion model accordingly; iteratively training the post-distortion model using the post-distortion model and the SPSA algorithm to obtain initial predistortion screening coefficients; iteratively training the predistortion model using the SPSA algorithm based on the digital predistortion model and the initial predistortion screening coefficients to obtain the final predistortion coefficients and the trained digital predistortion model; and using the trained digital predistortion model to perform nonlinear distortion compensation on the power amplifier, ensuring the accuracy of the predistortion parameters while reducing computational complexity.

The application discloses a power amplifier DPD coefficient estimation method based on an RLS algorithm, obtains an input signal of an input power amplifier (PA) and a feedback output signal after gain adjustment; performs time delay alignment processing on the input signal and the feedback signal; based on a memory polynomial model and a memory polynomial model of cross terms, constructs a nonlinear input matrix and a coefficient matrix of the PA; the recursive least square (RLS) algorithm based on Givens rotation of complex QR decomposition is adopted, a CORDIC algorithm is used to solve a rotation angle and a modulus value, the nonlinear input matrix and the coefficient matrix are iteratively calculated, and DPD coefficients are estimated. The memory polynomial model or the memory polynomial model of cross terms is adopted, so that the PA model is clearer, and data processing is facilitated; the RLS algorithm adopts the RLS expression based on Givens rotation of complex QR decomposition, direct solution of a matrix inverse is not needed, a large number of division operations are avoided, and the problem that the original RLS algorithm is not conducive to engineering implementation in DPD coefficient estimation is solved.

This disclosure provides an apparatus and method for compensating for nonlinearity in a power amplifier. An apparatus configured to perform wireless communication to accelerate a neural network includes: a predistortion circuit configured to generate a predistorted input signal by performing predistortion on an input signal based on a parameter set including multiple coefficients; a power amplifier configured to generate an output signal by amplifying an RF signal based on the predistorted input signal; and a parameter acquisition circuit configured to obtain second memory polynomial modeling information corresponding to an operating frequency band based on first memory polynomial modeling information corresponding to each of a plurality of frequency bands, and to use the second memory polynomial modeling information to obtain a parameter set according to an indirect learning structure.

This application provides a predistortion extended model and a method and apparatus for implementing predistortion, used to improve the linearity of a power amplifier (PA) and reduce the power consumption of a base station system under conditions of limited sampling rate. The method includes: determining the input and output predicted signals corresponding to any time interval between two adjacent sampling times of the power amplifier using linear interpolation; extending the GMP model by using the input and output predicted signals as the input and output signals of the generalized memory polynomial model (GMP) to obtain an extended GMP model; querying the first lookup table (LUT) value corresponding to the first input signal using the original GMP model, and querying the second LUT value corresponding to the first input signal using the extended model; and generating the predistortion signal based on the first and second LUT values.

The invention belongs to the technical field of data processing of the computer information disciplinary, and relates to a broadband power amplifier linearization method and system based on predistortion, and the method comprises the steps: collecting and preprocessing a radio frequency input baseband signal, an actual output baseband signal and tube core heat flow data of a broadband power amplifier, and obtaining an initial data set; calculating a transient heat dissipation gradient at each sampling moment based on the signal energy difference and the tube core temperature change rate; in combination with broadband distribution characteristics of radio frequency input baseband signals, broadband thermal teratogenesis factors at all sampling moments are obtained; performing dynamic tracking compensation on a cross term coefficient of a generalized memory polynomial based on the distortion factor to obtain an update coefficient; and the pre-distortion signal is reconstructed by utilizing the updating coefficient, so that high-linearity output of the broadband power amplifier is realized. The dynamic response speed of coefficient updating can be improved, thermally induced nonlinear distortion of a power amplifier is effectively offset, and the linearity and stability of a high-power broadband radio frequency test are improved.