Search

CN-122027420-A - Pulse modulation distortion pre-correction device and method and electronic equipment

CN122027420ACN 122027420 ACN122027420 ACN 122027420ACN-122027420-A

Abstract

The disclosure provides a pulse modulation distortion pre-correction device and method and electronic equipment. The device comprises a DFT interpolation module, a function reconstruction module, an auxiliary equation solving module and an error analysis module, wherein the DFT interpolation module is configured to acquire a time domain discrete data signal output by an audio device and convert the time domain discrete data signal into a high-resolution time domain sequence, the function reconstruction module is configured to move sampling points in the high-resolution time domain sequence into a normalized coordinate system and conduct interpolation to obtain an input signal equation, the sawtooth carrier generated by a carrier generator is interpolated into a carrier equation, the auxiliary equation solving module is configured to establish an auxiliary equation and estimate the root of the auxiliary equation, and the error analysis module is configured to conduct error analysis on the root of the auxiliary equation and output a corrected modulation signal when the root of the auxiliary equation meets an error threshold. The device disclosed by the disclosure replaces the original flow processed by the digital circuit with a software code, improves the pulse modulation precision by using a mathematical algorithm, and reduces the harmonic distortion.

Inventors

  • WANG XINCHENG
  • HU RUBO
  • XIE SIHUAI
  • WU XU

Assignees

  • 上海翰迈电子科技有限公司

Dates

Publication Date
20260512
Application Date
20260106

Claims (10)

  1. 1. A pulse modulation distortion pre-correction apparatus, comprising: A DFT interpolation module configured to acquire a time-domain discrete data signal output by an audio device and to convert the time-domain discrete data signal into a high-resolution time-domain sequence; A function reconstruction module configured to move sampling points in the high-resolution time domain sequence into a normalized coordinate system comprising a plurality of unit time coordinates, interpolate the sampling points moved into the normalized coordinate system to obtain an input signal equation, and interpolate a sawtooth carrier generated by a carrier generator into a carrier equation; an auxiliary equation solving module configured to establish an auxiliary equation from the input signal equation and the carrier equation and estimate a root of the auxiliary equation in a rightmost unit interval in the normalized coordinate system, and And the error analysis module is configured to perform error analysis on the root of the auxiliary square program and output a corrected modulation signal when the root of the auxiliary square program meets an error threshold value, wherein the modulation signal is positioned in the right-most unit time in the normalized coordinate system in the high-resolution time domain sequence.
  2. 2. The apparatus of claim 1, wherein the DFT interpolation module is configured to: Converting the time domain discrete data signal into a discrete frequency domain signal by fourier transform; interpolation of the discrete frequency domain signal using spectral interpolation to obtain spread spectrum data, and The spread spectrum data is converted into the high resolution time domain sequence by an inverse fourier transform.
  3. 3. The apparatus of claim 1, wherein the normalized coordinate system comprises n+1 unit time coordinates, each unit time coordinate having a start point with a vertical axis having only one corresponding discrete sample point.
  4. 4. The apparatus of claim 3, wherein the function reconstruction module is configured to: sequentially moving n+1 sampling points in the high-resolution time domain sequence into the normalized coordinate system from right to left; Interpolation of sampling points located in n+1 unit time coordinates in the normalized coordinate system by interpolation method to obtain an n-order polynomial, and The nth order polynomial is converted to the input signal equation.
  5. 5. The apparatus of claim 3, wherein the function reconstruction module is further configured to: after the corrected modulation signal is output, the sampling points in the (n+2) th high-resolution time domain sequence are moved into the normalized coordinate system, and the input signal equation is updated.
  6. 6. The apparatus of claim 1, wherein the auxiliary equation solving module is configured to: The auxiliary equation is set based on the carrier equation minus the input signal equation, and a difference of the carrier equation and the input signal equation is set to zero.
  7. 7. The apparatus of claim 1, wherein the auxiliary equation solving module further receives an error vector signal output by the error analysis module, and wherein the auxiliary equation solving module is configured to: Estimating the root of the auxiliary equation in the rightmost unit interval in the normalized coordinate system by using a bisection method, and And after receiving the error vector signal, re-estimating the interval where the estimated root of the auxiliary equation is located by using a bisection method to obtain a new root of the auxiliary equation.
  8. 8. The apparatus of claim 1, wherein the error analysis module is configured to: performing error calculation on the root of the auxiliary equation to obtain an error value; outputting a modulated signal located in the right unit time in the normalized coordinate system in the high-resolution time domain sequence after pre-correction when the error value meets an error threshold value, and And outputting an error vector signal to the auxiliary equation solving module when the error value does not meet an error threshold value.
  9. 9. A method for pre-correction of pulse modulation distortion, comprising: Acquiring a time domain discrete data signal output by an audio device and converting the time domain discrete data signal into a high resolution time domain sequence; Moving sampling points in the high-resolution time domain sequence into a normalized coordinate system comprising a plurality of unit time coordinates, interpolating the sampling points moved into the normalized coordinate system to obtain an input signal equation, and interpolating a sawtooth carrier generated by a carrier generator into a carrier equation; Establishing an auxiliary equation according to the input signal equation and the carrier equation, and solving the root of the auxiliary equation in the rightmost unit interval in the normalized coordinate system, and And carrying out error analysis on the root of the auxiliary square program, and outputting a pre-corrected modulation signal when the root of the auxiliary square program meets an error threshold value, wherein the modulation signal is positioned in the rightmost unit time in the normalized coordinate system in the high-resolution time domain sequence.
  10. 10. An electronic device, comprising: A memory configured to store an executable program, and A processor configured to execute the program to cause the electronic device to perform the pulse modulation distortion pre-correction method according to claim 9.

Description

Pulse modulation distortion pre-correction device and method and electronic equipment Technical Field The present disclosure relates to the field of digital signal processing technologies, and in particular, to a device and a method for pre-correcting pulse modulation distortion, and an electronic device. Background Unified pulse-width modulation (UPWM), since the processed modulated signal is a discrete pulse-code (PCM) signal, there is inherent harmonic distortion in the baseband compared to a natural pulse-width modulator (NPWM) constructed from a continuous-time signal and a triangular carrier. Since the first CD invention in the eighties of the last century, audio sources were digitized successively, and commonly used I 2 S, USB audio, bluetooth audio, and music and speech signals in mobile communications, the usual signal format PCM, but the sampling frequency and quantization bit number were different. In addition, UPWM is also a key technology in high-efficiency power electronics systems, such as signal processing in wind power, photovoltaic inverters, electric locomotives, electric car drives, coulometers, and the like, and is widely used. Therefore, researches in academia and engineering world are used for overcoming enthusiasm continuous rising of UPWM inherent distortion, achievements and patent layers are endless, and the researches are classified into three major categories of pseudo NPWM simulation method, predistortion method and negative feedback method, wherein the pseudo NPWM simulation method is the main stream, and the successful application proportion is the highest. However, interpolation adopted by the existing or disclosed pseudo NPWM simulation method generally has the defects of low precision, large residual distortion, complex circuit, large occupied silicon area and the like. Disclosure of Invention The embodiment of the disclosure provides a pulse modulation distortion pre-correction device and method, and electronic equipment, which can improve pulse modulation precision, reduce distortion and provide corrected pulse signals. In a first aspect, embodiments of the present disclosure provide a pulse modulation distortion pre-correction apparatus. The pulse modulation distortion pre-correction device comprises a DFT interpolation module, a function reconstruction module, an auxiliary equation solving module and an error analysis module, wherein the DFT interpolation module is configured to acquire a time domain discrete data signal output by audio equipment and convert the time domain discrete data signal into a high-resolution time domain sequence, the function reconstruction module is configured to move sampling points in the high-resolution time domain sequence into a normalized coordinate system comprising a plurality of unit time coordinates, interpolate the sampling points moved into the normalized coordinate system to obtain an input signal equation and interpolate a sawtooth carrier generated by a carrier generator into a carrier equation, the auxiliary equation solving module is configured to establish an auxiliary equation according to the input signal equation and the carrier equation and estimate the root of the auxiliary equation in the rightmost unit interval in the normalized coordinate system, and the error analysis module is configured to conduct error analysis on the root of the auxiliary equation and output a corrected modulation signal when the root of the auxiliary equation meets an error threshold value, and the modulation signal is located in the rightmost unit time in the normalized coordinate system in the high-resolution time domain sequence. In some implementations of the first aspect, the DFT interpolation module is configured to convert the time-domain discrete data signal into a discrete frequency-domain signal by fourier transform, interpolate the discrete frequency-domain signal with spectral interpolation to obtain spread spectrum data, and convert the spread spectrum data into the high-resolution time-domain sequence by inverse fourier transform. In some implementations of the first aspect, the normalized coordinate system includes n+1 unit time coordinates, and each unit time coordinate has only one corresponding discrete sampling point on a longitudinal axis of a starting point of the unit time coordinate. In some implementations of the first aspect, the function reconstruction module is configured to sequentially shift n+1 sampling points within the high resolution time domain sequence from right to left into the normalized coordinate system, interpolate the sampling points within n+1 unit time coordinates in the normalized coordinate system using interpolation to obtain an nth order polynomial, and convert the nth order polynomial to the input signal equation. In some implementations of the first aspect, the function reconstruction module is further configured to, after the corrected modulated signal is output, shift sample points within the