Overlap band phase compensation method and system based on secondary integer loss point and phase shift

Through the overlap band phase compensation method based on the secondary integer point drop and phase shift, the problems of overlap band signal amplitude cancellation and phase randomness in high-speed oscilloscopes are solved, and the effective compensation of overlap bands and the optimization of system resources are realized. Ensure consistency of phase frequency response after each boot.

CN118731824BActive Publication Date: 2024-11-15成都玖锦科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411196311.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-15
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

In the high-speed oscilloscope, the non-ideal characteristics of the hardware design of the subband decomposition filter during the frequency domain interleaving process lead to the phase difference of the same frequency signal, resulting in the cancellation of the amplitude of the overlapping band signal, and the phase changes randomly after the system reset, resulting in the The signal is distorted and the compensation process consumes a lot of system resources.

Method used

The overlap band phase compensation method based on the secondary integer point drop and phase shift is adopted. Through factory calibration and automatic calibration steps, the linear fitting algorithm is used to fit the primary integer point drop value and overlap band nonlinear phase difference data. And calculate the second-level integer drop point value and digital local oscillator coefficient to achieve effective compensation of the overlapping band.

Benefits of technology

The effectiveness of overlap band compensation is ensured, extreme peaks and depressions of noise amplification and system group delay are avoided, and the phase relationship between subbands is consistent every time the system is powered on, and there is no need to re-produce phase frequency calibration and update.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118731824B_ABST
    Figure CN118731824B_ABST
Patent Text Reader

Abstract

The present invention relates to an overlapping band phase compensation method and system based on secondary integer dropout and phase shift, belonging to the field of communication technology, and the phase compensation method comprises: a factory calibration step: calculating the phase difference based on the overlapping band sweep data, fitting the primary integer dropout value and the overlapping band nonlinear phase difference data using a linear fitting algorithm according to the angular frequency and phase difference of the overlapping band; a power-on automatic calibration step: sweeping the overlapping band, and calculating the secondary integer dropout value and the digital local oscillator coefficient according to the sweep result. The present invention first ensures that the overlapping band compensation is effective, that is, it meets the requirements of not amplifying noise after splicing and not having extreme peaks and depressions of the system group delay; it ensures that the relative relationship of the system group delay at the non-overlapping bands of each sub-band is consistent each time the system is powered on, that is, no phase-frequency calibration update is required after each self-start.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to an overlapping band phase compensation method and system based on secondary integer point loss and phase shifting. Background Art

[0002] As a representative of electronic measurement and testing instruments, high-speed oscilloscopes play an important role in wireless communication measurement, national defense and military industry, aerospace, and industrial medical instruments. At present, the highest indicators of single-chip ADCs of mature commercial products are 10.4GSa / s sampling rate, 8GHz bandwidth, and 12-bit effective bits, which are difficult to meet the indicator requirements of acquisition systems with high sampling rates and large bandwidths. Parallel sampling architecture is currently an important means to improve the performance indicators of acquisition systems. Compared with the disadvantage that the bandwidth in time interleaving depends on the bandwidth indicator of the single-chip ADC, the hybrid architecture of time interleaving and frequency domain interleaving can break through the limitations of single-chip ADCs, allowing the acquisition system to have a larger bandwidth while achieving a higher sampling rate. Specifically, this technology decomposes the acquisition system into multiple sub-bands, down-converts different sub-bands to low frequency bands, uses single-chip or multiple low-performance ADCs for sampling, and then achieves distortion-free sampling of the signal through interpolation filtering and up-mixing.

[0003] The non-ideal characteristics of the hardware design of the sub-band decomposition filter during the frequency domain interleaving process will cause the same-frequency signal to enter the two adjacent sub-bands at the same time. At the same time, the difference in hardware characteristics of different sub-bands will also introduce the phase difference of the same-frequency signal, which will eventually cause the amplitude of the signal after reconstruction and splicing at the overlap of the two sub-bands to cancel each other out; in addition, due to temperature factors, different analog circuits, calculator offsets, and ADC synchronous reset signal lights, the phase of the analog local oscillator signal is random at each system reset (restart), but once the system works stably, the corresponding phase value will be fixed and changed, which is specifically manifested as the frequency response difference between the self-bands each time the system starts. The existing technology uses cascaded integer delay / drop point, fractional delay FIR filters and nonlinear phase compensation IIR filters to make the phase difference between sub-bands close. This method consumes a lot of system computing resources during the calculation process. Considering that after the sub-band splicing, the system will cascade all-pass phase-frequency compensation filters to compensate for the nonlinear phase characteristics of the system; there are also single-subband phase compensation modules that delete nonlinear phase IIR filters and fractional filters to reduce the time and hardware resources consumed by system design.

[0004] However, the existing technologies focus on the phase difference of the overlapped band after compensation. In fact, the calibration of the overlapped band will also affect the overall frequency response characteristics of the subsequent system. For example, a small difference in integer delay loss will change the group delay of the sub-band, that is, the phase-frequency response of the broadband acquisition system will change, resulting in signal distortion. In order to ensure that the phase-frequency response within the system band meets the linear condition, the traditional method needs to re-perform the system's phase-frequency compensation. The steps of sampling and filter coefficient calculation in the compensation process will cause a waste of system resources and time consumption. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an overlapping band phase compensation method based on secondary integer point loss and phase shifting, thereby solving the shortcomings of the prior art.

[0006] The object of the present invention is achieved by the following technical solution: an overlapping band phase compensation method based on secondary integer point loss and phase shift, the phase compensation method comprising:

[0007] Factory calibration steps: Calculate the phase difference based on the overlap band sweep data, and use the linear fitting algorithm to fit the first-order integer loss point value and the overlap band nonlinear phase difference data according to the overlap band angular frequency and phase difference;

[0008] Automatic calibration steps at power on: Sweep the overlap band and calculate the secondary integer drop point value and digital local oscillator coefficient based on the sweep result.

[0009] The factory calibration steps specifically include the following:

[0010] Determine the frequency range of the overlapping band: preset the maximum amplitude frequency difference of the same frequency points of two sub-bands at the overlapping band boundary to be , and calculate the amplitude-frequency response of each sub-band transition band by sine sweep, setting the amplitude difference between the two sub-bands at the same frequency point to , and satisfies ,in, The digital angular frequency is The amplitude value of the low-frequency subband signal is The digital angular frequency is The amplitude value of the time-frequency sub-band signal;

[0011] Overlapping band nonlinear phase difference extraction: The frequency range of the overlapping band is divided into N The corresponding overlapping band sweep sequence is , the sweep step frequency is The signal source inputs a sine sweep signal with different frequencies, and uses the sine three-parameter estimation algorithm or Fourier transform method to estimate the phase of the high-frequency sub-band for different frequencies. and the low frequency subband phase , and calculate the phase difference of the same frequency point between the high-frequency and low-frequency sub-bands as , ω1 represents the first frequency sweep;

[0012] Extract the linear phase difference and the first-order dropout value within the overlapping band: Set the optimization formula , obtained by least squares method The optimal solution as well as The optimal solution They are , , and then the linear phase difference is obtained , and the nonlinear phase difference , and the first-level loss point value ,like If is negative, it means integer delay in high frequency sub-band, otherwise in low frequency sub-band, where is the phase difference between sub-bands, k is the fitted linear phase difference slope, and θ is the linear phase difference offset.

[0013] The power-on automatic calibration step specifically includes the following contents:

[0014] After the first level drop point is turned on, overlap band data acquisition: Set the sweep frequency step frequency to After the first-level point loss, the signal source inputs a sine sweep signal with different frequencies, and the phase of the same frequency point between sub-bands is estimated for different frequencies using the sine three-parameter estimation algorithm or the Fourier transform method. and , and calculate the phase difference of the same frequency point between sub-bands as ;

[0015] Secondary loss point calculation: Use linear fitting algorithm to obtain the overlap band phase difference slope , and the secondary loss value is ,in, Indicates rounding down. Indicates the multiple of the system storage frequency and sampling frequency;

[0016] Update the digital local oscillator coefficient: According to the formula Calculate the phase difference after the overlapping band is lost, and the digital local oscillator coefficient of the high-frequency sub-band is expressed as , then according to the formula The updated digital local oscillator coefficients are calculated, where the initial local oscillator phase value is , the angular frequency is , is the number of sampling points.

[0017] An overlapping band phase compensation system based on secondary integer point loss and phase shifting, the system comprising a factory calibration unit and a power-on automatic calibration unit;

[0018] The factory calibration unit calculates the phase difference based on the overlapping band frequency sweep data, and uses a linear fitting algorithm to fit the first-order integer loss point value and the overlapping band nonlinear phase difference data according to the overlapping band angular frequency and phase difference;

[0019] The power-on automatic calibration unit performs a frequency sweep on the overlapping band, and calculates the secondary integer drop point value and the digital local oscillator coefficient according to the frequency sweep result.

[0020] The factory calibration unit includes an overlapping band frequency range determination subunit, an overlapping band nonlinear phase difference extraction subunit, and an overlapping band range linear phase difference and first-level loss point value extraction subunit;

[0021] The overlapping band frequency range determination subunit: presets the maximum amplitude frequency difference of the same frequency points of two sub-bands at the overlapping band boundary to be , and calculate the amplitude-frequency response of each sub-band transition band by sine sweep, setting the amplitude difference between the two sub-bands at the same frequency point to , and satisfies ,in, The digital angular frequency is The amplitude value of the low-frequency subband signal is The digital angular frequency is The amplitude value of the time-frequency sub-band signal;

[0022] Overlapping band nonlinear phase difference extraction: The frequency range of the overlapping band is divided into N The corresponding overlapping band sweep sequence is , the sweep step frequency is The signal source inputs a sine sweep signal with different frequencies, and uses the sine three-parameter estimation algorithm or Fourier transform method to estimate the phase of the high-frequency sub-band for different frequencies. and the low frequency subband phase , and calculate the phase difference of the same frequency point between the high-frequency and low-frequency sub-bands as , ω1 represents the first frequency sweep;

[0023] Extract the linear phase difference and the first-order dropout value within the overlapping band: Set the optimization formula , obtained by least squares method The optimal solution as well as The optimal solution They are , , and then the linear phase difference is obtained , and the nonlinear phase difference , and the first-level loss point value ,like If is negative, it means integer delay in high frequency sub-band, otherwise in low frequency sub-band, where is the phase difference between sub-bands, k is the fitted linear phase difference slope, and θ is the linear phase difference offset.

[0024] The power-on automatic calibration unit comprises a first-level point loss after the overlap band data acquisition subunit, a second-level point loss calculation subunit and a digital local oscillator coefficient update subunit;

[0025] After the first-level drop point is turned on, the overlapping band data acquisition subunit: set the sweep frequency step frequency to After the first-level point loss, the signal source inputs a sine sweep signal with different frequencies, and the phase of the same frequency point between sub-bands is estimated for different frequencies using the sine three-parameter estimation algorithm or the Fourier transform method. and , and calculate the phase difference of the same frequency point between sub-bands as ;

[0026] Secondary loss point calculation: Use linear fitting algorithm to obtain the overlap band phase difference slope , and the secondary loss value is ,in, Indicates rounding down. Indicates the multiple of the system storage frequency and sampling frequency;

[0027] Update the digital local oscillator coefficient: According to the formula Calculate the phase difference after the overlapping band is lost, and the digital local oscillator coefficient of the high-frequency sub-band is expressed as , then according to the formula The updated digital local oscillator coefficients are calculated, where the initial local oscillator phase value is , the angular frequency is , is the number of sampling points.

[0028] The present invention has the following advantages: the overlapping band phase compensation method based on secondary integer point loss and phase shifting first ensures that the overlapping band compensation is effective, that is, it meets the requirements of not amplifying noise and not having extreme peaks and depressions in system group delay after splicing; it ensures that the relative relationship of the system group delay in the non-overlapping bands of each sub-band is consistent each time the system is powered on, that is, there is no need to perform phase-frequency calibration update each time after self-startup. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the process of the present invention;

[0030] Figure 2 A schematic diagram of the phase difference of the overlapping band after two startup compensations of the present invention;

[0031] Figure 3 This is a schematic diagram of the phase difference of the overlapping band after the power-on compensation when only integer points are lost twice;

[0032] Figure 4 A schematic diagram of group delay after two startup compensations of the present invention;

[0033] Figure 5 This is a schematic diagram of the group delay after startup compensation with only integer loss points twice;

[0034] Figure 6 FIG. 4 is a diagram showing the compensation effect of the method according to the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the drawings is not intended to limit the scope of protection of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application. The present invention is further described below in conjunction with the drawings.

[0036] The present invention specifically relates to an overlapping band compensation method based on secondary integer point loss and phase shifting. The phase difference between sub-bands caused by the system radio frequency hardware and the sub-band analysis filter group will not change with repeated power-on of the system, so the impact on the phase-frequency response of the system can be considered to be constant. However, there may be a different delay between adjacent acquisition boards each time they are powered on, and the delay presents an integer multiple relationship of a fixed value related to the logical data storage, that is, the difference in group delay between adjacent sub-bands is an integer multiple of the fixed value. Therefore, this method includes two steps: first-level integer dropout during factory calibration and second-level integer dropout during power-on automatic calibration: in factory calibration, firstly, the phase difference is calculated based on the overlapping band sweep data, and the first-level integer dropout value and the overlapping band nonlinear phase difference data are fitted using a linear fitting algorithm according to the angular frequency and phase difference of the overlapping band; in the subsequent power-on automatic calibration, the first-level integer dropout is turned on and fixed, in which case the overlapping band is swept, and the second-level integer dropout value and the digital local oscillator offset are calculated according to the sweep result; in the process of logically implementing the overlapping band calibration, the first-level dropout is performed when the sub-bands are spliced ​​and remains unchanged, and the second-level dropout is performed at the acquisition board of the system and needs to be recalculated each time the system is powered on. This method can achieve the goal of not needing to regenerate the phase-frequency calibration coefficient each time the system is powered on, based on the successful compensation correction of the overlapping band phase.

[0037] like Figure 1 As shown, specifically including the following:

[0038] S1, factory calibration steps;

[0039] S1.1. Determine the frequency range of the overlap band;

[0040] The frequency range of the overlapping band is determined based on the amplitude-frequency response characteristics of the transition band between two adjacent sub-bands. Specifically, the maximum amplitude-frequency difference between the same frequency points of the two sub-bands at the boundary of the overlapping band is first preset based on engineering experience. , in dB. The amplitude-frequency response of each sub-band transition band is calculated by sine frequency sweep, and then the amplitude difference between two sub-bands at the same frequency point is defined as:

[0041] (1)

[0042] in, The digital angular frequency is The amplitude value of the low-frequency subband signal is The digital angular frequency is The amplitude value of the high-frequency sub-band signal. At this time, the overlapping band frequency interval range should be satisfied.

[0043] (2)

[0044] The above formula shows that when the absolute value of the amplitude difference between the same frequency points of the two sub-bands is less than , the frequency point is within the overlapping band.

[0045] S1.2, overlapping band nonlinear phase difference extraction;

[0046] The frequencies within the overlapping band are divided into The corresponding overlapping band scanning sequence is , ω1 represents the first frequency sweep, and the sweep step frequency is The signal source inputs a sine sweep signal with different frequencies, and uses the sine three-parameter estimation algorithm or Fourier transform method to estimate the high-frequency sub-band phase for different frequencies. and the low frequency subband phase , and then calculate the phase difference of the same frequency point between sub-bands as:

[0047] (3)

[0048] S1.3, extracting the linear phase difference and the first-level dropout value within the overlapping band;

[0049] The optimal solution is:

[0050] (4)

[0051] in is an overlapping band sweep sequence, is the phase difference between sub-bands, k is the fitted linear phase difference slope, and θ is the linear phase difference offset.

[0052] Taking the least squares method as an example, we get The optimal solution ,as well as The optimal solution :

[0053] (5)

[0054] (6)

[0055] Then we get the linear phase difference , and the nonlinear phase difference , and the first-level loss point value ,like If it is negative, it means integer delay is performed in the high-frequency sub-band, otherwise it is performed in the low-frequency sub-band.

[0056] S2, automatic calibration steps at startup;

[0057] S2.1, overlap band data acquisition after the first level drop point is turned on;

[0058] Based on the primary dropout value obtained from factory calibration, there may be a fixed delay deviation between sub-band boards each time the system is powered on. , which is associated with a logical data store, where is the system sampling rate, is the clock frequency of the logic storage data. To correct the fixed delay deviation, set the sweep step frequency to After the first-level point loss, the signal source inputs a sine sweep signal with different frequencies, and the phase of the same frequency point between sub-bands is estimated for different frequencies using the sine three-parameter estimation algorithm or the Fourier transform method. and ; Then calculate the phase difference of the same frequency point between sub-bands as .

[0059] S2.2, Second level loss point calculation;

[0060] Referring to step S1.3, use the linear fitting algorithm to obtain the overlap band phase difference slope , then the secondary loss point value is calculated as follows: Indicates rounding down. It is a multiple of the system storage frequency and the sampling frequency. The calculation formula is as follows: , is the system sampling rate, The clock frequency at which the logic stores data.

[0061] (7)

[0062] S2.3, update digital local oscillator coefficients;

[0063] First, the following formula is used to calculate the phase difference after the overlapping band is lost;

[0064] (8)

[0065] The digital local oscillator coefficient of the high frequency sub-band is , where the initial value of the local oscillator phase is , the angular frequency is , is the number of sampling points, then the updated digital local oscillator coefficient is calculated according to formula (9).

[0066] (9)

[0067] The present invention uses a high-speed digital storage oscilloscope with a sampling rate of 80GSa / s and a bandwidth of 18GHz built based on broadband interleaved sampling technology and a signal source that can generate a sinusoidal signal with a frequency range of 20GHz. The signal source generates a sinusoidal swept frequency signal and inputs it into the oscilloscope. The oscilloscope collects the time domain signals of the two sub-bands without phase compensation at the overlapping band for analysis and calculation compensation.

[0068] like Figure 2 and Figure 3 As shown, the high-speed digital storage oscilloscope is restarted twice, and each time it is restarted, the method of the present invention and the method of only integer loss are used for calibration, and the overlap band phase difference curve after calibration is drawn; it can be seen that the phase difference curve after compensation by the method of the present invention is basically consistent, while the phase difference after calibration by the method of only first-order integer loss is different due to the change of analog local oscillator phase and the change of board synchronization state caused by the system power-on again, which leads to the change of group delay after the final splicing. Figure 4 and Figure 5 As shown, the system group delay response after overlapping band compensation of the present invention is basically consistent in two startups. The method of performing overlapping band compensation by only performing first-level integer loss will result in inconsistent system group delay after compensation.

[0069] like Figure 6As shown, in order to verify that the automatic calibration of the power-on overlapping band of the present invention will not affect the change of the system group delay (phase-frequency response), an overlapping band calibration is performed according to the method of the present invention when the power is turned on for the first time, and then a square wave signal with a fundamental frequency of 50MHz is input and the phase-frequency calibration coefficient is generated. Restart the system, compensate for the overlapping band phase difference in the second power-on state, and use the phase-frequency compensation filter generated for the first power-on. The illustrated calibration waveform after the second power-on is consistent with the reference waveform, which shows that the old phase-frequency calibration filter coefficient is still in effect. Therefore, using the method of the present invention to compensate for the overlapping band will not cause changes in the system group delay, and there is no need to calibrate the phase frequency again.

[0070] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used for various other combinations, modifications and improvements, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. An overlapping band phase compensation method based on secondary integer loss and phase shifting, characterized in that: The phase compensation method comprises: Factory calibration steps: Calculate the phase difference based on the overlap band sweep data, and use the linear fitting algorithm to fit the first-order integer loss point value and the overlap band nonlinear phase difference data according to the overlap band angular frequency and phase difference; Automatic calibration steps at power-on: sweep the overlap band, and calculate the secondary integer loss point value and digital local oscillator coefficient based on the sweep result; The factory calibration steps specifically include the following: Determine the frequency range of the overlapping band: preset the maximum amplitude frequency difference of the same frequency points of two sub-bands at the overlapping band boundary to be , and calculate the amplitude-frequency response of each sub-band transition band by sine sweep, setting the amplitude difference between the two sub-bands at the same frequency point to , and satisfies ,in, The digital angular frequency is The amplitude value of the low-frequency subband signal is The digital angular frequency is The amplitude value of the time-frequency sub-band signal; Overlapping band nonlinear phase difference extraction: The frequencies in the overlapping band are divided into N The corresponding overlapping band sweep sequence is , the sweep step frequency is The signal source inputs a sine sweep signal with different frequencies, and uses the sine three-parameter estimation algorithm or Fourier transform method to estimate the phase of the high-frequency sub-band for different frequencies. and the low frequency subband phase , and calculate the phase difference of the same frequency point between the high-frequency and low-frequency sub-bands as , ω 1 indicates the first sweep frequency; Extract the linear phase difference and first-order integer loss value within the overlapping band: Set the optimization formula , obtained by least squares method The optimal solution as well as The optimal solution They are , , and then the linear phase difference is obtained , and the nonlinear phase difference , and the first-level integer loss value ,like If is negative, it means integer delay in high frequency sub-band, otherwise in low frequency sub-band, where is the phase difference between sub-bands, k is the fitted linear phase difference slope, and θ is the linear phase difference offset; The power-on automatic calibration step specifically includes the following contents: After the first-level integer drop point is turned on, the overlapping band data is acquired: set the sweep step frequency to After first-level integer loss, the signal source inputs a sine sweep signal with different frequencies, and the phase of the same frequency point between sub-bands is estimated for different frequencies using the sine three-parameter estimation algorithm or the Fourier transform method. and , and calculate the phase difference of the same frequency point between sub-bands as ; Secondary integer loss point calculation: Use linear fitting algorithm to obtain the overlap band phase difference slope , and the secondary integer loss value is ,in, Indicates rounding down. Indicates the multiple of the system storage frequency and sampling frequency; Update the digital local oscillator coefficient: According to the formula Calculate the phase difference after the overlapping band is lost, and the digital local oscillator coefficient of the high-frequency sub-band is expressed as , then according to the formula The updated digital local oscillator coefficients are calculated, where the initial local oscillator phase value is , the angular frequency is , is the number of sampling points; The system includes a factory calibration unit and a power-on automatic calibration unit; The factory calibration unit calculates the phase difference based on the overlapping band frequency sweep data, and uses a linear fitting algorithm to fit the first-order integer loss point value and the overlapping band nonlinear phase difference data according to the overlapping band angular frequency and phase difference; The power-on automatic calibration unit is used to sweep the overlapping band and calculate the secondary integer loss point value and the digital local oscillator coefficient according to the sweep result; The factory calibration unit includes an overlapping band frequency range determination subunit, an overlapping band nonlinear phase difference extraction subunit, and an overlapping band range linear phase difference and first-level integer loss point value extraction subunit; The overlapping band frequency range determination subunit: presets the maximum amplitude frequency difference of the same frequency points of two sub-bands at the overlapping band boundary to be , and calculate the amplitude-frequency response of each sub-band transition band by sine sweep, setting the amplitude difference between the two sub-bands at the same frequency point to , and satisfies ,in, The digital angular frequency is The amplitude value of the low-frequency subband signal is The digital angular frequency is The amplitude value of the time-frequency sub-band signal; Overlapping band nonlinear phase difference extraction: The frequencies in the overlapping band are divided into N The corresponding overlapping band sweep sequence is , the sweep step frequency is The signal source inputs a sine sweep signal with different frequencies, and uses the sine three-parameter estimation algorithm or Fourier transform method to estimate the phase of the high-frequency sub-band for different frequencies. and the low frequency subband phase , and calculate the phase difference of the same frequency point between the high-frequency and low-frequency sub-bands as , ω 1 indicates the first sweep frequency; Extract the linear phase difference and first-order integer loss value within the overlapping band: Set the optimization formula , obtained by least squares method The optimal solution as well as The optimal solution They are , , and then the linear phase difference is obtained , and the nonlinear phase difference , and the first-level integer loss value ,like If is negative, it means integer delay in high frequency sub-band, otherwise in low frequency sub-band, where is the phase difference between sub-bands, k is the fitted linear phase difference slope, and θ is the linear phase difference offset; The power-on automatic calibration unit comprises a primary integer loss point start-up overlap band data acquisition subunit, a secondary integer loss point calculation subunit and a digital local oscillator coefficient update subunit; After the first-level integer loss point is turned on, the overlapping band data acquisition subunit: set the sweep frequency step frequency to After first-level integer loss, the signal source inputs a sine sweep signal with different frequencies, and the phase of the same frequency point between sub-bands is estimated for different frequencies using the sine three-parameter estimation algorithm or the Fourier transform method. and , and calculate the phase difference of the same frequency point between sub-bands as ; Secondary integer loss point calculation: Use linear fitting algorithm to obtain the overlap band phase difference slope , and the secondary integer loss value is ,in, Indicates rounding down. Indicates the multiple of the system storage frequency and sampling frequency; Update the digital local oscillator coefficient: According to the formula Calculate the phase difference after the overlapping band is lost, and the digital local oscillator coefficient of the high-frequency sub-band is expressed as , then according to the formula The updated digital local oscillator coefficients are calculated, where the initial local oscillator phase value is , the angular frequency is , is the number of sampling points.