A method of analog-to-digital local oscillator synchronization based on minimum mean square error
By calculating the initial phase compensation of the digital local oscillator as a nonlinear planning problem, the optimization algorithm and multi-frequency sweep method of the overlap band frequency range can realize the modulus local oscillator synchronization of the broadband interleaved sampling system, solving the nonlinear characteristics The synchronization problem below reduces the phase error between subbands and ensures the consistency of amplitude and frequency response.
Patent Information
- Application Number
- CN202411164580.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-08-23
AI Technical Summary
When the nonlinear characteristics of the existing broadband interwoven sampling system exist, the modulus local oscillator synchronization is difficult to achieve, resulting in the failure of the sampling sequence synchronization between subbands, and the overlap band phase difference may differ after the power is turned on, affecting the consistency of the amplitude and frequency response. .
The calculation of the initial phase compensation of the digital local oscillator is converted into a nonlinear planning problem, and the global optimal solution is obtained using the optimization algorithm. Through the multi-frequency sweep method of the overlap band frequency range, the nonlinear phase difference of the overlap band is calculated, and Based on the delay loss point, each subband is phase compensated to realize serial synchronization processing to realize modulo local oscillator synchronization.
Calculate the optimal initial phase compensation value of digital local oscillator under the minimum mean square error criterion, reduce the phase error between analog local oscillator and digital local oscillator, improve the nonlinear characteristic problem, ensure subband consistency compensation, and reduce overlap The influence of phase difference on amplitude-frequency response.
Smart Images

Figure CN118694364B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to signal processing, in particular to an analog-to-digital local oscillation synchronization method based on minimum mean square error. Background Art
[0002] With the rapid development of signal processing technology, electronic signals present high frequency, wide bandwidth, and high dynamic characteristics. Their complexity and transient characteristics also increase dramatically, resulting in a growing demand for broadband ultra-high-speed signal acquisition systems. The traditional time-interleaved sampling system architecture achieves a higher sampling rate by using multiple analog-to-digital converters (ADCs) for parallel acquisition. However, due to the limitations of the front-end analog bandwidth and the ADC's own bandwidth, this architecture cannot improve the input bandwidth of the entire system. In recent years, the broadband interleaved sampling system architecture has broken through the constraints of the time-interleaved sampling system on the input bandwidth. While increasing the sampling rate, it also expands the input bandwidth of the system. Specifically, the broadband interleaved sampling system introduces an analog down-conversion circuit before ADC acquisition, uses a mixer group to down-convert different high-frequency sub-band signals to a low-frequency range, and then samples them using the same ADC chip. Finally, the sampled data is processed and reconstructed in the digital domain to restore the input signal. The characteristic of this system architecture is that in the analog domain, an analog local oscillator is used to move the signal at a high frequency to a low frequency for time-interleaved sampling, and then in the digital domain, a digital local oscillator is used to restore the signal at a low frequency to a high frequency for reconstruction.
[0003] One of the important prerequisites for the broadband interleaved sampling system to achieve perfect signal reconstruction is the phase synchronization between the analog local oscillator and the digital local oscillator. This is because after analog down-conversion and digital up-conversion, the output signal of the system will introduce the initial phases of the analog local oscillator signal and the digital local oscillator signal. In theory, in order not to introduce additional phase errors, the initial phases of the digital local oscillator and the analog local oscillator between adjacent sub-bands should be consistent. However, due to reasons such as lack of synchronization between boards, the initial phases of the analog local oscillator and the digital local oscillator will be different, resulting in the inability to synchronize the sampling sequences between sub-bands.
[0004] Existing Solution 1: The patent "A Method for Synchronizing Analog-to-Digital Local Oscillators in Digital Bandwidth Alternating Systems" with application number CN202010834613.0 provides a solution for synchronization of analog-to-digital local oscillators, but this method is extremely dependent on the linear characteristics of the system. When the system has a certain degree of nonlinear characteristics, the initial phases of the digital local oscillators that need to be compensated at different frequency points are different. If a single-frequency signal is still used for analog-to-digital local oscillator synchronization processing, the system will introduce a large analog-to-digital local oscillator phase error at other frequency points, which will in turn affect the sub-band consistency compensation at the back end. Therefore, this method cannot meet the needs of broadband interleaved acquisition systems with a certain degree of nonlinear characteristics; Existing Solution 2: The paper "Research and Implementation of Correction Algorithms for Ultra-Wideband Acquisition Systems Based on Subband Decomposition" improves the nonlinear characteristics problem to a certain extent, but the overlapping band phase difference after analog-to-digital local oscillator synchronization after each startup may have a certain degree of difference, resulting in changes in the amplitude-frequency response after splicing near the center frequency of the overlapping band. Summary of the invention
[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an analog-to-digital local oscillator synchronization method based on minimum mean square error, which converts the initial phase compensation calculation of the digital local oscillator of each sub-band into a nonlinear programming problem, and then uses the optimization algorithm involved to obtain the global optimal solution, and finally uses the global optimal solution to calculate the initial phase of the digital local oscillator of the corresponding sub-band to achieve analog-to-digital local oscillator synchronization.
[0006] The object of the present invention is achieved through the following technical scheme: a method for modulus local oscillation synchronization based on minimum mean square error, comprising the following steps:
[0007] S1. For a broadband interleaved sampling system comprising multiple sub-bands, determining the overlapping band frequency range between adjacent sub-bands;
[0008] S2. Obtain the amplitude-frequency response and phase-frequency response of different frequency points between two adjacent sub-bands, and calculate the nonlinear phase difference of the overlapping band;
[0009] S3. Perform overlapping phase compensation on each sub-band based on delay loss point;
[0010] S4. Calculate the phase difference and initial phase compensation value of the overlapping band after the delay loss point, and realize the analog-to-digital local oscillator synchronization of the entire broadband interleaved sampling system based on the serial synchronization processing framework.
[0011] The beneficial effects of the present invention are: (1) The present invention uses a multi-frequency point scanning method in an overlapping frequency band range, and can calculate the optimal digital local oscillator initial phase compensation value under the minimum mean square error criterion, which not only improves the problem of nonlinear characteristics of the system to a certain extent, but also can minimize the phase error between the analog local oscillator and the digital local oscillator, laying a favorable foundation for the sub-band consistency compensation at the back end.
[0012] (2) The present invention sets the digital local oscillator value to be compensated as the average of the maximum and minimum phase differences after the delay point loss, and sets the ideal value after analog-to-digital local oscillator synchronization to zero or the nonlinear phase difference of the overlapping band. After the analog-to-digital local oscillator synchronization of the oscilloscope at different power-on times, the amplitude-frequency curves of the sub-bands after splicing are closer, indicating that the influence of the phase difference between the sub-bands on the amplitude after splicing has been effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 It is a schematic diagram of the synchronization process of the present invention;
[0014] Figure 2 It is a schematic diagram of the architecture of a broadband interleaved sampling system;
[0015] Figure 3 Schematic diagram of phase difference and amplitude-frequency response before overlapping band compensation;
[0016] Figure 4 Schematic diagram of phase difference after synchronization under different conditions;
[0017] Figure 5 Schematic diagram of the amplitude-frequency response after synchronization in different situations. DETAILED DESCRIPTION
[0018] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.
[0019] like Figure 1 As shown, a method for analog-to-digital local oscillator synchronization based on minimum mean square error includes the following steps:
[0020] S1. For a broadband interleaved sampling system comprising multiple sub-bands, determining the overlapping band frequency range between adjacent sub-bands;
[0021] Taking a system consisting of 4 sub-bands as an example, the broadband interleaved sampling system is Figure 2 As shown, each sub-band corresponds to one processing channel. Due to the roll-off characteristics of the analog filter, there will be a repeated frequency range between adjacent sub-bands. This range is called the overlapping band. The signal within the frequency range of the overlapping band will be output in two adjacent sub-bands at the same time.
[0022] The overlapping frequency interval is determined according to the amplitude-frequency response characteristics of the transition band between two adjacent sub-bands. Specifically, the amplitude-frequency response overlapping band threshold between two sub-bands is set to , in dB. Input sinusoidal test signals of different frequencies by frequency sweeping, and calculate the i The frequency response of the sub-band is:
[0023] (1)
[0024] in, and Respectively i The amplitude-frequency response and phase-frequency response of each sub-band. It means to find the phase angle;
[0025] Calculate the p Sub-band and p +1 sub-band amplitude difference at different frequency points , in dB, we get:
[0026] (2)
[0027] No. p Sub-band and p The frequency range repeated between the +1 sub-bands is defined as p Overlapping band, p The set of overlapping frequency points C is defined as:
[0028] (3)
[0029] S2. Obtain the amplitude-frequency response and phase-frequency response of different frequency points between two adjacent sub-bands, and calculate the nonlinear phase difference of the overlapping band;
[0030] The output signal of the broadband interleaved sampling system is obtained by step frequency sweeping. During the sine frequency sweeping process, for each frequency point of the sine frequency sweeping signal input to the broadband interleaved sampling system, each sub-band has a corresponding output signal;
[0031] For any two adjacent sub-bands corresponding to an overlapping band, the Fourier transform method is used to estimate the different frequency points between the two adjacent sub-bands. , The amplitude-frequency response and And the phase frequency response and According to formula (3), find the frequency points within the overlapping band and calculate the phase difference of the same frequency points in adjacent sub-bands :
[0032] (4)
[0033] in, N is the number of overlapping frequency points.
[0034] Phase Difference Untangle and get the phase difference after untangle , use the least squares method to fit a straight line to calculate the linear phase difference between adjacent sub-bands in the overlapping band , which can be expressed as:
[0035] (5)
[0036] in, Indicates calculation of average value; It is the phase difference slope introduced by the different linear delays between the two sub-bands in the overlapping band. b It is the initial phase part of the phase difference that does not change with frequency and is composed of the analog local oscillator and the digital local oscillator.
[0037] Finally, calculate the nonlinear phase difference of N frequency points , the calculation formula is:
[0038] (6)
[0039] For the two sub-bands corresponding to each overlapping band, the above steps are repeated to obtain the nonlinear phase difference of each frequency point in each overlapping band.
[0040] S3. Perform overlapping phase compensation on each sub-band based on delay loss point;
[0041] The phase compensation of the overlapping band includes delay loss and phase shift. Due to the different delay times of the sub-band signals and the nonlinear characteristics introduced by the mixing filter, the phase difference of the overlapping band includes nonlinear phase difference, fixed phase difference of the analog local oscillator and linear phase difference. ,in is the delay difference between the two sub-bands. The linear phase difference and the modulus local oscillator phase difference can be achieved by delay loss and phase shift, respectively. In order to simplify the phase compensation operation of the overlapping band, the present invention uses the least squares method to calculate the delay loss value, that is, the phase difference after unwrapping the same frequency point of the adjacent sub-band Substituting into equation (5) we can calculate the slope of the phase difference , then The integer part of the delay loss value , the compensated overlapping band phase difference is:
[0042] (7)
[0043] Then Perform digital local oscillator initial phase processing.
[0044] S4. Calculate the phase difference and initial phase compensation value of the overlapping band after the delay loss point, and realize the analog-to-digital local oscillator synchronization of the entire broadband interleaved sampling system based on the serial synchronization processing framework.
[0045] Calculation of the phase difference of the overlapping band after delay loss:
[0046] Suppose the signal of the mth frequency point after the delay and loss of the i-th subband is for:
[0047] (8)
[0048] in, Indicates i Sub-band m frequency points, n represents discrete time; For the i Sub-band m The amplitude of each frequency point; For the i The initial phase of the simulated local oscillator of the sub-band, For the i The initial phase of the digital local oscillator of the sub-band; For the i Sub-band m The nonlinear phase deviation caused by the nonlinear characteristics of the frequency point broadband interleaved sampling system;
[0049] Using Fast Fourier Transform to get the signal The initial value of , define p Overlapping bands ( p Sub-band and p + 1 sub-band) is:
[0050] (9)
[0051] Calculation of initial phase compensation value of digital local oscillator
[0052] (1) Construction of target optimization function
[0053] Assume that the initial phase value of the digital local oscillator to be compensated is , under the minimum mean square error criterion, the objective optimization function is constructed as:
[0054] (10)
[0055] Indicates constraints;
[0056] in,
[0057] (11)
[0058] For the pThe ideal value of the phase difference of the overlapping bands. Due to the changes in the phase state of the analog local oscillator and the digital local oscillator and the inconsistent delay of the sub-band signal path each time the machine is turned on (each sub-band signal passes through a different mixer), that is, the randomness of the machine time, this leads to differences in the phase difference curve after removing the linear phase difference each time. In order to ensure that the phase difference of the overlapping band after each phase compensation is similar when other conditions remain unchanged, the concept of ideal value is introduced, and the calculation of this value will be described below.
[0059] (2) Calculation of ideal phase difference in overlapping band
[0060] There are two main methods for obtaining the ideal value of the overlapping band phase difference. One is to set the ideal value to 0, and the other is to use the nonlinear phase difference of the overlapping band. The specific operation of the second method is: by collecting the output signal of the overlapping band before phase compensation after the oscilloscope is turned on for many times, the phase difference of adjacent sub-bands at different frequency points is obtained, and the estimated values of the nonlinear phase deviation are calculated according to equations (4) to (6) , statistical overlap band The distribution of The sum and average result is taken as the p The ideal value of the phase difference of the overlapping band .
[0061] (3) Calculation of initial phase compensation value of digital local oscillator
[0062] For formula (10), the interval Divided into Q equal parts, the interval is , the value space of the initial phase of the digital local oscillator to be compensated is ,Will Substituting into equation (10), we can find the optimal solution that minimizes the objective function. value, and As the initial phase value of the digital local oscillator of the p+1th sub-band to be compensated .
[0063] Serial synchronization processing framework construction
[0064] Serial synchronization processing means: first, the digital local oscillator initial phase of the second sub-band is compensated according to the first and second sub-bands, so that the second sub-band reaches the analog-to-digital local oscillator synchronization state; then, the digital local oscillator initial phase of the third sub-band is compensated according to the second and third sub-bands, so that the third sub-band reaches the analog-to-digital local oscillator synchronization state; and so on, the third and fourth sub-bands compensate the digital local oscillator initial phase of the fourth sub-band, so that the fourth sub-band reaches the analog-to-digital local oscillator synchronization state. At this point, the analog-to-digital local oscillator synchronization processing of the entire broadband interleaved sampling system is completed in a serial manner.
[0065] According to the idea of the analog-to-digital local oscillator serial synchronization processing framework, the steps for each sub-band to achieve analog-to-digital local oscillator synchronization are divided into the following steps, which will be described in detail below:
[0066] Step 1: Initialization, let p = 1, that is, the overlapped band 1 is first subjected to analog-to-digital local oscillator synchronization processing. In particular, in the broadband interleaved sampling system, subband 1 does not need mixing, so it can be directly considered that subband 1 has reached the analog-to-digital local oscillator synchronization state, that is, .
[0067] Step 2: Loop and calculate the phase difference between adjacent sub-bands in each overlapping band according to formula (9): .
[0068] Step 3: Solve the problem according to the methods described in (1) to (3). p +1 sub-band digital local oscillator initial phase compensation value .
[0069] Step 4: Calculate the final initial phase of the local oscillator signal of the p+1th subband. p +1 sub-band for phase shifting. The phase value to be shifted is . Order No. p The initial phase of the +1 subband is , the initial phase value to be compensated is , and then increase the phase of the digital local oscillator of all sub-bands , and get the final p The initial phase of the local oscillator signal of the +1 sub-band is , which can generate the p +1 digital local oscillator signal of sub-band.
[0070] Step 5: Judgement p Is it greater than 3? If it is greater than 3, the analog-to-digital local oscillator synchronization processing of the entire system is completed; otherwise, let p = p +1, the step jumps to step 2, and continues with steps 2 to 5 until p If it is greater than 3, the loop ends.
[0071] In the embodiments of the present application, the technical effects of the present invention are described in detail below in combination with a simulation test in a certain scenario:
[0072] 1. Test conditions and contents
[0073] The experimental equipment is an oscilloscope and a signal waveform generator with a sampling rate of 80 GSa / s and a bandwidth of 18 GHz based on a broadband interleaved sampling system. Assuming that the 1st, 2nd and 3rd sub-bands of the broadband interleaved sampling system have been synchronized, the following takes the analog-to-digital local oscillator synchronization in the overlapping bands corresponding to the 3rd and 4th sub-bands as an example for detailed explanation.
[0074] A signal waveform generator is used to generate a sinusoidal test signal of equal amplitude and interval to the oscilloscope. The frequency range is 15.5 GHz to 16.5 GHz and the frequency interval is 10 MHz. It is known that the initial phase of the digital local oscillator between sub-bands 1 and 2 and between sub-bands 2 and 3 needs to be compensated. and , the initial phase of the 3rd and 4th sub-band digital local oscillator is and .
[0075] 2. Test result analysis
[0076] Reference Figure 3 After the oscilloscope is powered on three times, the phase difference of the overlapping band is different at different power-on times, and the different phase differences of adjacent sub-bands in the overlapping band will affect the amplitude-frequency response of the overlapping band after splicing. The reason for this phenomenon is the change in the phase state of the analog local oscillator and the digital local oscillator and the inconsistent delay of the sub-band signal path each time the power is turned on. Therefore, in order to make the amplitude-frequency response of the overlapping band after splicing consistent after each power-on, the phase difference of the overlapping band should be corrected.
[0077] Reference Figure 4 The phase correction of the overlapping band includes linear phase removal (integer loss point) and digital local oscillator initial phase compensation. The new overlapping band phase difference data after compensation is obtained by using the integer loss point. Under the minimum mean square error criterion, the compensation value of the digital local oscillator initial phase is obtained by solving the target optimization function. ,Will As the new digital local oscillator initial phase to meet the requirements of analog-to-digital local oscillator synchronization in the system. Figure 4 It can be seen that after the overlapping band phase compensation, the phase difference between adjacent sub-bands in the overlapping band is between -0.4768 and 0.7965 rad / s.
[0078] Reference Figure 5 , under the action of the new digital local oscillator initial phase, the signal after the two sub-bands are spliced is obtained again, and the amplitude-frequency response after splicing is obtained according to Fourier transform. Theoretically, for the same machine, after the linear phase difference of the overlapping bands is eliminated and the analog-to-digital local oscillator is synchronized at different startup times, the amplitude-frequency response of the overlapping bands after the sub-bands are spliced should be consistent. From the figure, it can be observed that the traditional method of setting the digital local oscillator compensation value to the average of the maximum and minimum values of the phase difference after the delay loss point and the method of setting the ideal value to zero or the nonlinear phase difference of the overlapping band, the amplitude-frequency response of the overlapping bands after the sub-bands are spliced is almost the same at the three startup times of the oscilloscope, indicating that all three methods effectively reduce the influence of the phase difference of the overlapping band on the amplitude-frequency response after splicing, which lays a favorable foundation for the subsequent full-pass amplitude-frequency compensation.
[0079] Table 1 Comparison of RMS errors of different modulus local oscillator synchronization methods
[0080] First boot and second boot Second and third startup First and third startup average Existing Solution 2 0.1005 0.1248 0.0597 0.0950 The ideal value is zero 0.0709 0.1129 0.0604 0.0841 The present invention 0.0616 0.0901 0.0605 0.0707
[0081] Referring to Table 1, after overlapping band phase compensation and analog-to-digital local oscillator synchronization processing, the average root mean square error of the traditional method is 0.0950, while when the ideal phase difference is zero, the average root mean square error is 0.0841, and the root mean square error of the present invention is the smallest, indicating that the method of setting the ideal value as the nonlinear phase difference further improves the effect of overlapping band compensation.
[0082] The above is 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 in other combinations, modifications and environments, 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 depart 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. A method for modulus local oscillator synchronization based on minimum mean square error, characterized in that: The following steps are involved: S1. For a broadband interleaved sampling system comprising multiple sub-bands, determining the overlapping band frequency range between adjacent sub-bands; S2. Obtain the amplitude-frequency response and phase-frequency response of different frequency points between two adjacent sub-bands, and calculate the nonlinear phase difference of the overlapping band; S3. Perform overlapping phase compensation on each sub-band based on delay loss point; S4. Calculate the phase difference and initial phase compensation value of the overlapping band after the delay loss point, and realize analog-to-digital local oscillator synchronization of the entire broadband interleaved sampling system based on the serial synchronization processing framework; Calculating the phase difference of the overlapping band after delay loss includes: Suppose the signal of the mth frequency point after the delay and loss of the i-th subband is for: (8) in, , Indicates i Sub-band m frequency points, n represents discrete time; For the i Sub-band m The amplitude of each frequency point; For the i The initial phase of the simulated local oscillator of the sub-band, For the i The initial phase of the digital local oscillator of the sub-band; For the i Sub-band m The nonlinear phase deviation caused by the nonlinear characteristics of the frequency point broadband interleaved sampling system; Using Fast Fourier Transform to get the signal The initial value of , define p The phase difference of the overlapping bands is: (9) Among them, p The overlapping zone refers to the p Sub-band and p + 1 The frequency range repeated between sub-bands; Calculating the initial phase compensation value of the digital local oscillator includes: (1) Construction of target optimization function Assume that the initial phase value of the digital local oscillator to be compensated is , under the minimum mean square error criterion, the objective optimization function is constructed as: (10) Indicates constraints; in, (11) For the p The ideal value of the phase difference of the overlapping bands; (2) Calculation of ideal phase difference in overlapping band There are two methods to determine the ideal value of the overlap band phase difference: One is to set the ideal value to 0; The second is to use the nonlinear phase difference of the overlapping band. The specific process is: by repeatedly collecting the output signal of the overlapping band before phase compensation after the oscilloscope is turned on, the phase difference of adjacent sub-bands at different frequency points is obtained, and the estimated value of the nonlinear phase deviation is calculated multiple times. , statistical overlap band The distribution of The sum and average result is taken as the p The ideal value of the phase difference of the overlapping band ; Indicates the different frequencies between two adjacent sub-bands. , N is the number of overlapping band frequencies; (3) Calculation of initial phase compensation value of digital local oscillator For formula (10), the interval Divided into Q equal parts, the interval is , the value space of the initial phase of the digital local oscillator to be compensated is ,Will Substituting into equation (10), we can find the optimal solution that minimizes the objective function. value, and As the initial phase value of the digital local oscillator of the p+1th sub-band to be compensated .
2. The method for analog-to-digital local oscillation synchronization based on minimum mean square error according to claim 1, characterized in that: In step S1, the overlapping band refers to a repeated frequency range between adjacent sub-bands, and the signal within the frequency range in the overlapping band will be output in two adjacent sub-bands at the same time.
3. The method for analog-to-digital local oscillation synchronization based on minimum mean square error according to claim 1, characterized in that: The step S1 comprises: S101. Set the overlap threshold of the amplitude-frequency response between two sub-bands to , the unit is dB; Input sinusoidal test signals of different frequencies by frequency sweeping and calculate the i Subband frequency response ,get: (1) in, and Respectively i The amplitude-frequency response and phase-frequency response of each sub-band; It means to find the phase angle; S102. Calculate the p Sub-band and p +1 sub-band at different frequencies The amplitude difference , in dB, we get: (2) No. p Sub-band and p The frequency range repeated between the +1 sub-bands is defined as p Overlapping band, p The set of overlapping frequency points C is defined as: (3) in, Represents the number of subbands in a wideband interleaved sampling system.
4. The method for analog-to-digital local oscillation synchronization based on minimum mean square error according to claim 3, characterized in that: The step S2 comprises: S201. Obtain the output signal of the broadband interleaved sampling system by sine frequency sweep: During the sine frequency sweep process, for each frequency point of the sine frequency sweep signal input to the broadband interleaved sampling system, each sub-band has a corresponding output signal; For any two adjacent sub-bands corresponding to an overlapping band, the Fourier transform method is used to estimate the different frequency points between the two adjacent sub-bands. , The amplitude-frequency response and And the phase frequency response and , find the frequency points within the overlapping band and calculate the phase difference of the same frequency points in adjacent sub-bands : (4) in, N is the number of overlapping band frequencies; S202. Phase difference Untangle and get the phase difference after untangle , use the least squares method to fit a straight line to calculate the linear phase difference between adjacent sub-bands in the overlapping band , expressed as: (5) in, Indicates calculation of average value; It is the phase difference slope introduced by the different linear delays between the two sub-bands in the overlapping band. It is the initial phase part of the phase difference that does not change with frequency and is composed of the analog local oscillator and the digital local oscillator. Finally, calculate: (6) in, For the j Nonlinear phase difference of each frequency point; S203. Repeat steps S201 to S202 for each of the two sub-bands corresponding to the overlapping band.
5. The method for analog-to-digital local oscillation synchronization based on minimum mean square error according to claim 4, characterized in that: The step S3 comprises: The least squares method is used to calculate the delay loss value, that is, the phase difference after unwrapping the same frequency points of adjacent sub-bands Substituting into equation (5) we can calculate the slope of the phase difference , then The integer part of the delay loss value ,get: (7) in, It is the phase difference of overlapping band after delay loss compensation.
6. The method for analog-to-digital local oscillator synchronization based on minimum mean square error according to claim 1, characterized in that: In step S4, based on the serial synchronization processing framework, the analog-to-digital local oscillator synchronization of the entire broadband interleaved sampling system is realized, including: Step 1: Initialization, let p = 1, that is, the overlap band 1 is first subjected to analog-to-digital local oscillator synchronization processing. In the broadband interleaved sampling system, subband 1 does not need to be mixed, so it is directly considered that subband 1 has reached the analog-to-digital local oscillator synchronization state, that is, ; Step 2: Loop and calculate the phase difference between adjacent sub-bands in each overlapping band according to formula (9): ; Step 3: Solve p +1 sub-band digital local oscillator initial phase compensation value ; Step 4: Calculate the final initial phase of the local oscillator signal of the p+1th subband: p +1 sub-band for phase shifting, so that p The initial phase of the +1 subband is , the initial phase value to be compensated is , and then increase the phase of the digital local oscillator of all sub-bands , and get the final p The initial phase of the local oscillator signal of the +1 sub-band is , thus generating the p +1 sub-band digital local oscillator signal; Step 5: Judgement p Is it greater than , if greater than , then the analog-to-digital local oscillator synchronization processing of the entire system is completed; otherwise, let p = p +1, the step jumps to step 2, and continues with steps 2 to 5 until p Greater than , ending the loop.
