SAR ADC nonlinear calibration algorithm and device based on harmonic distortion component extraction

Abstract

The invention discloses an SAR ADC nonlinear calibration algorithm and device based on harmonic distortion component extraction. The SAR ADC nonlinear calibration algorithm comprises the steps of obtaining a digital output matrix of an SAR ADC; separating the digital output matrix according to columns to construct a new digital output matrix; calculating a digital summation vector according to the ideal weight vector and the new digital output matrix; calculating a normalized frequency of the plurality of odd harmonics; for each odd harmonic, the method comprises the following steps: carrying out decimation filtering on a digital summation vector according to a normalized frequency to obtain a harmonic vector; constructing a new linear combination expression of a digital output matrix and a harmonic vector according to the to-be-estimated harmonic weight vector; solving the linear combination expression to obtain a to-be-estimated harmonic weight vector; calculating a total to-be-estimated harmonic weight vector according to the to-be-estimated harmonic weight vectors of all odd harmonics; and calculating a calibration weight vector according to the ideal weight vector and the total to-be-estimated harmonic weight vector. According to the invention, no circuit needs to be embedded, the portability is high, the hardware overhead is small, the calibration time is short, and the calibration precision is high.

Description

Technical Field

[0001] The invention belongs to the technical field of integrated circuits, and in particular relates to a SAR ADC nonlinear calibration algorithm and a device based on harmonic distortion component extraction. Background Art

[0002] Analog-to-Digital Converter (ADC) is an essential module for converting continuous analog signals into discrete digital signals. There are many types of ADCs, but the Successive Approximation Register ADC (SAR ADC) has been widely used in portable consumer electronics, medical equipment, industrial automation, communication systems, sensor interfaces, and digital acquisition due to its unique advantages such as simple structure, small area, low power consumption, and more suitable for the continuous evolution of advanced processes. In the above application fields, higher requirements are placed on the performance of SAR ADC. The capacitor-based SAR ADC obtains the digital code corresponding to the input signal by successively comparing the input voltage signal and the voltage value generated by the capacitor array from high to low. The voltage value generated by the capacitor array is determined by the comparison result obtained. The accuracy of SAR ADC is limited by the noise of each module and the mismatch of capacitors. The noise can be achieved through circuit design methods, but the capacitor mismatch is more complicated and difficult to handle.

[0003] Capacitor mismatch caused by CMOS manufacturing variations is a key factor that must be considered. It causes nonlinear problems in the Digital-to-Analog Converter (DAC), which in turn causes errors in the comparator; these errors produce harmonic distortion in the spectrum, which degrades the overall performance. Dynamic Element Matching (DEM) is a method that can be used to reduce mismatch errors, but DEM is expensive to implement in high-resolution designs because it requires thermometer encoding of the calibration DAC, which requires complex control logic. Although DAC mismatch errors can also be shaped by Mismatch Error Shaping (MES) technology, applying the reverse least significant bit (LSB) operation sacrifices 6dB of dynamic range. Noise Shaping (NS) technology requires embedding a noise shaping module and a DAC calibration module in the ADC. Although noise shaping is achieved in-band, it adds additional area power consumption and makes the timing more complex. The offset double-injection digital least mean square (LMS) algorithm calibration scheme requires two independent and biased N-bit SAR ADCs and an embedded digital calibration module, which requires a large area, has high design complexity and poor portability.

[0004] Accurate calibration technology can reduce the power consumption of ADC when it is working, improve the energy efficiency of the system, and reduce the deviation of ADC under different working conditions, thereby improving the stability, reliability and accuracy of the system. Therefore, it is urgent to design an off-chip digital calibration solution that does not require the circuit module to be embedded in the SAR ADC, has strong portability, small area and high calibration accuracy. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a SAR ADC nonlinear calibration algorithm and device based on harmonic distortion component extraction. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction, the calibration algorithm comprising:

[0007] Quantize the input signal using SAR ADC to obtain a digital output matrix;

[0008] Separating the digital output matrix by columns to construct a new digital output matrix;

[0009] Calculating a digital summation vector based on the ideal weight vector and the new digital output matrix;

[0010] Normalizing the frequency of the input signal to obtain normalized frequencies of multiple odd harmonics;

[0011] For each odd harmonic, the execution steps include: extracting and filtering the digital summation vector according to the normalized frequency of the current odd harmonic to obtain the harmonic vector corresponding to the current odd harmonic; constructing a linear combination expression of the new digital output matrix and the harmonic vector according to the harmonic weight vector to be estimated; solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic;

[0012] Calculating a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics;

[0013] A calibration weight vector is calculated according to the ideal weight vector and the total harmonic weight vector to be estimated, and the calibration weight vector is returned to the SAR ADC.

[0014] In one embodiment of the present invention, the input signal is a standard sine wave signal.

[0015] In one embodiment of the present invention, the new digital output matrix constructed is expressed as follows:

[0016] D new =[D0,…,D n ,…,D N-1 ];

[0017] Among them, D new Represents the new digital output matrix, D n It represents the n+1th vector in the new digital output matrix, that is, all elements in the n+1th column of the digital output matrix. The value of n ranges from 0 to N-1, and N represents the number of conversion bits of the SAR ADC.

[0018] In one embodiment of the present invention, the digital sum vector is calculated, and the formula is expressed as:

[0019]

[0020] Among them, D sum represents the digital sum vector, D n Represents the n+1th vector in the new digital output matrix, consisting of 2 0 , 2 1 , ..., 2 N-1 Construct the ideal weight vector.

[0021] In one embodiment of the present invention, the plurality of odd harmonics include 3rd harmonic, 5th harmonic, 7th harmonic and 9th harmonic.

[0022] In one embodiment of the present invention, the digital summation vector is subjected to decimation filtering according to the normalized frequency of the current odd harmonics, comprising:

[0023] The digital sum vector is subjected to decimation and filtering processing according to the normalized frequency of the current odd harmonics by using the fir1 function and the filter function, and the group delay caused by the decimation is removed.

[0024] In one embodiment of the present invention, the linear combination expression constructed is expressed as follows:

[0025]

[0026] Among them, D new Represents the new digital output matrix, W k Indicates the harmonic weight vector to be estimated corresponding to the current odd harmonic, k ranges from 1 to K, K represents the number of odd harmonics, Y k Indicates the harmonic vector corresponding to the current odd harmonic, Y′ k =D new W k .

[0027] In one embodiment of the present invention, solving the linear combination expression includes:

[0028] The linear combination expression is solved using the least squares method.

[0029] In one embodiment of the present invention, the calibration algorithm further comprises:

[0030] Calculate the level signal before calibration according to the new digital output matrix and the ideal weight vector;

[0031] Calculate a calibrated level signal according to the new digital output matrix and the calibration weight vector;

[0032] Perform FFT processing on the level signal before calibration and the level signal after calibration respectively, compare the two FFT processing results, and determine whether to return the calibration weight vector to the SAR ADC according to the comparison result.

[0033] In a second aspect, an embodiment of the present invention provides a SAR ADC nonlinear calibration device based on harmonic distortion component extraction, the calibration device comprising:

[0034] A quantization module, used to quantize the input signal using the SAR ADC to obtain a digital output matrix;

[0035] A construction module, used for separating the digital output matrix by columns to construct a new digital output matrix;

[0036] A first calculation module, used for calculating a digital sum vector according to an ideal weight vector and the new digital output matrix;

[0037] A normalization module, used for normalizing the frequency of the input signal to obtain normalized frequencies of multiple odd harmonics;

[0038] The estimation module is used for each odd harmonic, and the execution steps include: extracting and filtering the digital summation vector according to the normalized frequency of the current odd harmonic to obtain the harmonic vector corresponding to the current odd harmonic; constructing a linear combination expression of the new digital output matrix and the harmonic vector according to the harmonic weight vector to be estimated; solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic;

[0039] A second calculation module is used to calculate a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics;

[0040] The third calculation module is used to calculate a calibration weight vector according to the ideal weight vector and the total harmonic weight vector to be estimated, and return the calibration weight vector to the SAR ADC.

[0041] Beneficial effects of the present invention:

[0042] The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction proposed in the present invention is a method that can quickly solve the weights required for SAR ADC nonlinear calibration using only one SAR ADC. It takes the ideal binary weight as the real weight, and constantly approaches the real weight value during the calibration process, and constructs a method for estimating the harmonic weight according to the harmonic component. In the calibration process, the influence of different harmonic components on the weight is extracted, and then the calibrated weight is calculated according to the estimated harmonic weight. The amount of calculation is small, and the linear combination expression is solved by mathematical algorithms. The speed of solution is fast. In general, the present invention reduces the harmonic distortion and nonlinearity caused by SAR ADC capacitor mismatch, and improves the conversion accuracy of SAR ADC; the present invention does not need to embed any circuit modules in the SAR ADC, such as digital calibration modules, noise shaping modules, dynamic element matching modules, etc., and has high portability, low hardware overhead, short calibration time, high calibration accuracy, and has a wider application prospect.

[0043] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1It is a flow chart of a SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction provided by an embodiment of the present invention;

[0045] Figure 2 is a structural schematic diagram of a test system provided by an embodiment of the present invention;

[0046] Figure 3 It is a structural schematic diagram of a SAR ADC nonlinear calibration device based on harmonic distortion component extraction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0048] The conversion accuracy of SAR ADC is mainly affected by non-ideal factors such as mismatch, parasitics and noise of unit capacitors in DAC, among which unit capacitor mismatch has the greatest impact. In the method of improving ADC conversion accuracy, digital calibration is more convenient and can realize more complex functions. The existing digital calibration needs to embed relevant modules in the circuit to realize the function, such as: embedding noise shaping, digital calibration and DAC calibration modules. Based on the above situation, an embodiment of the present invention proposes a SAR ADC nonlinear calibration algorithm and device. The algorithm does not need to embed any circuit modules in the SAR ADC, so it can be applied to most SAR ADCs. Specifically:

[0049] First, see Figure 1 The embodiment of the present invention provides a SARADC nonlinear calibration algorithm based on harmonic distortion component extraction, which is applied to Figure 2 The test system shown in the figure has a SAR ADC integrated in Figure 2 In the test board, the SAR ADC has an internal register that is used to store the calibrated weights. The signal source and other test equipment input signals to the SAR ADC, and the SAR ADC quantizes the output data on the test version. Then, on the algorithm executable device, such as a desktop computer or a laptop, the quantized output data is directly used to implement calibration. The specific calibration algorithm includes:

[0050] S10. quantize the input signal using the SAR ADC to obtain a digital output matrix.

[0051] In the embodiment of the present invention, a signal is first input to the SAR ADC by a test device such as a signal source. The input signal form is not limited, and a standard sine wave signal is preferably input for testing. The input signal is quantized in the SAR ADC to obtain a digital output matrix. The digital output matrix represents multiple samplings including information of the input signal, harmonic distortion components and noise. Due to the influence of harmonics, noise and other factors on the circuit, the quantization result of each sampling may be different.

[0052] If the conversion bit number of the SAR ADC is N, the number of sampling times on the SAR ADC is M, and after the M samples are quantized by the SAR ADC, an M×N digital output matrix is ​​formed, and each element in the digital output matrix is ​​0 or 1.

[0053] A five-bit standard SAR ADC with a range of 3.3V is used as an example to illustrate the implementation of the algorithm proposed in the present invention, but the algorithm proposed in the present invention is not limited to the calibration of the above five-bit standard SAR ADC, and can be applied to most SAR ADCs.

[0054] For example, if a 5-bit standard SAR ADC with a range of 3.3V is used to measure a 1.65V DC voltage, assuming that there is no redundant bit, the quantization result is 00001, with the highest bit on the right and the lowest bit on the left. Taking an ideal square wave signal as an example, its valley voltage is 0V and its peak voltage is 3.3V. The above SAR ADC quantization method is used to measure the input square wave signal. Specifically:

[0055] For example, in an ideal situation, that is, when the square wave signal is not affected by harmonic components and noise, the digital output matrix formed by 6 samplings and quantization by a 5-bit standard SAR ADC is: Each row represents the result of a sampling after quantization by a five-bit standard SAR ADC. The SAR ADC quantization output at the valley is 00000, and the SAR ADC quantization output at the peak is 11111. If the input square wave signal is affected by harmonic components and noise, the SAR ADC quantization output at the valley may no longer be 00000, but 10000. Similarly, the SAR ADC quantization output at the peak may no longer be 11111, but 01111. Then the digital output matrix formed by the quantization of 6 samples by the five-bit standard SAR ADC becomes The changes of "1" and "0" in the digital output matrix reflect the influence of harmonic distortion components and noise on the input square wave signal.

[0056] S20, separating the digital output matrix by columns to construct a new digital output matrix.

[0057] In the embodiment of the present invention, the digital output matrix obtained in S10 is separated by columns, and the digital output matrix is ​​re-expressed to obtain a new digital output matrix, which is expressed as follows:

[0058] D new =[D0,…,D n ,…,D N-1 ] (1);

[0059] Among them, D new Represents the new digital output matrix, D n represents the n+1th vector in the new digital output matrix, that is, all elements in the n+1th column of the digital output matrix, where n ranges from 0 to N-1, D n is an M×1 vector, D n Each element in is either 0 or 1.

[0060] For example, the digital output matrix is Then D0=[100100] T , D1=[010101] T , D2=[010101] T , D3=[010101] T , D4=[010101] T , [] T Represents a transpose operation.

[0061] S30, calculating a digital summation vector according to the ideal weight vector and the new digital output matrix.

[0062] The embodiment of the present invention is based on the ideal weight vector and the new digital output matrix D new Calculate the digital sum vector, the ideal weight vector is denoted by W ideal =[2 0 …2 n …2 N-1 ] T , where weight 2 n With D n Correspondingly, the digital sum vector is calculated, and the formula is expressed as:

[0063]

[0064] Among them, D sum represents the digital sum vector, D n Represents the n+1th vector in the new digital output matrix, D sum is an M×1 vector. For example, the digital output matrix is Then D is calculated by formula (2): sum =[130031030] T .

[0065] S40, normalizing the frequency of the input signal to obtain normalized frequencies of a plurality of odd harmonics.

[0066] The embodiment of the present invention can normalize the frequency of the input signal to obtain normalized frequencies including the fundamental wave, even harmonics, and odd harmonics. The fundamental wave corresponds to the input signal and needs to be retained. The impact of the even harmonics on the calibration is relatively small and can be ignored. The focus here is on the odd harmonics. Among the odd harmonics, the 3rd harmonic, 5th harmonic, 7th harmonic and 9th harmonic have the greatest impact on the calibration. Therefore, the normalized frequencies of multiple odd harmonics are obtained, more specifically, the normalized frequencies corresponding to the 3rd harmonic, 5th harmonic, 7th harmonic and 9th harmonic are obtained.

[0067] Next, for each odd harmonic, the steps include:

[0068] S50 , performing decimation filtering on the digital sum vector according to the normalized frequency of the current odd harmonic to obtain a harmonic vector corresponding to the current odd harmonic.

[0069] The embodiment of the present invention uses the built-in fir1 function and filter function in MATLAB to calculate the digital summation vector D according to the normalized frequency of the current odd harmonics. sum Perform extraction and filtering, and remove the group delay caused by extraction to obtain the harmonic component corresponding to the current odd harmonic, recorded as Y k , Y k is a vector of size M×1. Digital summation vector D sum It can characterize whether the input signal is an ideal signal. For different odd harmonics, different harmonic components are obtained after extraction and filtering. For how to extract and filter, please refer to the prior art and will not be described here.

[0070] S60. Construct a linear combination expression of a new digital output matrix and a harmonic vector according to the harmonic weight vector to be estimated.

[0071] The embodiment of the present invention constructs a linear combination expression of a new digital output matrix and a harmonic vector according to the harmonic weight vector to be estimated, and the formula is expressed as follows:

[0072] ||D new W k -Y k ||2=||Y′ k -Y k ||2 (3);

[0073] Among them, D new Represents the new digital output matrix, W k Represents the harmonic weight vector to be estimated corresponding to the current odd harmonic, W kis an N×1 vector, k ranges from 1 to K, K represents the odd harmonic number, Y k Indicates the harmonic vector corresponding to the current odd harmonic, Y′ k =D new W k It can be seen that the embodiment of the present invention constructs a method of estimating harmonic components according to harmonic weights to evaluate the impact of different harmonic components on calibration. Here, K=4, W1 corresponds to the weight of the 3rd harmonic, W2 corresponds to the weight of the 5th harmonic, W3 corresponds to the weight of the 7th harmonic, and W4 corresponds to the weight of the 9th harmonic.

[0074] S70, solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic.

[0075] The embodiment of the present invention solves the linear combination expression, including: using the least square method (Recursive Least Square, referred to as RLS), but not limited to the least square method, to solve the linear combination expression shown in formula (3) to obtain the harmonic weight vector W to be estimated corresponding to the current odd harmonic k .

[0076] S80, calculating a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics.

[0077] The embodiment of the present invention can obtain the harmonic weight vectors to be estimated corresponding to all odd harmonics of interest through S50 to S70, such as W1, W2, W3, and W4, and calculate the total harmonic weight vector to be estimated W using the weights of all harmonics that affect the calibration. Total , the formula is:

[0078] W Total =W1+W2+W3+W4 (4).

[0079] 90. Calculate a calibration weight vector according to the ideal weight vector and the total harmonic weight vector to be estimated, and return the calibration weight vector to the SAR ADC.

[0080] The embodiment of the present invention uses an ideal weight vector W ideal Subtract the total harmonic weight vector W to be estimated Total The calibration weight vector for removing the influence of odd harmonics, i.e., the 3rd harmonic, the 5th harmonic, the 7th harmonic, and the 9th harmonic, is obtained, denoted as W cal Finally, the calibration weight vector W cal Sent back to the internal register of the SAR ADC for calibration of subsequent circuits.

[0081] Furthermore, in order to further ensure the calibration weight vector W of the returned SAR ADC calTo improve the reliability of the SAR ADC, the SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction proposed in the embodiment of the present invention further includes:

[0082] According to the new digital output matrix D new and the ideal weight vector W ideal Calculate the level signal before calibration, recorded as V ideal ; According to the new digital output matrix D new and the calibration weight vector W cal Calculate the calibrated level signal, recorded as V cal ; Perform FFT processing on the level signal before calibration and the level signal after calibration respectively, and compare the two FFT processing results, and determine whether to calibrate the weight vector W according to the comparison result cal Return to SAR ADC. How to calculate V ideal 、V cal Refer to the example of S10, and the FFT processing refers to the prior art. Finally, by comparing the results, we can see the improvement of conversion accuracy, spurious-free dynamic range and linearity before and after calibration. The calibration weight vector W with improved indicators cal Returns the internal registers of the SAR ADC, the calibration weight vector W for which the index is not improved cal There is no need to return to the internal register of the SAR ADC yet, and continue the calibration process of S10 to S90 until the comparison results show that the index has improved. The calibration weight vector W is cal Returns the internal register of the SAR ADC. Generally speaking, a reasonable algorithm design can obtain a calibration weight vector W that improves the performance in the first attempt. cal .

[0083] It should be noted here that the embodiments of the present invention are not only applicable to scenarios without redundant bits, but also to scenarios with redundant bits. For the case of redundant bits, the ideal weight vector can be designed according to the actual digital output matrix. The entire algorithm flow is similar to the case of redundant bits, and will not be described in detail here.

[0084] In summary, the SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction proposed in the embodiment of the present invention is a method that can quickly solve the weight required for SAR ADC nonlinear calibration using only one SAR ADC. It takes the ideal binary weight as the real weight, and constantly approaches the real weight value during the calibration process, and constructs a method for estimating the harmonic weight according to the harmonic component. The influence of different harmonic components on the weight is extracted during the calibration, and then the calibration binary weight is calculated according to the estimated harmonic weight. The amount of calculation is small, and the linear combination expression is solved by mathematical algorithms. The speed of solution is fast. In general, the embodiment of the present invention reduces the harmonic distortion and nonlinearity caused by the capacitor mismatch of the SAR ADC, improves the conversion accuracy of the SAR ADC, and accelerates the conversion efficiency of the SAR ADC; the embodiment of the present invention does not need to embed any circuit modules in the SAR ADC, such as digital calibration modules, noise shaping modules, dynamic element matching modules, etc., and is implemented without making any changes to the original circuit. It has high portability, low hardware overhead, short calibration time, high calibration accuracy, and has a wider application prospect.

[0085] Second, see Figure 3 The embodiment of the present invention provides a SARADC nonlinear calibration device based on harmonic distortion component extraction, the calibration device comprising:

[0086] A quantization module, used to quantize the input signal using the SAR ADC to obtain a digital output matrix;

[0087] A construction module, used for separating the digital output matrix by columns and constructing a new digital output matrix;

[0088] A first calculation module, used for calculating a digital sum vector according to an ideal weight vector and a new digital output matrix;

[0089] A normalization module, used for normalizing the frequency of an input signal to obtain normalized frequencies of a plurality of odd harmonics;

[0090] The estimation module is used for each odd harmonic, and the execution steps include: extracting and filtering the digital sum vector according to the normalized frequency of the current odd harmonic to obtain the harmonic vector corresponding to the current odd harmonic; constructing a new digital output matrix and a linear combination expression of the harmonic vector according to the harmonic weight vector to be estimated; solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic;

[0091] A second calculation module is used to calculate a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics;

[0092] The third calculation module is used to calculate the calibration weight vector according to the ideal weight vector and the total harmonic weight vector to be estimated, and return the calibration weight vector to the SAR ADC.

[0093] As for the calibration device embodiment of the second aspect, since it is basically similar to the calibration algorithm embodiment of the first aspect, the description is relatively simple, and the relevant parts may refer to the partial description of the calibration algorithm embodiment of the first aspect.

[0094] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0095] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art may understand and implement other variations of the disclosed embodiments by viewing the specification and its drawings. In the specification, the word "comprising" does not exclude other components or steps, and "a" or "one" does not exclude multiple situations. Certain measures are recorded in different embodiments, but this does not mean that these measures cannot be combined to produce good results.

[0096] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction, It is characterized in that The calibration algorithm includes: Quantize the input signal using SAR ADC to obtain a digital output matrix; Separating the digital output matrix by columns to construct a new digital output matrix; Calculating a digital summation vector based on the ideal weight vector and the new digital output matrix; Normalizing the frequency of the input signal to obtain normalized frequencies of multiple odd harmonics; For each odd harmonic, the execution steps include: extracting and filtering the digital summation vector according to the normalized frequency of the current odd harmonic to obtain the harmonic vector corresponding to the current odd harmonic; constructing a linear combination expression of the new digital output matrix and the harmonic vector according to the harmonic weight vector to be estimated; solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic; Calculating a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics; A calibration weight vector is calculated according to the ideal weight vector and the total harmonic weight vector to be estimated, and the calibration weight vector is returned to the SAR ADC.

2. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The input signal is a standard sine wave signal.

3. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The new digital output matrix constructed is expressed as follows: D new =[D 0 ,…,D n ,…,D N-1 ]; Among them, D new Represents the new digital output matrix, D n It represents the n+1th vector in the new digital output matrix, that is, all elements in the n+1th column of the digital output matrix. The value of n ranges from 0 to N-1, and N represents the number of conversion bits of the SAR ADC.

4. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that Calculate the digital sum vector, the formula is expressed as: Among them, D sum represents the digital sum vector, D n Represents the n+1th vector in the new digital output matrix, consisting of 2 0 , 2 1 , ..., 2 N-1 Construct the ideal weight vector.

5. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The plurality of odd harmonics include a 3rd harmonic, a 5th harmonic, a 7th harmonic, and a 9th harmonic.

6. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The digital summation vector is subjected to decimation filtering according to the normalized frequency of the current odd harmonics, comprising: The digital sum vector is subjected to decimation and filtering processing according to the normalized frequency of the current odd harmonics by using the fir1 function and the filter function, and the group delay caused by the decimation is removed.

7. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The linear combination expression constructed is expressed as follows: ||D new W k -Y k || 2 =||Y′ k -Y k || 2 ; Among them, D new Represents the new digital output matrix, W k Indicates the harmonic weight vector to be estimated corresponding to the current odd harmonic, k ranges from 1 to K, K represents the number of odd harmonics, Y k Indicates the harmonic vector corresponding to the current odd harmonic, Y k '=D new W k .

8. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that Solving the linear combination expression includes: The linear combination expression is solved using the least squares method.

9. The SAR ADC nonlinear calibration algorithm based on harmonic distortion component extraction according to claim 1, It is characterized in that The calibration algorithm also includes: Calculate the level signal before calibration according to the new digital output matrix and the ideal weight vector; Calculate a calibrated level signal according to the new digital output matrix and the calibration weight vector; Perform FFT processing on the level signal before calibration and the level signal after calibration respectively, compare the two FFT processing results, and determine whether to return the calibration weight vector to the SAR ADC according to the comparison result.

10. A SAR ADC nonlinear calibration device based on harmonic distortion component extraction, It is characterized in that The calibration device comprises: A quantization module, used to quantize the input signal using the SAR ADC to obtain a digital output matrix; A construction module, used for separating the digital output matrix by columns to construct a new digital output matrix; A first calculation module, used for calculating a digital sum vector according to an ideal weight vector and the new digital output matrix; A normalization module, used for normalizing the frequency of the input signal to obtain normalized frequencies of multiple odd harmonics; The estimation module is used for each odd harmonic, and the execution steps include: extracting and filtering the digital summation vector according to the normalized frequency of the current odd harmonic to obtain the harmonic vector corresponding to the current odd harmonic; constructing a linear combination expression of the new digital output matrix and the harmonic vector according to the harmonic weight vector to be estimated; solving the linear combination expression to obtain the harmonic weight vector to be estimated corresponding to the current odd harmonic; A second calculation module is used to calculate a total harmonic weight vector to be estimated according to the harmonic weight vectors to be estimated corresponding to all odd harmonics; The third calculation module is used to calculate a calibration weight vector according to the ideal weight vector and the total harmonic weight vector to be estimated, and return the calibration weight vector to the SAR ADC.

Images