DTC nonlinear calibration circuit based on adaptive normalization

By using an adaptively normalized DTC nonlinear calibration circuit, the adaptation problem of the third-order DTC nonlinear calibration circuit in different Σ-Δ modulator operating modes is solved, the calibration coefficient jitter is reduced, the calibration effect is improved, and the compensation capability of sub-quantization units is achieved.

CN123113276APending Publication Date: 2026-11-17BEIHANG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611618228.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the existing technology, the third-order nonlinear calibration circuit of DTC cannot be adapted to the working mode of different Σ-Δ modulators, resulting in mutual interference between gain calibration and third-order nonlinear calibration, large fluctuations in calibration coefficients, and insufficient subquantization unit compensation capability.

Method used

A DTC nonlinear calibration circuit based on adaptive normalization is adopted. Through a dynamic normalization circuit, a distributed sensing third-order basis function generation circuit, and an adaptive update circuit, the normalization denominator and orthogonalization coefficients are dynamically adjusted according to the working mode and output code distribution of the Σ-Δ modulator to generate a third-order calibration basis function. The third-order nonlinear compensation circuit is then used for compensation to achieve cumulative compensation of subquantization units.

Benefits of technology

It reduces the statistical coupling between third-order calibration and gain calibration, maintains the stability of calibration coefficients, improves the effect of third-order nonlinear calibration, is applicable to different modulator operating modes, and achieves sub-quantization unit compensation capability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113276A_ABST
    Figure CN123113276A_ABST
Patent Text Reader

Abstract

This invention discloses a DTC nonlinear calibration circuit based on adaptive normalization, belonging to the field of analog mixed-signal integrated circuits. The circuit includes a dynamic normalization circuit, a distributed sensing third-order basis function generation circuit, a third-order nonlinear compensation circuit, and an adaptive update circuit. The dynamic normalization circuit selects the normalization denominator and normalizes the control code according to the Σ-Δ modulator operating mode; the distributed sensing third-order basis function generation circuit determines the orthogonality coefficients based on the current modulator output code pattern distribution, generating a third-order calibration basis function that is statistically orthogonal to the first-order gain basis function; the third-order nonlinear compensation circuit generates a compensation amount to correct the phase error; the adaptive update circuit updates the third-order nonlinear calibration coefficients based on the sign of the compensated phase error signal and the amplitude of the third-order basis function. This invention can adapt to different modulator operating modes, reduce the mutual coupling between gain calibration and third-order calibration, reduce coefficient jitter, and improve calibration convergence speed and steady-state accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of analog and mixed-signal integrated circuit technology, and in particular to a DTC nonlinear calibration circuit based on adaptive normalization. Background Technology

[0002] The Digital-to-Time Converter (DTC) is a key module in a fractional-frequency division phase-locked loop (PLL). It converts digital control codes into corresponding time delays to compensate for fractional phase errors during the fractional-frequency division process. A PLL typically includes a Sigma-Delta Modulator (Σ-Δ modulator), whose output is a time-varying digital control sequence that drives the DTC to generate the corresponding time delay.

[0003] Ideally, the time delay generated by DTC should maintain a linear relationship with the input digital control code. However, due to factors such as circuit structure, component mismatch, and inconsistency of delay units, actual DTC exhibits integral nonlinearity (INL), causing the actual time delay to deviate from the ideal relationship, generating residual phase errors, and potentially creating fractional spurious noise and increasing in-band phase noise.

[0004] The nonlinearity error of DTC can be divided into even-function error components and odd-function error components. For difference DTC, even-function errors can be offset to some extent through the difference structure, but odd-function errors, which are mainly composed of cubic terms, are usually difficult to eliminate effectively, becoming the main limiting factor for further improving the linearity of DTC and reducing fractional spuriousness.

[0005] Existing techniques typically employ first-order basis functions to calibrate DTC gain error and fixed standard third-order Legendre polynomials to calibrate third-order nonlinear error. However, different Σ-Δ modulators vary in order, structure, and operating mode, resulting in differences in the effective range and probability distribution of their output codes. Therefore, fixed normalization parameters and fixed third-order calibration basis functions cannot guarantee good applicability and orthogonality across different modulator operating modes. This can easily lead to mutual interference between gain calibration and third-order nonlinear calibration, resulting in calibration coefficient oscillations, reduced convergence speed, or increased steady-state error.

[0006] Furthermore, the traditional symbol-based least mean square (LMS) algorithm typically symbolizes both the error signal and the calibration basis function simultaneously, losing the amplitude information of the calibration basis function. When the amplitude of the calibration basis function is small, the update amount may still remain large, easily causing the calibration coefficients to fluctuate near the steady state. In the digital fixed-point compensation process, the fractional part of the product may also be directly truncated, making it impossible to retain compensation amounts smaller than one phase error quantization unit, resulting in insufficient sub-quantization unit compensation capability.

[0007] Therefore, it is necessary to provide a DTC third-order nonlinear adaptive calibration circuit that can adjust the third-order calibration basis function according to the actual output code distribution of the Σ-Δ modulator, adapt to different modulator operating modes, reduce the mutual influence between gain calibration and third-order nonlinear calibration, and retain the subquantization unit compensation amount. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a DTC nonlinear calibration circuit based on adaptive normalization, which solves the following technical problems in the prior art: (1) The fixed normalization strategy cannot adapt to the change in the effective range of DTC control codes under different modulator working modes; (2) The standard third-order Legendre polynomial cannot be adapted to the output distribution of the non-uniform modulator, which leads to the problem of mutual coupling between the third-order calibration and the gain calibration. (3) The traditional symbol-to-symbol update method loses the amplitude information of the third-order basis function, resulting in large fluctuations in the calibration coefficient.

[0009] This invention discloses a DTC nonlinear calibration circuit based on adaptive normalization, including a dynamic normalization circuit, a distributed sensing third-order basis function generation circuit, a third-order nonlinear compensation circuit, and an adaptive update circuit. The dynamic normalization circuit selects the corresponding normalization denominator according to the current Σ-Δ modulator operating mode, and performs normalization processing on the control code of the digital time converter to generate a normalized control quantity. The distributed sensing third-order basis function generation circuit determines the orthogonalization coefficients based on the normalized control quantity and the output code distribution corresponding to the current Σ-Δ modulator operating mode, and generates a third-order calibration basis function based on the orthogonalization coefficients, so that the statistical correlation between the third-order calibration basis function and the first-order gain basis function is lower than a preset threshold. The third-order nonlinear compensation circuit generates a third-order nonlinear compensation quantity based on the third-order nonlinear calibration coefficient and the third-order calibration basis function, and uses the third-order nonlinear compensation quantity to compensate the original phase error signal, and outputs the compensated phase error signal. The adaptive update circuit updates the third-order nonlinear calibration coefficients based on the sign of the compensated phase error signal and the amplitude value of the third-order calibration basis function.

[0010] Optionally, the digital time converter is a differential digital time converter, the dynamic normalization circuit obtains the positive branch control code and the negative branch control code, takes the difference between the positive branch control code and the negative branch control code as the differential control code, and divides the differential control code by the normalization denominator to obtain the normalized control quantity.

[0011] Optionally, the dynamic normalization circuit selects one of multiple preset normalization denominator values ​​as the normalization denominator for the current sampling period according to the Σ-Δ modulator operating mode; wherein the normalization denominator value corresponding to the second modulator operating mode is greater than the normalization denominator value corresponding to the first modulator operating mode, so that the normalization control quantity under different modulator operating modes is kept within a preset effective range.

[0012] Optionally, the digital time converter is a single-ended digital time converter, and the dynamic normalization circuit obtains the single-ended control code and the control code range center point. The single-ended control code is subtracted from the control code range center point and then divided by the normalization denominator to obtain the normalized control quantity.

[0013] Optionally, the distributed sensing third-order basis function generation circuit calculates the orthogonalization coefficient based on the second-order and fourth-order statistical moments of the normalized control quantity corresponding to the current Σ-Δ modulator operating mode, and then subtracts the product of the orthogonalization coefficient and the normalized control quantity from the cube of the normalized control quantity, and multiplies it by the amplitude scaling factor to obtain the third-order calibration basis function.

[0014] Optionally, the third-order nonlinear compensation circuit includes a fixed-point multiplier and a subtractor; the fixed-point multiplier multiplies the third-order nonlinear calibration coefficient with the third-order calibration basis function to generate the third-order nonlinear compensation amount; the subtractor subtracts the third-order nonlinear compensation amount from the original phase error signal to output the compensated phase error signal.

[0015] Optionally, the third-order nonlinear compensation circuit further includes a fractional residual feedback circuit, which includes a residual adder, an integer extraction circuit, a fractional residual update circuit, and a fractional residual feedback register. The residual adder adds the fixed-point product of the fixed-point multiplier to the fractional residual stored in the fractional residual feedback register to obtain a product with residuals. The integer extraction circuit extracts the integer part from the product with residuals as the third-order nonlinear compensation amount. The fractional residual update circuit calculates the updated fractional residual based on the product with residuals and the third-order nonlinear compensation amount, and stores it in the fractional residual feedback register.

[0016] Optionally, the adaptive update circuit includes a sign function module, a multiplier, and a coefficient accumulator; the sign function module signs the compensated phase error signal; the multiplier multiplies the adaptive step size, the sign of the compensated phase error signal, and the third-order calibration basis function of the current sampling period to generate an update amount; the coefficient accumulator adds the current third-order nonlinear calibration coefficients to the update amount to obtain the updated third-order nonlinear calibration coefficients.

[0017] Optionally, when the normalized control quantity corresponding to the current Σ-Δ modulator operating mode follows a uniform distribution, the distribution-sensing third-order basis function generation circuit uses a preset standard third-order Legendre polynomial as the third-order calibration basis function.

[0018] Optionally, the circuit has a modular calibration structure that can configure the control code input method, normalized denominator value, orthogonalization coefficient and adaptive update step size according to the type of digital time converter and the Σ-Δ modulator operating mode, so as to adapt to differential digital time converters or single-ended digital time converters, and to adapt to multiple modulator operating modes.

[0019] Compared with the prior art, the present invention has at least the following beneficial effects: (1) The present invention determines the third-order orthogonalization coefficients based on the current modulator output code pattern distribution, instead of using the standard third-order Legendre polynomial in all modulator modes. Therefore, it can reduce the statistical coupling between third-order calibration and gain calibration.

[0020] (2) The present invention dynamically determines the normalized denominator based on the effective DTC gain control word and the modulator working mode, so that the normalized control quantity under different modulator modes is kept within a similar effective range.

[0021] (3) In the adaptive update process, the present invention retains the amplitude information of the third-order calibration basis function. Compared with the update method where both the error signal and the regression quantity are signed, it can reduce the excessive update in the small regression quantity region and the jitter of the calibration coefficient.

[0022] (4) The present invention sets a fractional residual feedback structure in the third-order nonlinear compensation circuit so that the compensation amount less than one phase error quantization unit can be accumulated in multiple sampling periods, thereby realizing the average compensation of sub-quantization unit.

[0023] (5) The present invention reduces the component related to the third-order nonlinear error in the phase error signal and improves the third-order nonlinear calibration effect by means of closed-loop feedback between the original phase error signal, the third-order nonlinear compensation quantity and the compensated phase error signal.

[0024] (6) The present invention adopts a modular calibration structure, which can configure the control code input method, normalization parameters, orthogonalization coefficients and adaptive update step size according to the specific structure of the digital time converter, the modulator working mode and the modulator output code pattern distribution. It can be applied to differential digital time converters and single-ended digital time converters. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the embodiments will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the accompanying drawings. The accompanying drawings are schematic and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 The signal flow graph for calibrating the DTC nonlinear calibration circuit based on adaptive normalization according to the present invention is shown.

[0027] Figure 2a The original phase error signal, the third-order nonlinear compensation amount, and the compensated phase error signal are obtained from the adaptive normalization-based DTC nonlinear calibration circuit of the present invention in the first modulator operating mode.

[0028] Figure 2b The original phase error signal, the third-order nonlinear compensation amount, and the compensated phase error signal are obtained from the adaptive normalization-based DTC nonlinear calibration circuit of the present invention in the second modulator operating mode.

[0029] Figure 3 This is the convergence result of the third-order nonlinear calibration coefficients of the DTC nonlinear calibration circuit based on adaptive normalization in different modulator operating modes. Detailed Implementation

[0030] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other.

[0031] like Figure 1 As shown, this invention provides a DTC nonlinear calibration circuit based on adaptive normalization, which is a third-order nonlinear adaptive calibration circuit for a digital-to-time converter (DTC) based on modulator distributed sensing orthogonal normalization. The DTC nonlinear calibration circuit includes a dynamic normalization circuit, a distributed sensing third-order basis function generation circuit, a third-order nonlinear compensation circuit, and an adaptive update circuit.

[0032] The DTC nonlinear calibration circuit of this invention receives the effective gain control word of the digital time converter and the Σ-Δ modulator operating mode signal, and generates a normalized control quantity and a third-order calibration basis function according to the modulator operating mode. After third-order nonlinear compensation, the digital time converter control code is compared with the original phase error signal output by the phase error detector. Correction is performed to obtain the compensated phase error. Based on the adaptive update circuit, the third-order nonlinear calibration coefficients for the k-th sampling period are... The update is performed, and the third-order nonlinear calibration coefficients for the k-th sampling period are updated. Feedback is sent to a third-order nonlinear compensation circuit to form a closed-loop calibration process.

[0033] The dynamic normalization circuit is located at the input of the third-order nonlinear calibration circuit. It is used to receive input signals such as the positive branch control code, negative branch control code, effective gain control word, and Σ-Δ modulator operating mode signal applied to the digital time converter, and generate a normalized control quantity based on the above input signals.

[0034] In this invention, the digital time converter is disposed in the fractional phase compensation path of the fractional frequency divider phase-locked loop. The positive and negative branch control codes of the digital time converter are generated by the phase control circuit in the fractional frequency divider phase-locked loop and applied to the positive and negative branches of the differential digital time converter, respectively. The dynamic normalization circuit obtains the positive and negative branch control codes from the phase control circuit and performs differential and normalization processing on the control codes.

[0035] Furthermore, the digital time converter is a differential digital time converter. For a differential digital time converter, differential operations are performed to obtain the differential control code for the k-th sampling period. The expression is:

[0036] in, This represents the positive branch control code of the k-th sampling period. This represents the negative branch control code for the kth sampling period.

[0037] Differential control code based on the k-th sampling period The ratio of the differential control code to the normalized denominator in the k-th sampling period is obtained, i.e., the normalized control quantity, expressed as:

[0038] in, This represents the normalized control quantity in the k-th sampling period. This represents the normalized denominator for the k-th sampling period.

[0039] Furthermore, according to the Σ-Δ modulator operating mode, the expression for the normalized denominator of the k-th sampling period is:

[0040] Where m=1 represents the first modulator operating mode and m=2 represents the second modulator operating mode; This indicates the effective gain control word of the currently active differential-to-digital-time converter. In this embodiment, the first modulator operating mode corresponds to the first-order digital Σ-Δ modulator operating mode, and the second modulator operating mode corresponds to the first-order multi-stage noise-shaping (MASH) 1-1 operating mode. These two operating modes are only used to indicate different output code patterns of the Σ-Δ modulator; the Σ-Δ modulator operating mode is selected or configured by the mode configuration circuit in the fractional-order frequency-locked loop.

[0041] The differential control code effective range of the differential digital time converter in the second modulator operating mode is larger than that in the first modulator operating mode. Therefore, a larger normalized denominator is used to keep the normalized control quantity in the two Σ-Δ modulator operating modes within a similar effective range.

[0042] The normalized control quantity is subjected to amplitude limiting to satisfy: .

[0043] Furthermore, for single-ended digital-to-time converters, the single-ended control code adopts... This indicates that the center point of the control code is adopted. The expression for the normalized control quantity in the k-th sampling period is:

[0044] in, This indicates the center point of the control code range for a single-ended digital time converter.

[0045] It should be noted that, Indicates the first The normalized control quantity for each sampling period is the discrete sampled value of the normalized control quantity within a specific sampling period. This represents the random or statistical variable corresponding to the normalized control quantity, used to describe the overall probability distribution of the Σ-Δ modulator output code after normalization. Therefore, it is used when calculating the basis function for a single sampling period. When performing statistical averaging, statistical moment calculation, and correlation analysis, the following methods are used: .

[0046] The distributed sensing third-order basis function generation circuit generates control quantities based on normalized control quantities. Based on the current Σ-Δ modulator output code pattern distribution, a third-order calibration basis function is generated, expressed as:

[0047] in, Indicates the modulator operating mode The corresponding third-order calibration basis function; Indicates the amplitude scaling factor; Indicates the modulator operating mode The corresponding orthogonalization coefficients; This represents the random or statistical variable corresponding to the normalized control quantity, used to describe the overall probability distribution of the Σ-Δ modulator output code after normalization.

[0048] Furthermore, the expression for the orthogonalization coefficients is:

[0049] in, Represents the second-order statistical moments; This represents the fourth-order statistical moment.

[0050] Furthermore, the expression for the first-order gain basis function is:

[0051] Furthermore, by setting the modulator operating mode corresponding orthogonalization coefficients This makes the statistical correlation between the third-order calibration basis function and the first-order gain basis function close to zero: .

[0052] Furthermore, when inputting the first Normalized control quantity per sampling period It follows a uniform distribution and the normalized interval is At that time, there were: ,when At that time, the third-order calibration basis functions degenerate into standard third-order Legendre polynomials: .

[0053] In the generation of third-order basis functions, Q8 quantization represents the Q8 fixed-point format. The Q8 fixed-point format indicates that the fractional part uses 8-bit quantization, and the integer code corresponding to the value 1 is... .

[0054] The third-order calibration basis function, after Q8 quantization, is expressed as:

[0055] When the modulator output code pattern distribution is not uniform, the code pattern determined based on the current modulator output distribution is used. This reduces the statistical correlation between the third-order calibration basis function and the first-order gain basis function.

[0056] Furthermore, the third-order nonlinear compensation circuit includes a fixed-point multiplier, a subtractor, and a fractional residual feedback circuit.

[0057] Specifically, the fixed-point multiplier receives the third-order nonlinear calibration coefficients of the k-th sampling period. and third-order calibration basis functions The third-order nonlinear compensation quantity is generated, and its expression is:

[0058] in, Indicates the first The third-order nonlinear compensation quantity for each sampling period.

[0059] Specifically, the subtractor receives the original phase error signal from the kth sampling period. and third-order nonlinear compensation quantity The output compensated phase error signal is expressed as follows:

[0060] in, Indicates the first The compensated phase error signal for each sampling period.

[0061] The fractional residual feedback circuit includes a Q16 fixed-point multiplier, a residual adder, an integer extraction circuit, a fractional residual update circuit, and a fractional residual feedback register.

[0062] Specifically, the Q16 fixed-point multiplier outputs a fixed-point product with a 16-bit fractional part. The expression is:

[0063] in, Represents the third-order nonlinear calibration coefficients in Q8 format for the k-th sampling period; Let represent the third-order calibration basis function in Q8 format for the k-th sampling period.

[0064] Specifically, the residual adder adds the fixed-point product of the k-th sampling period to the fractional residual stored in the previous sampling period, resulting in:

[0065] in, This indicates the first time after adding residual feedback. Fixed-point product of each sampling period; This represents the fractional residual stored in the fractional residual feedback register at the beginning of the k-th sampling period. It is calculated by the fractional residual update circuit of the previous sampling period, that is, the update obtained in the previous sampling period. Used as the residual feedback quantity for the current sampling period.

[0066] Specifically, the integer extraction circuit uses a floor function, which is represented by a sign-down function. Indicates the integer third-order nonlinear compensation quantity. for:

[0067] Specifically, the fractional residual update circuit calculates the fractional residual for the (k+1)th sampling period based on the integer compensation amount. The expression is:

[0068] Specifically, the decimal residual feedback register uses code variables. express, Used for saving Therefore, the first Calculated from each sampling period It is written to the fractional residual feedback register, and in the first... Each sampling period is used as the residual feedback quantity.

[0069] When the absolute value of an unquantized third-order nonlinear compensation quantity is less than one phase error quantization unit, the compensation quantity is not directly discarded, but is stored in the code variable as a fractional residual. The error is stored in the corresponding fractional residual feedback register and continues to accumulate in subsequent sampling periods until an output integer compensation amount is formed.

[0070] The adaptive update circuit includes a sign function, an adaptive step size input, a multiplication and update quantity generation circuit, a coefficient accumulator, and a third-order basis function amplitude input.

[0071] Specifically, the update amount in the kth sampling period The expression is:

[0072] in, Indicates adaptive step size, Represents a symbolic function.

[0073] Furthermore, the sign function The expression is:

[0074] in, The signal to be judged represents the input sign function. In this invention, Indicates the first Compensated phase error signal for each sampling period .

[0075] Specifically, the coefficient accumulator generates the third-order nonlinear calibration coefficients to be used in the (k+1)th sampling period based on the current third-order nonlinear calibration coefficients and the update amount. The expression is:

[0076] in, Indicates the first The third-order nonlinear calibration coefficients currently stored for each sampling period, This represents the third-order nonlinear calibration coefficients obtained after the coefficient accumulator is updated, which are used in subsequent sampling periods. Updated third-order nonlinear calibration coefficients. Feedback to Figure 1 The third-order nonlinear compensation circuit shown is used to generate a third-order nonlinear compensation value in subsequent sampling periods.

[0077] Unlike the sign-sign least mean square update method where both the error signal and the regression quantity are signed, this invention only signs the compensated phase error signal and retains the amplitude information of the third-order calibration basis function.

[0078] To illustrate the effectiveness of the method proposed in this invention, the following detailed description of the above technical solution is provided through several specific embodiments. The specific implementation steps are as follows: Example 1: Differential Digital Time Converter First Modulator Mode Calibration; This embodiment employs a differential digital time converter structure, including positive branch control codes. and negative branch control code .

[0079] In the first modulator operating mode, the normalized denominator is: ; The normalized control quantity is: ; The orthogonalization coefficients corresponding to the first modulator operating mode are: ; The third-order calibration basis function is: ; The third-order calibration basis function is quantized by Q8 and then input into the third-order nonlinear compensation circuit.

[0080] Example 2: Differential Digital Time Converter Second Modulator Mode Calibration; The difference between this embodiment and Embodiment 1 is that the modulator is switched to the second modulator operating mode.

[0081] In the second modulator operating mode, the normalized denominator is: ; The orthogonalization coefficients corresponding to the second modulator operating mode are: ; The third-order calibration basis function is: ; With the same effective digital-to-digital converter gain control word, the second modulator operating mode uses a larger normalized denominator to accommodate its larger effective range of differential control codes.

[0082] Example 3: Adaptive update of third-order nonlinear calibration coefficients; The compensation circuit generates a third-order nonlinear compensation quantity based on the third-order nonlinear calibration coefficients and the third-order calibration basis function: ; The compensated phase error signal is: ; The adaptive update circuit determines the sign of the signal based on the compensated phase error signal and retains the amplitude information of the third-order calibration basis function. ; When there is a systematic error in the compensated phase error signal that is related to the third-order calibration basis function, the third-order nonlinear calibration coefficients are updated in the direction of reducing the related error.

[0083] Example 4: Joint operation of third-order calibration and gain calibration; The gain calibration loop uses first-order basis functions: ; The third-order nonlinear calibration loop uses: ; The two calibration loops share the compensated phase error signal: ; The gain calibration loop updates the gain calibration coefficients based on the correlation between the compensated phase error signal and the first-order basis function, while the third-order nonlinear calibration loop updates the third-order nonlinear calibration coefficients based on the correlation between the compensated phase error signal and the third-order calibration basis function.

[0084] When the following conditions are met: At this time, the mutual influence between the two calibration loops is reduced.

[0085] Example 5: Joint operation of third-order calibration and gain calibration; This embodiment employs a single-ended digital time converter structure. The single-ended digital time converter includes a single-ended control code. .

[0086] The normalized control value for a single-ended digital time converter is: ; Single-ended digital time converters require explicit control code center point subtraction. Differential digital time converters, on the other hand, reduce the impact of common-mode center point offset on the normalized control quantity by subtracting the positive and negative branch control codes.

[0087] A single-ended digital time converter can employ the same third-order basis function generation circuit, third-order nonlinear compensation circuit, fractional residual feedback circuit, and adaptive update circuit as a differential digital time converter.

[0088] Example 6: Joint operation of third-order calibration and gain calibration; The third-order nonlinear calibration coefficients and third-order calibration basis functions are represented in Q8 format.

[0089] The multiplier output is: ; The product of the decimal residuals is: ; Integer compensation amount is: ; The residual is updated as follows: ; When the absolute value of the compensation is less than one phase error quantization unit, the compensation is retained. Instead of being directly truncated, it is in the middle.

[0090] Example 7: Calibration control status; The calibration circuit can be configured with the following control functions: 1. Automatic third-order nonlinear calibration enabled; 2. Manual input of third-order nonlinear calibration coefficients; 3. Automatic calibration coefficient freezing; 4. Third-order nonlinear compensation bypass; 5. Clear the third-order nonlinear calibration accumulator; 6. Readback of third-order nonlinear calibration coefficients; 7. Effective digital-to-time converter gain control word readback; 8. Modulator operating mode configuration; 9. Adaptive update step size configuration.

[0091] Manual third-order nonlinear calibration coefficients can be used for loading offline measurement results, functional verification, and calibration coefficient comparison. Automatic third-order nonlinear calibration coefficients are used for normal background calibration.

[0092] Example 8: Calibration Verification; like Figure 2a As shown, in the first modulator operating mode, the original phase error signals are displayed respectively. Third-order nonlinear compensation quantity and the compensated phase error signal .

[0093] The original phase error signal contains significant third-order nonlinear error variations. After third-order nonlinear compensation, the amplitude of the compensated phase error signal is significantly reduced, mainly retaining quantization noise and other unmodeled error components.

[0094] like Figure 2b As shown, in the second modulator operating mode, the original phase error signal is also compensated for using the corresponding normalized denominator and orthogonalization coefficients in a third-order nonlinear manner. The compensated phase error signal is significantly reduced compared to the original phase error signal, indicating that the calibration circuit can adapt to the output code pattern distribution under different modulator operating modes.

[0095] Example 9: Convergence Verification of Third-Order Nonlinear Calibration Coefficients like Figure 3 As shown, the third-order nonlinear calibration coefficients are recorded in the first modulator operating mode and the second modulator operating mode, respectively. The process of change over simulation time.

[0096] In the first modulator operating mode, the third-order nonlinear calibration coefficients gradually enter the stable region after an initial rapid adjustment, with a convergence time of approximately 0.578 milliseconds in 90 percent of cases.

[0097] In the second modulator operating mode, the third-order nonlinear calibration coefficients also enter the stable region after initial adjustment, with a convergence time of approximately 0.630 milliseconds in 90 percent of cases.

[0098] The above results show that the present invention can select the corresponding normalization parameters and orthogonalization coefficients according to different modulator operating modes, and complete the adaptive update of the third-order nonlinear calibration coefficients.

[0099] It should be noted that the specific convergence time is related to the modulator output code pattern distribution, the nonlinearity error amplitude of the digital-to-time converter, the quantization bit width, and the adaptive update step size. Without altering the basic technical concept of this invention, the adaptive update step size and fixed-point data bit width can be adjusted according to specific application requirements.

[0100] The above verification mainly targets the third-order phase-closed-loop error in the differential digital-time converter, the first modulator operating mode, and the second modulator operating mode. For different digital-time converter structures, different modulator output codes, and other error sources, the control code mapping relationship, normalization parameters, timing relationships, and orthogonality coefficients should be re-determined according to the specific circuit structure.

[0101] The calibration circuit of this invention has a modular structure, allowing configuration of the control code input method, normalized denominator, orthogonalization coefficients of the third-order calibration basis function, fixed-point bit width, update step size, and register interface according to the specific structure of the digital time converter and the modulator operating mode. Therefore, this invention can be applied to both differential and single-ended digital time converters, and is adaptable to third-order nonlinear calibration under various modulator operating modes.

[0102] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. An adaptive normalization based DTC nonlinear calibration circuit, characterized by, This includes a dynamic normalization circuit, a distributed sensing third-order basis function generation circuit, a third-order nonlinear compensation circuit, and an adaptive update circuit; The dynamic normalization circuit selects the corresponding normalization denominator according to the current Σ-Δ modulator operating mode, and performs normalization processing on the control code of the digital time converter to generate a normalized control quantity. The distributed sensing third-order basis function generation circuit determines the orthogonalization coefficients based on the normalized control quantity and the output code distribution corresponding to the current Σ-Δ modulator operating mode, and generates a third-order calibration basis function based on the orthogonalization coefficients, so that the statistical correlation between the third-order calibration basis function and the first-order gain basis function is lower than a preset threshold. The third-order nonlinear compensation circuit generates a third-order nonlinear compensation quantity based on the third-order nonlinear calibration coefficient and the third-order calibration basis function, and uses the third-order nonlinear compensation quantity to compensate the original phase error signal, and outputs the compensated phase error signal. The adaptive update circuit updates the third-order nonlinear calibration coefficients based on the sign of the compensated phase error signal and the amplitude value of the third-order calibration basis function.

2. The DTC non-linear calibration circuit of claim 1, wherein, The digital time converter is a differential digital time converter. The dynamic normalization circuit obtains the positive branch control code and the negative branch control code, takes the difference between the positive branch control code and the negative branch control code as the differential control code, and divides the differential control code by the normalization denominator to obtain the normalized control quantity.

3. The DTC non-linear calibration circuit of claim 2, wherein, The dynamic normalization circuit selects one of multiple preset normalization denominator values ​​as the normalization denominator for the current sampling period according to the Σ-Δ modulator operating mode; wherein, the normalization denominator value corresponding to the second modulator operating mode is greater than the normalization denominator value corresponding to the first modulator operating mode, so that the normalization control quantity under different modulator operating modes is kept within a preset effective range.

4. The DTC non-linear calibration circuit of claim 1, wherein, The digital time converter is a single-ended digital time converter. The dynamic normalization circuit obtains the single-ended control code and the control code range center point. The single-ended control code is subtracted from the control code range center point and then divided by the normalization denominator to obtain the normalized control quantity.

5. The DTC nonlinear calibration circuit of claim 1, wherein, The distributed sensing third-order basis function generation circuit calculates the orthogonalization coefficients based on the second and fourth statistical moments of the normalized control quantity corresponding to the current Σ-Δ modulator operating mode. Then, it subtracts the product of the orthogonalization coefficients and the normalized control quantity from the cube of the normalized control quantity and multiplies it by the amplitude scaling factor to obtain the third-order calibration basis function.

6. The DTC non-linear calibration circuit of claim 1, wherein, The third-order nonlinear compensation circuit includes a fixed-point multiplier and a subtractor; the fixed-point multiplier multiplies the third-order nonlinear calibration coefficient with the third-order calibration basis function to generate the third-order nonlinear compensation amount; the subtractor subtracts the third-order nonlinear compensation amount from the original phase error signal to output the compensated phase error signal.

7. The DTC non-linear calibration circuit of claim 6, wherein, The third-order nonlinear compensation circuit further includes a fractional residual feedback circuit, which comprises a residual adder, an integer extraction circuit, a fractional residual update circuit, and a fractional residual feedback register. The residual adder adds the fixed-point product of the fixed-point multiplier to the fractional residual stored in the fractional residual feedback register to obtain a product with residuals. The integer extraction circuit extracts the integer part from the product with residuals as the third-order nonlinear compensation amount. The fractional residual update circuit calculates the updated fractional residual based on the product with residuals and the third-order nonlinear compensation amount, and stores it in the fractional residual feedback register.

8. The DTC nonlinear calibration circuit of claim 1, wherein, The adaptive update circuit includes a sign function module, a multiplier, and a coefficient accumulator; the sign function module signs the compensated phase error signal; the multiplier multiplies the adaptive step size, the sign of the compensated phase error signal, and the third-order calibration basis function of the current sampling period to generate the update amount; The coefficient accumulator adds the current third-order nonlinear calibration coefficient to the update amount to obtain the updated third-order nonlinear calibration coefficient.

9. The DTC nonlinear calibration circuit of claim 5, wherein, When the normalized control quantity corresponding to the current Σ-Δ modulator operating mode follows a uniform distribution, the distribution-sensing third-order basis function generation circuit uses a preset standard third-order Legendre polynomial as the third-order calibration basis function.

10. The DTC nonlinear calibration circuit of claim 1, wherein, The circuit has a modular calibration structure, which can configure the control code input method, normalized denominator value, orthogonalization coefficient and adaptive update step size according to the type of digital time converter and the working mode of Σ-Δ modulator, so as to adapt to differential digital time converters or single-ended digital time converters, and to adapt to various modulator working modes.