Dc anchored nrz amplitude overlap calibration method and system

Abstract

This invention relates to a DC-anchored NRZ amplitude overlap calibration method and system, belonging to the field of high-bandwidth digital communication technology. It addresses the amplitude overlap distortion caused by signal combining in NRZ mode during time-interleaved DACs, as well as the technical problems of existing pre-compensation schemes, such as high-frequency gain amplification leading to sub-DAC encoding overflow, low-frequency gain drift, and the difficulty in simultaneously achieving compensation accuracy and overflow suppression. The method includes: constructing the target frequency response and equivalent amplitude-frequency response of the FIR pre-encoder; determining the compensation conditions of the FIR pre-encoder and converting them into a system of real-valued linear equations; solving the equations and introducing regularization parameters and DC-anchored soft constraints to obtain the initial coefficients of the FIR pre-encoder; applying norm constraints; solving the FIR pre-encoder coefficient optimization problem; and achieving NRZ amplitude overlap calibration. This invention is applicable to applications such as 5G communication, ultra-wideband radar, and high-speed test and measurement instruments.

Description

Technical Field

[0001] This invention belongs to the field of high-bandwidth digital communication technology, specifically relating to a DC-anchored NRZ amplitude overlap calibration technology. Background Technology

[0002] As fields such as 5G communication, ultra-wideband radar, and high-speed test and measurement instruments evolve towards higher bandwidth and higher sampling rates, signal processing systems place stringent demands on the performance of digital-to-analog converters (DACs). The sampling rate of a single-channel DAC is limited by the physical limits of semiconductor technology, making it difficult to meet the application requirements of ultra-high-speed scenarios.

[0003] Time-interleaved DAC (TI-DAC) technology increases the system's equivalent sampling rate to M times that of a single-channel sub-DAC by sampling M low-rate sub-DACs in parallel and interleaved manner according to multi-phase clocks. It has become a core technology solution to break through the single-channel rate bottleneck and achieve ultra-high-speed signal output, and has been widely used in various high-end electronic devices.

[0004] In practical applications of TI-DAC, the non-return-to-zero (NRZ) code mode has become the mainstream signal output mode due to its simple structure and strong timing compatibility. However, in NRZ mode, each sub-DAC maintains the input digital sample for one sampling period. When the output signals of M sub-DACs are combined, the signals of adjacent channels overlap in the time domain, causing amplitude overlap distortion in the combined signal—specifically manifested as abnormal amplitude superposition, blurred edges, and spectral broadening, severely damaging the integrity of the output signal and directly affecting the system's transmission accuracy and dynamic performance. Therefore, digital pre-compensation technology must be used to suppress this amplitude superposition effect to ensure that the TI-DAC's output signal can accurately reproduce the amplitude characteristics of the input digital signal.

[0005] To address the NRZ amplitude overlap problem, various digital pre-compensation schemes have been proposed in this field, the most common being an approximate implementation based on FIR filters. The core idea of ​​this type of scheme is to design the frequency domain response of the pre-encoder to counteract the amplitude overlap equivalent operator effect generated during sub-DAC combining, thereby making the signal amplitude response within the passband tend to be flat. However, existing pre-compensation techniques suffer from intractable technical bottlenecks, specifically manifested in the following three aspects: 1. High-frequency gain amplification triggers sub-DAC encoding overflow. An ideal precoder exhibits a recursive frequency response, with the gain increasing approximately exponentially with frequency. Even with a finite-order FIR approximation, the gain amplification effect remains significant in the high-frequency range. Due to the limited bit width of the sub-DAC, the peak value of the amplified signal easily exceeds the full-amplitude range, leading to code overflow, which in turn causes signal clipping distortion, compressing the system's effective bandwidth and dynamic performance.

[0006] 2. Existing compromise strategies have obvious flaws. To mitigate the overflow problem, existing technologies mostly employ two approaches: one is to limit the effective compensation bandwidth, sacrificing high-frequency compensation effects to suppress gain, which contradicts the core objective of TI-DAC—high speed and high bandwidth; the other is to scale the precoder coefficients as a whole, forcibly reducing the amplitude but simultaneously weakening the accuracy of low-frequency compensation, thus failing to completely eliminate amplitude overlap distortion.

[0007] 3. The coefficient optimization lacks stability and the low-frequency gain is prone to drift. Some schemes use weighted least squares (WLS) plus regularization optimization coefficients, but do not constrain the low-frequency gain reference, which makes the gain in the DC and near-DC frequency bands prone to drift and the output signal has a fixed amplitude offset. At the same time, the lack of "input-independent" amplitude response constraints means that even in the low-frequency band, overflow may be triggered by the instantaneous peak value of the input, resulting in insufficient stability.

[0008] In summary, existing TI-DAC NRZ amplitude overlap compensation technology consistently fails to simultaneously meet the three core requirements of "suppressing amplitude overlap distortion, avoiding code overflow, and maintaining low-frequency gain stability." To address this, a novel precoding calibration method that balances compensation accuracy, overflow suppression, and gain stability is urgently needed. Summary of the Invention

[0009] This invention provides a method based on The norm-constrained DC anchored NRZ amplitude overlap calibration method and system aims to solve the amplitude overlap distortion caused by signal combining in NRZ mode in time-interleaved DACs, as well as the technical problems of existing pre-compensation schemes, such as high-frequency gain amplification causing sub-DAC encoding overflow, easy low-frequency gain drift, and difficulty in balancing compensation accuracy and overflow suppression.

[0010] Firstly, the present invention aims to provide a DC-anchored NRZ amplitude overlap calibration method, comprising the following steps: S1: Construct the frequency response and equivalent amplitude-frequency response of the FIR precoder; S2: Determine the compensation conditions of the FIR precoder based on the equivalent amplitude-frequency response, and convert them into a system of real-valued linear equations; S3: The real-valued linear equation system is solved using the frequency-weighted least squares method, while introducing regularization parameters and DC anchoring soft constraint terms. S4: Solve for the weighted regularized least squares to obtain the initial coefficients of the FIR precoder, and apply... Norm constraints are used to determine the optimization problem of FIR precoder coefficients. The optimal coefficients of the FIR precoder are obtained by solving the problem. The optimal FIR precoder is determined and used to precode the digital signal input to the TI-DAC to achieve NRZ amplitude overlap calibration.

[0011] Furthermore, a preferred embodiment is provided: the frequency response of the FIR precoder is expressed as: , in, For the FIR precoder order, For FIR precoder coefficients, This refers to the digital angular frequency.

[0012] Furthermore, a preferred embodiment is provided: the equivalent amplitude-frequency response is expressed as: , in, This is the equivalent operator for NRZ amplitude overlap.

[0013] Furthermore, a preferred embodiment is provided: the compensation condition of the FIR precoder is expressed as: , in, , indicating the upper limit of the effective passband.

[0014] Furthermore, a preferred embodiment is provided: the transformation steps of the real-valued linear equation system include: For offline selection Each passband frequency sampling point , The product of the FIR precoder frequency response and the NRZ amplitude overlap equivalent operator is expanded into real and imaginary components, and a system of real-valued linear equations is established based on the compensation conditions.

[0015] Furthermore, a preferred embodiment is provided: the DC soft anchoring constraint term is expressed as: , in, This represents the weighting coefficient for DC soft anchoring. It is a column vector consisting entirely of 1s.

[0016] Furthermore, a preferred solution is provided: the FIR precoder coefficient optimization problem is expressed as: , in, This indicates the upper bound of the maximum permissible amplification factor. It is a diagonal weight matrix. For regularization parameters, Represents a matrix of real coefficients. Let be the objective vector of the real-valued linear equation system.

[0017] Secondly, the purpose of this invention is to provide a DC-anchored NRZ amplitude overlap calibration system, said system being implemented based on the DC-anchored NRZ amplitude overlap calibration method as described in any one or more of the above-mentioned schemes, said system comprising: Target frequency response construction module: used to construct the frequency response and equivalent amplitude frequency response of the FIR precoder; Compensation condition determination module: used to determine the compensation conditions of the FIR precoder based on the equivalent amplitude-frequency response, and convert it into a system of real-valued linear equations; Constraint introduction module: used to solve the real-valued linear equation system using the frequency-weighted least squares method, while introducing regularization parameters and DC anchored soft constraint terms; Optimal coefficient solving module: used to solve for the initial coefficients of the FIR precoder using weighted regular least squares, and to apply a weighted regular least squares method to the initial coefficients of the FIR precoder. Norm constraints are used to determine the optimization problem of FIR precoder coefficients. The optimal coefficients of the FIR precoder are obtained by solving the problem. The optimal FIR precoder is determined and used to precode the digital signal input to the TI-DAC to achieve NRZ amplitude overlap calibration.

[0018] Thirdly, the present invention aims to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, and when the processor runs the computer program stored in the memory, the processor executes the DC anchoring NRZ amplitude overlap calibration method according to any one or more of the above-described schemes.

[0019] Fourthly, the present invention aims to provide a computer-readable storage medium for storing a computer program that executes the DC anchoring NRZ amplitude overlap calibration method described in any one or more of the above-described schemes.

[0020] Compared with the prior art, the advantages of the present invention are: 1. This invention is achieved through... Norm constraints limit the maximum amplitude response of the precoder from an input-independent perspective, effectively reducing the risk of sub-DAC encoding overflow, while avoiding the loss of compensation accuracy caused by overall coefficient scaling.

[0021] 2. The method described in this invention introduces a DC soft anchoring constraint to fix the zero-frequency gain reference, which solves the problem of easy low-frequency gain drift in existing solutions, ensures that the output signal has no fixed amplitude offset, and improves the system's working stability.

[0022] 3. This invention uses an FIR pre-encoder + frequency-weighted least squares + regularization design to suppress excessive high-frequency gain amplification without limiting the effective compensation bandwidth. This meets the core application requirements of TI-DAC for high speed and high bandwidth, and avoids the compromise of sacrificing bandwidth for overflow suppression in existing technologies.

[0023] 4. This invention combines the frequency band selectivity error control and regularization mechanism of WLS to effectively suppress the amplification of pre-encoder coefficients, significantly improve the numerical stability of solving linear equations, and is suitable for pre-encoder designs of different orders.

[0024] This invention is applicable to application scenarios such as 5G communication, ultra-wideband radar, and high-speed test and measurement instruments. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0026] Figure 1 This is a flowchart of the DC anchoring NRZ amplitude overlap calibration method according to a specific embodiment of the present invention; Figure 2 This is a graph showing the amplitude-frequency response curve of the pre-encoder according to a specific embodiment of the present invention; Figure 3 This is a fitting diagram of the 60th-order pre-encoder as described in a specific embodiment of the present invention; Figure 4 This is a fitting diagram of the 150th-order pre-encoder as described in a specific embodiment of the present invention; Figure 5 This is a time-domain curve of the coefficients of the 60th-order pre-encoder as described in a specific embodiment of the present invention; Figure 6 This is a time-domain curve of the coefficients of the 150th-order pre-encoder as described in a specific embodiment of the present invention; Figure 7 This is a schematic diagram of the TI-DAC pre-encoder structure according to a specific embodiment of the present invention. Detailed Implementation

[0027] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0028] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0030] Implementation Method 1 Reference Figure 2 Description of this implementation method This implementation proposes a method based on A norm-constrained DC-anchored NRZ amplitude overlap calibration method is proposed. This method designs a pre-encoder for amplitude overlap compensation by constraining the high-frequency gain to avoid dynamic range limitations. First, an NRZ model is established for the TI-DAC passband frequency sampling points, and a linear representation of the FIR frequency response is constructed using cosine basis expansion. The compensation conditions are decomposed into real and imaginary parts to form a system of real-valued linear equations. Then, the frequency-weighted least squares (WLS) method is introduced to solve the linear equations, and regularization and DC soft anchoring constraints are added to enhance the stability of the solution. Finally, the initial coefficients are obtained by solving the weighted regularized least squares method, and then... Norm scaling limits the magnitude of the coefficients to prevent overflow and obtains the optimal coefficients for the pre-encoder.

[0031] This implementation mainly optimizes the amplitude overlap problem caused by the NRZ-DAC in the TI-DAC during the combining process, while reducing the risk of DAC gain overflow that the pre-encoder may cause in the high-frequency band.

[0032] In TI-DAC, Each sub-DAC is driven by a multi-phase clock during interleaving, sampling sequentially with adjacent channels having a phase difference of [missing information]. Taking a two-channel system as an example, The sampling period of the TI-DAC is The frequency domain transfer function of the precoded code is expressed as: , (1) The discrete-time frequency response of the precoder ,in, , The frequency response of the pre-encoder is the continuous-time angular frequency. Figure 2 As shown.

[0033] As can be seen from formula (1), the pre-encoder adopts a recursive structure at the digital angular frequency. Its gain tends towards infinity. Since the gain of the precoder exhibits an approximately exponential increase with frequency, its amplitude modulation degree varies significantly across different frequency bands. The gain in the high-frequency region is significantly increased, causing the peak value of the codeword input to the sub-DAC to be amplified, easily exceeding the full-scale range of the DAC. Therefore, in practical applications, it is usually necessary to suppress the high-frequency gain of the precoder by limiting the effective compensation bandwidth to avoid dynamic range limitations. This implementation uses a finite-order FIR approximation of the precoder's frequency response and introduces an amplitude constraint mechanism to effectively reduce the risk of DAC code overflow while ensuring compensation accuracy.

[0034] The method described in this embodiment specifically includes the following steps: S1: Construct the frequency response and equivalent amplitude-frequency response of the FIR precoder.

[0035] S2: Determine the compensation conditions of the FIR precoder based on the equivalent amplitude-frequency response, and convert them into a system of real-valued linear equations; S3: The real-valued linear equation system is solved using the frequency-weighted least squares method, while introducing regularization parameters and DC anchoring soft constraint terms. S4: Solve for the weighted regularized least squares to obtain the initial coefficients of the FIR precoder, and apply... Norm constraints are used to determine the optimization problem of FIR precoder coefficients. The optimal coefficients of the FIR precoder are obtained by solving the problem. The optimal FIR precoder is determined and used to precode the digital signal input to the TI-DAC to achieve NRZ amplitude overlap calibration.

[0036] Furthermore, the frequency response of the FIR precoder is: , (2) in, For the FIR precoder order, For FIR precoder coefficients, This refers to the digital angular frequency.

[0037] The equivalent amplitude-frequency response is: .

[0038] To avoid target frequency response Numerical amplification at high frequencies leads to a deterioration of the TI-DAC condition number. This implementation introduces an NRZ amplitude overlap equivalent operator. The ideal precoder satisfies the compensation condition within the effective passband as follows: , (3) in, , indicating the upper limit of the effective passband.

[0039] Let the pre-encoder coefficients be Offline frequency sampling points are , Then (2) can be expressed as: , (4) Therefore, the precoding frequency response can be expressed as: , (5) in, Representation matrix with vector The product of the first product Each component. Representation matrix with vector The product of the first product Each component. The NRZ amplitude overlap equivalent operator can be expressed as follows for each frequency point: (6) From (4) and (6), p. The equivalent amplitude-frequency response of each component can be expanded as follows: (7) Therefore, the real part of the compensation condition and the virtual part It can be represented as: (8) Substituting (8) into (3), the compensation form is transformed into the form of a real linear system: (9) Therefore, the precoder coefficient optimization problem is transformed into solving a system of linear equations, which can be expressed as: (10) Since the pre-encoder's impact on the TI-DAC differs significantly between the passband and transition band, this embodiment employs the WLS method for pre-encoder coefficient optimization. This method enables band-selective error control and, by introducing regularization, effectively suppresses pre-encoder coefficient amplification, thereby significantly improving numerical stability. However, WLS cannot directly constrain the pre-encoder's gain reference at low frequencies and may lead to uncontrolled time-domain output peaks. Therefore, this embodiment further introduces a DC anchoring constraint to fix the zero-frequency gain reference, while simultaneously applying a ℓ1 norm constraint to the pre-encoder coefficients. This limits the pre-encoder's maximum amplitude response from an input-independent perspective, effectively reducing the risk of DAC code overflow while ensuring passband compensation accuracy.

[0040] Furthermore, The frequency response of the pre-encoder at DC is expressed as: , (11) The NRZ magnitude overlap equivalent operator at DC is According to (3), it can be calculated Therefore, the DC soft anchor term is described as follows: (12) Where μ is the weighting coefficient of DC soft anchoring, It is a column vector of all 1s. Then (10) can be further expressed as: (13) in, It is a diagonal weight matrix. This is the regularization parameter. Adding DC soft anchoring can lock the low-frequency gain reference, preventing overall scaling drift. Let the input signal of the pre-encoder be... The output sequence after encoder is Then the input-output relationship can be expressed as: , (14) Taking the absolute values ​​of both sides of the equation and applying the triangle inequality, equation (14) can be transformed into: (15) For any sequence, it exists. ,in express Norm, which is the maximum absolute value of the magnitude of a sequence. According to (15), for any variable... The upper bound of the output sequence is represented as: (16) For both sides of the equation, the variables Taking the maximum value simultaneously, equation (16) can be expressed as: (17) This inequality indicates that the output peak of the pre-encoder is at most amplified. times, therefore constrained It can provide an input-independent upper bound for the output peak, reducing the risk of sub-DAC encoding overflow.

[0041] Therefore, the precoder coefficient optimization problem can be expressed as: (18) in, This indicates the upper bound of the maximum permissible amplitude amplification factor. Unless otherwise stated, in this embodiment... The value is 1.5.

[0042] Implementation Method 2 Reference Figure 3 , Figure 4 This implementation method is described below.

[0043] This implementation method is based on the method described in Implementation Method 1. Further examples illustrate the norm-constrained DC-anchored NRZ amplitude overlap calibration method.

[0044] In this embodiment, the TI-DAC sampling rate is 25 GSa / s, and the passband bandwidth is 6 GHz. The frequency response diagrams of precoders with different orders are shown below. Figure 3 , Figure 4 As shown in the figure, the left vertical axis represents the pre-encoder frequency response, and the right vertical axis represents the equivalent amplitude-frequency response AH. The equivalent amplitude-frequency response of the WLS design is... The frequency fluctuation is relatively large, approximately 0.2 dB. Scaling 0.5 and the method proposed in this invention can approximate the ideal frequency response very well. As the order increases, the overall fitting effect of each method improves; however, Scaling 0.5 introduces approximately 6 dB of amplitude attenuation at low frequencies, thus reducing the compensation accuracy. In contrast, the method proposed in this invention has the best overall fitting effect and the smallest amplitude deviation in the Nyquist region.

[0045] The time-domain comparison of precoder coefficients of different orders is shown in the figure below. Figure 4 , Figure 5 As shown in Table 1, the specific indicators are as follows.

[0046] Table 1 Performance metrics of precoders with different compensation strategies

[0047] As shown in Table 1, the Scaling 0.5 method forcibly reduces the coefficient amplitude through overall scaling, resulting in the lowest overflow risk, but it also weakens the compensation capability. The WLS method has the largest values ​​for ℓ1 and ℓ2, indicating a larger coefficient size, the highest peak amplification effect, and a greater likelihood of DAC encoding overflow. In contrast, the method proposed in this invention falls between the other two methods in terms of various indicators, effectively suppressing the coefficient size while ensuring compensation effectiveness. Moreover, as the order increases, the coefficient growth is more controlled, thus further reducing codeword amplification and overflow risk.

[0048] Implementation Method 3 Reference Figure 1 , Figure 5 This implementation method is described below.

[0049] This implementation method is based on the method described in Implementation Method 1. Further examples illustrate the norm-constrained DC-anchored NRZ amplitude overlap calibration method.

[0050] The structural diagram of the TI-DAC precoder optimization scheme proposed in this embodiment is shown below. Figure 5 As shown, the input signal is calibrated by a pre-encoder to compensate for the amplitude overlap effect caused by the NRZ-DAC during interleaving. By introducing ℓ1 norm constraints and DC anchoring, the pre-encoder coefficients are constrained, thereby suppressing codeword peak amplification and reducing the risk of DAC encoding overflow. After precoding and TI-DAC output, the goal is to make the output signal amplitude response as close as possible to the input signal within the passband of interest, achieving amplitude consistency.

[0051] The implementation method described herein is based on The flowchart of the norm-constrained DC anchored NRZ amplitude overlap calibration method is as follows: Figure 1 As shown, it includes the following steps: (1) Pre-encoder modeling. First, the equivalent operator is introduced. Target frequency response of pre-encoder Construct an equivalent amplitude-frequency response.

[0052] (2) Solve the imaginary and real parts of the complex equivalent amplitude-frequency response. Based on the equivalent amplitude-frequency response conditions, the precoder design problem is transformed into a system of matrix linear equations.

[0053] (3) By introducing ℓ1 norm constraints and DC anchoring soft constraints, the DC gain is fixed, solution drift is avoided, and the upper bound of the peak is determined. The optimal coefficients of the pre-encoder are obtained by combining weights and regularization parameters. .

[0054] A schematic diagram of the TI-DAC precoder optimization scheme is shown below. Figure 5As shown, the input signal is calibrated by a pre-encoder to compensate for the amplitude overlap effect caused by the NRZ-DAC during interleaving. By introducing ℓ1 norm constraints and DC anchoring, the pre-encoder coefficients are constrained, thereby suppressing codeword peak amplification and reducing the risk of DAC encoding overflow. After precoding and TI-DAC output, the goal is to make the output signal amplitude response as close as possible to the input signal within the passband of interest, achieving amplitude consistency.

[0055] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A DC anchor NRZ amplitude overlap calibration method, characterized in that, Includes the following steps: S1: Construct the frequency response and equivalent amplitude-frequency response of the FIR precoder; S2: Determine the compensation conditions of the FIR precoder based on the equivalent amplitude-frequency response, and convert them into a system of real-valued linear equations; S3: The real-valued linear equation system is solved using the frequency-weighted least squares method, while introducing regularization parameters and DC anchoring soft constraint terms. S4: Solve for the weighted regularized least squares to obtain the initial coefficients of the FIR precoder, and apply... Norm constraints are used to determine the optimization problem of FIR precoder coefficients. The optimal coefficients of the FIR precoder are obtained by solving the problem. The optimal FIR precoder is determined and used to precode the digital signal input to the TI-DAC to achieve NRZ amplitude overlap calibration.

2. The DC anchored NRZ amplitude swing overlap calibration method of claim 1, wherein, The frequency response of the FIR precoder is expressed as: ， wherein, is the FIR precoder order, is the FIR precoder coefficient, is the digital angular frequency.

3. The DC anchored NRZ amplitude swing overlap calibration method of claim 2, wherein, The equivalent amplitude-frequency response is expressed as: ， in, This is the equivalent operator for NRZ amplitude overlap.

4. The DC anchored NRZ amplitude overlap calibration method according to claim 3, characterized in that, The compensation condition of the FIR precoder is expressed as follows: ， in, , indicating the upper limit of the effective passband.

5. The DC anchored NRZ amplitude overlap calibration method according to claim 3, characterized in that, The transformation steps of the real-valued linear equation system include: For offline selection Each passband frequency sampling point , The product of the FIR precoder frequency response and the NRZ amplitude overlap equivalent operator is expanded into real and imaginary components, and a system of real-valued linear equations is established based on the compensation conditions.

6. The DC anchored NRZ amplitude overlap calibration method according to claim 1, characterized in that, The DC soft anchoring constraint term is expressed as follows: ， in, This represents the weighting coefficient for DC soft anchoring. It is a column vector consisting entirely of 1s.

7. The DC anchored NRZ amplitude overlap calibration method according to claim 1, characterized in that, The FIR precoder coefficient optimization problem is expressed as: ， in, This indicates the upper bound of the maximum permissible amplification factor. It is a diagonal weight matrix. For regularization parameters, Represents a matrix of real coefficients. Let be the objective vector of the real-valued linear equation system.

8. A DC-anchored NRZ amplitude overlap calibration system, characterized in that, The system is implemented based on the DC anchored NRZ amplitude overlap calibration method as described in any one of claims 1-7, and the system includes: Target frequency response construction module: used to construct the frequency response and equivalent amplitude frequency response of the FIR precoder; Compensation condition determination module: used to determine the compensation conditions of the FIR precoder based on the equivalent amplitude-frequency response, and convert it into a system of real-valued linear equations; Constraint introduction module: used to solve the real-valued linear equation system using the frequency-weighted least squares method, while introducing regularization parameters and DC anchored soft constraint terms; Optimal coefficient solving module: used to solve for the initial coefficients of the FIR precoder using weighted regular least squares, and to apply a weighted regular least squares method to the initial coefficients of the FIR precoder. Norm constraints are used to determine the optimization problem of FIR precoder coefficients. The optimal coefficients of the FIR precoder are obtained by solving the problem. The optimal FIR precoder is determined and used to precode the digital signal input to the TI-DAC to achieve NRZ amplitude overlap calibration.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the DC anchoring NRZ amplitude overlap calibration method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that performs the DC anchored NRZ amplitude overlap calibration method according to any one of claims 1-7.

Images