Arbitrary wave generator predistortion method
By acquiring the time-domain pulse step response signal of an arbitrary wave generator, establishing a time-domain distortion model, and using a first-order IIR or FIR filter for pre-distortion processing, the problem of long-term overshoot distortion caused by the non-ideal characteristics of the hardware link is solved, and simple and efficient signal correction is achieved.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2026-03-05
Smart Images

Figure CN2024117465_05032026_PF_FP_ABST
Abstract
Description
A predistortion method for arbitrary wave generators
[0001] Cross-reference to related applications
[0002] This application claims priority to Chinese Patent Application No. 202411212798.6, filed on August 30, 2024, entitled "A Predistortion Method for an Arbitrary Wave Generator", the entire contents of which are incorporated herein by reference. Technical Field
[0003] This disclosure relates to the field of precision test and measurement technology, and in particular to a predistortion method for an arbitrary wave generator. Background Technology
[0004] Arbitrary wave generators, as general-purpose electronic instruments, can generate various realistic waveforms and excitation signals, and are widely used in modern electronic test and measurement, electronic simulation, and other fields. However, due to many non-ideal characteristics in the entire hardware link from the analog circuit of the arbitrary wave generator to the load, such as amplifier ringing, parasitic capacitance and inductance in the circuit, and impedance mismatch, the output signal of the arbitrary wave generator exhibits varying degrees of distortion when it reaches the load. This distortion includes ringing, overshoot, and reflected standing waves. These distortions are often unacceptable in some precision electronic test and measurement applications.
[0005] Currently, waveform predistortion is commonly used to reduce waveform distortion caused by non-ideal characteristics of hardware links. The basic principle of this method is to first measure the frequency domain response of the link, and then take the reciprocal of that response to obtain the predistorted response. The original waveform of the arbitrary wave generator undergoes this predistortion processing before output, thus canceling out these non-ideal characteristics when passing through the actual link, thereby reducing waveform distortion at the load end. However, this method has drawbacks such as large measurement and computational requirements, and poor performance in correcting overshoot distortion over long periods.
[0006] Summary of the Invention
[0007] This disclosure provides a predistortion method for an arbitrary wave generator to correct distortion signals of the long-term overshoot type. The method is simple to measure, requires little computation, and has a good correction effect.
[0008] According to one aspect of this disclosure, a predistortion method for an arbitrary wave generator is provided, the predistortion method for an arbitrary wave generator comprising:
[0009] Obtain the time-domain pulse step response signal output by the arbitrary wave generator;
[0010] A time-domain distortion model is established based on the time-domain pulse step response signal.
[0011] The first pre-distortion signal is obtained by correcting the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model.
[0012] Optionally, correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model includes:
[0013] A first-order IIR filter is obtained based on the time-domain distortion model.
[0014] The distortion model parameters of the first-order IIR filter are obtained based on the time-domain pulse step response signal and the time-domain distortion model.
[0015] The distortion model parameters of the first-order IIR filter are used to correct the long-term overshoot type distortion signal in the time-domain pulse step response signal.
[0016] Optionally, obtaining the first-order IIR filter based on the time-domain distortion model includes:
[0017] The time-domain distortion model is sequentially subjected to Laplace transform, predistortion transform, and bilinear transform to obtain a first-order IIR filter.
[0018] Optionally, obtaining the distortion model parameters of the first-order IIR filter based on the time-domain pulse step response signal and the time-domain distortion model includes:
[0019] The distortion model parameters of the first-order IIR filter are obtained by fitting the time-domain pulse step response signal and the time-domain distortion model.
[0020] Optionally, the distortion model parameters of the first-order IIR filter include the amplitude and time constant of the distorted signal.
[0021] Optionally, the step of correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal based on the distortion model parameters of the first-order IIR filter and the first-order IIR filter includes:
[0022] The distortion model parameters of the first-order IIR filter are input into the first-order IIR filter to compensate for the long-term overshoot type distortion signal in the time-domain pulse step response signal.
[0023] Optionally, the time-domain distortion model is:
[0024] y(t)=(1+Ae -t / τ )·u(t)
[0025] Where y(t) is the voltage across the load, u(t) is the step response, A is the amplitude of the distorted signal, and τ is the time constant of the distorted signal.
[0026] Optionally, after correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain the first pre-distortion signal, the method further includes:
[0027] The impulse response is obtained by performing a differential operation on the first predistorted signal.
[0028] The predistorted impulse response is obtained based on the impulse response and the unit impulse function;
[0029] The first predistorted signal is predistorted based on the predistorted impulse response to obtain the second predistorted signal.
[0030] Optionally, before performing differential operations on the first predistorted signal, the method further includes:
[0031] The first predistortion signal is normalized.
[0032] Optionally, obtaining the predistorted impulse response based on the impulse response and the unit impulse function includes: obtaining the predistorted impulse response using the L2 regularization method based on the impulse response and the unit impulse function;
[0033] The step of predistorting the first predistorted signal according to the predistorted impulse response to obtain the second predistorted signal includes: using the predistorted impulse response as the coefficient of an FIR filter, and predistorting the first predistorted signal according to the FIR filter to obtain the second predistorted signal.
[0034] The technical solution of this disclosure provides a predistortion method for an arbitrary wave generator. This method includes: acquiring the time-domain pulse step response signal output by the arbitrary wave generator; establishing a time-domain distortion model based on the time-domain pulse step response signal; and correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain a first predistortion signal. Therefore, this method can achieve the following: obtaining the predistortion response by measuring the time-domain pulse step response signal of the arbitrary wave generator is simpler than the traditional method of directly measuring the system's frequency response. Furthermore, by establishing a time-domain distortion model, the long-time overshoot type distortion signal can be corrected, and compared with existing methods, the correction effect is better and the computational load is smaller.
[0035] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 is a flowchart of an arbitrary wave generator predistortion method provided in an embodiment of this disclosure;
[0038] Figure 2 is a schematic diagram of the distorted waveform acquired by the oscilloscope provided in the embodiment of this disclosure;
[0039] Figure 3 is a partial enlarged view of Figure 2 provided in the embodiment of this disclosure;
[0040] Figure 4 is a schematic diagram of the circuit structure formed by the arbitrary wave generator and the load provided in the embodiment of this disclosure;
[0041] Figure 5 is a flowchart of another arbitrary wave generator predistortion method provided in the embodiments of this disclosure;
[0042] Figure 6 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure;
[0043] Figure 7 is a schematic diagram of fitting the time-domain pulse step response signal and the time-domain distortion model provided in the embodiments of this disclosure;
[0044] Figure 8 is a schematic diagram of the signal waveform after pre-distortion by a first-order IIR filter provided in an embodiment of this disclosure;
[0045] Figure 9 is a partial enlarged view of Figure 8 provided in an embodiment of this disclosure;
[0046] Figure 10 is a schematic diagram of the waveform after pre-distortion by multiple IIR filters provided in an embodiment of this disclosure;
[0047] Figure 11 is a partial enlarged view of Figure 10 provided in an embodiment of this disclosure;
[0048] Figure 12 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure;
[0049] Figure 13 is a schematic diagram of a distorted signal with a short residual time after pre-distortion by a first-order IIR filter, provided in an embodiment of this disclosure.
[0050] Figure 14 is a partial enlarged view of Figure 13 provided in an embodiment of this disclosure;
[0051] Figure 15 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure;
[0052] Figure 16 is a diagram showing the effect of pre-distortion processing of the FIR filter provided in the embodiments of this disclosure. Detailed Implementation
[0053] To enable those skilled in the art to better understand the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present disclosure.
[0054] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0055] The inventors discovered that the arbitrary wave generator, as a general-purpose electronic instrument, can generate various realistic waveforms and excitation signals, and has wide applications in modern electronic test and measurement, electronic simulation, and other fields. However, due to many non-ideal characteristics in the entire hardware link from the analog circuit of the arbitrary wave generator to the load, such as amplifier ringing, parasitic capacitance and inductance in the circuit, and impedance mismatch, the output signal of the arbitrary wave generator experiences varying degrees of distortion when it reaches the load. This distortion includes ringing, overshoot, and reflected standing waves. These distortions are often unacceptable in some precision electronic test and measurement applications.
[0056] Currently, waveform predistortion is commonly used to reduce waveform distortion caused by non-ideal characteristics of hardware links. The basic principle of this method is to first measure the frequency domain response of the link, and then take the reciprocal of that response to obtain the predistorted response. The original waveform of the arbitrary wave generator undergoes this predistortion processing before output, thus canceling out these non-ideal characteristics when passing through the actual link, thereby reducing waveform distortion at the load end.
[0057] In practical implementation, there are generally two methods to obtain the frequency response of an arbitrary wave generator hardware link. One method is direct measurement: the arbitrary wave generator outputs single-frequency sine wave signals of different frequencies within the frequency band in sequence at certain frequency steps. The amplitude and phase of each frequency signal are then acquired using an oscilloscope, thus obtaining the frequency response of the hardware link across the entire frequency band. The drawback of this method is that it requires extensive testing to obtain accurate results. The other method utilizes the principle of transforming the system's impulse response to the frequency domain to obtain the system's frequency response. An ideal step signal is output by the arbitrary wave generator, and the step signal is acquired using an oscilloscope to obtain the system's step response. A difference operation is performed on the step response to obtain the system's impulse response, and then a Fourier transform is performed on the impulse response to obtain the system's frequency response. While this method does not require extensive testing, obtaining accurate frequency response results requires performing FFT operations on a large number of waveform points, resulting in a significant computational burden.
[0058] After obtaining the system's frequency response, there are generally two methods to predistort the original waveform data. One method is to use a higher-order digital filter to approximate the system's inverse frequency response, and then use that filter to predistort the data. The other method is to perform an FFT on the original waveform, multiply it by the reciprocal of the FFT result of the system's impulse response, and finally perform an inverse FFT to obtain the predistorted waveform.
[0059] While the aforementioned waveform predistortion method can theoretically handle distortion caused by any non-ideal characteristics, it has drawbacks in practical applications, such as large computational load, difficulty in real-time processing, insufficient frequency response resolution leading to poor overshoot correction effect over long periods, and sensitivity to noise.
[0060] Therefore, this disclosure provides an arbitrary wave generator predistortion method that can correct distortion signals of the long-term overshoot type, and the measurement method is simple, the amount of calculation is small, and the correction effect is good.
[0061] Figure 1 is a flowchart of a pre-distortion method for an arbitrary wave generator provided in an embodiment of this disclosure. This embodiment can be applied to correct the distortion signal output by an arbitrary wave generator. As shown in Figure 1, the method includes:
[0062] S110. Obtain the time-domain pulse step response signal output by the arbitrary wave generator.
[0063] Among them, the time-domain pulse step response signal of any wave generator can be acquired by an oscilloscope, and the measurement method is simpler compared with the traditional method of directly measuring the frequency response of the system.
[0064] Among them, the time-domain pulse step response signal contains distortion signals, such as long-term overshoot type distortion signals.
[0065] Figure 2 is a schematic diagram of the distorted waveform acquired by the oscilloscope provided in this embodiment of the present disclosure, and Figure 3 is a partial enlarged view of Figure 2 provided in this embodiment of the present disclosure. Referring to Figures 2 and 3, the pulse step response waveform acquired by the oscilloscope has a larger amplitude and a longer duration of distortion compared to the ideal pulse step response.
[0066] S120. Establish a time-domain distortion model based on the time-domain pulse step response signal.
[0067] Waveform distortion is mainly caused by amplifier ringing, parasitic capacitance and inductance in the circuit, impedance mismatch, and uneven system frequency response. Among these, parasitic capacitance and inductance in the circuit generally lead to overshoot distortion that lasts for a longer period of time.
[0068] Figure 4 is a schematic diagram of the circuit structure formed by the arbitrary wave generator and the load provided in the embodiment of this disclosure. Referring to Figure 4, U s (t) represents the output voltage of the arbitrary wave generator, R s Let y(t) be the output impedance of the arbitrary wave generator, L be the parasitic inductance of the line, C be the parasitic capacitance of the line, R be the load resistance, and y(t) be the voltage across the load.
[0069] WhenU s When u(t) = u(t) (u(t) is the unit step function), the arbitrary wave generator outputs a step voltage. The voltage at the load terminal can be derived from the response of the RLC circuit as follows:
[0070] Wherein, are c2, c3, s1, s2 and R s The constants related to R, L, and C. From the above equation, it can be seen that the parasitic capacitance and inductance in the circuit will superimpose the distortion of the multi-exponential response on the time-domain step waveform.
[0071] For example, taking first-order exponential response distortion as an example, a time-domain distortion model is established based on the time-domain pulse step response signal. Assume the step response to be played by the arbitrary wave generator is:
[0072] x(t)=u(t)
[0073] The response at the load end after exponential distortion, i.e., the time-domain distortion model, is:
[0074] y(t)=(1+Ae -t / τ )·u(t)
[0075] Where y(t) is the voltage across the load, u(t) is the step response, A is the amplitude of the distorted signal, and τ is the time constant of the distorted signal.
[0076] S130. Correct the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain the first pre-distortion signal.
[0077] Specifically, by establishing a time-domain distortion model, long-term overshoot distortion signals in the time-domain pulse step response signal can be calibrated, thereby compensating for these distortion signals and achieving pre-distortion processing of arbitrary wave generators. Furthermore, compared to existing methods that directly compensate for the system's frequency response, the calibration effect is better.
[0078] In this embodiment, the working principle of the arbitrary wave generator predistortion method is as follows: Referring to Figure 1, firstly, the time-domain pulse step response signal output by the arbitrary wave generator is acquired. Then, a corresponding time-domain distortion model is established based on the time-domain pulse step response signal. Finally, the long-time overshoot type distortion signal in the time-domain pulse step response signal is corrected according to the time-domain distortion model, thereby compensating for the long-time overshoot type distortion signal and ensuring that all long-time overshoot distortion is compensated. Furthermore, compared with the traditional method of directly compensating the system frequency response, establishing a time-domain distortion model corresponding to the time-domain pulse step response signal results in a better correction effect for long-time overshoot distortion.
[0079] The technical solution of this embodiment provides a predistortion method for an arbitrary wave generator. This method includes: acquiring the time-domain pulse step response signal output by the arbitrary wave generator; establishing a time-domain distortion model based on the time-domain pulse step response signal; and correcting the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain a first predistortion signal. Therefore, this method can achieve the following: obtaining the predistortion response by measuring the time-domain pulse step response signal of the arbitrary wave generator is simpler than the traditional method of directly measuring the system's frequency response. Furthermore, by establishing a time-domain distortion model, the long-term overshoot type distortion signal can be corrected, and compared with existing methods, the correction effect is better and the computational load is smaller.
[0080] Figure 5 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure. Optionally, based on the above embodiments and referring to Figure 5, the method includes the following steps:
[0081] S210. Obtain the time-domain pulse step response signal output by the arbitrary wave generator.
[0082] S220. Establish a time-domain distortion model based on the time-domain pulse step response signal.
[0083] S230. Obtain a first-order IIR filter based on the time-domain distortion model.
[0084] In this process, after establishing the corresponding time-domain distortion model based on the time-domain pulse step response signal, a first-order IIR filter can be obtained from the time-domain distortion model, so that the distortion signal of long-term overshoot type can be corrected by the first-order IIR filter.
[0085] S240. Obtain the distortion model parameters of the first-order IIR filter based on the time-domain pulse step response signal and the time-domain distortion model.
[0086] Optionally, the distortion model parameters of the first-order IIR filter include the amplitude and time constant of the distorted signal.
[0087] S250. Based on the distortion model parameters of the first-order IIR filter and the first-order IIR filter, correct the long-term overshoot type distortion signal in the time-domain pulse step response signal.
[0088] Specifically, a first-order IIR filter is first obtained through a time-domain distortion model. Then, the distortion model parameters of the first-order IIR filter are obtained based on the time-domain pulse step response signal and the time-domain distortion model. Finally, the distortion signal of long-term overshoot type in the time-domain pulse step response signal is corrected based on the distortion model parameters of the first-order IIR filter and the first-order IIR filter. Thus, long-term overshoot distortion can be corrected with a small number of IIR filters (such as first-order). Compared with existing technologies, the correction effect is better, the computation is smaller, and it is easier to implement.
[0089] In this embodiment, the working principle of the arbitrary wave generator predistortion method is as follows: Referring to Figure 5, firstly, the time-domain pulse step response signal output by the arbitrary wave generator is acquired. Then, a time-domain distortion model is established based on the time-domain pulse step response signal. Secondly, a first-order IIR filter is obtained based on the time-domain distortion model. The distortion model parameters of the first-order IIR filter are obtained based on the time-domain pulse step response signal and the time-domain distortion model. Finally, the long-term overshoot distortion signal in the time-domain pulse step response signal is corrected based on the distortion model parameters of the first-order IIR filter and the first-order IIR filter. Therefore, it can be seen that long-term overshoot distortion can be corrected using a small number of IIR filters (such as first-order). Compared with existing technologies, the correction effect is better, the computational load is smaller, and it is easier to implement. Furthermore, obtaining the predistortion response by measuring the time-domain pulse step response signal of the arbitrary wave generator is a simpler measurement method compared to the traditional direct measurement of the system's frequency response.
[0090] Figure 6 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure. Optionally, based on the above embodiments and referring to Figure 6, the method includes the following steps:
[0091] S310: Obtain the time-domain pulse step response signal output by the arbitrary wave generator.
[0092] S320. Establish a time-domain distortion model based on the time-domain pulse step response signal.
[0093] S330. The time-domain distortion model is sequentially subjected to Laplace transform, predistortion transform and bilinear transform to obtain a first-order IIR filter.
[0094] Among them, after establishing a time-domain distortion model based on the time-domain impulse step response signal, performing a Laplace transform on the time-domain distortion model yields the complex frequency domain response that causes the distortion as follows:
[0095] Then, by performing a predistortion transform on the complex frequency domain response, the predistortion frequency response that can cancel out the line frequency response causing the distortion is:
[0096] Finally, by performing a bilinear transform on the predistortion frequency response, we can obtain the z-transform of the IIR digital filter that approximately realizes the predistortion frequency response as follows:
[0097] in,
[0098] in,
[0099] in,
[0100] in,
[0101] λ=2τ(A+1)+T s
[0102] Among them, T s Let A be the sampling time, A be the amplitude of the distorted signal, and τ be the time constant of the distorted signal.
[0103] Therefore, it can be seen that the first-order exponential response distortion can be corrected by using a first-order IIR filter. Using a small number of filters can reduce the amount of computation and improve the correction effect. Moreover, it is easy to implement through the time-domain distortion model and the first-order IIR filter.
[0104] S340. Fit the time-domain pulse step response signal and the time-domain distortion model to obtain the distortion model parameters of the first-order IIR filter.
[0105] One way to fit the time-domain pulse step response signal and the time-domain distortion model is to use a fitting function, such as the curve_fit function.
[0106] Fitting the time-domain pulse step response signal to the time-domain distortion model can reduce the impact of measurement noise.
[0107] Figure 7 is a schematic diagram of fitting the time-domain pulse step response signal and the time-domain distortion model provided in the embodiments of this disclosure. Referring to Figure 7, curve S1 is the time-domain pulse step response signal, curve S2 is the time-domain distortion model, and the fitting of the time-domain pulse step response signal and the time-domain distortion model is shown in Figure 7.
[0108] S350. Input the distortion model parameters of the first-order IIR filter into the first-order IIR filter to compensate for the long-term overshoot type distortion signal in the time-domain pulse step response signal.
[0109] The coefficients of a first-order IIR filter are obtained by inputting the distortion model parameters of the first-order IIR filter into it. This first-order IIR filter is then used to pre-distort the time-domain pulse step response signal, correcting long-term overshoot distortion and compensating for the distortion response of the actual link. Figure 8 is a schematic diagram of the signal waveform after pre-distortion by a first-order IIR filter according to an embodiment of this disclosure, and Figure 9 is a partial enlarged view of Figure 8 according to an embodiment of this disclosure. As can be seen from Figures 8 and 9, the fitted distortion response is compensated for after compensation by the first-order IIR filter.
[0110] It should be noted that if multiple exponential response distortions are superimposed in the actual circuit, multiple first-order IIR filters can be used for multiple corrections. Figure 10 is a schematic diagram of the waveform after pre-distortion by multiple IIR filters provided in this embodiment of the present disclosure, and Figure 11 is a partial enlarged view of Figure 10 provided in this embodiment of the present disclosure. When the distorted waveform has multiple long-term overshoot distortions superimposed, the above process can be repeated multiple times to obtain multiple IIR filters, and the overshoot distortion can be compensated by multiple IIR filters. The waveform after pre-distortion processing by multiple IIR filters is shown in Figures 10 and 11.
[0111] In this embodiment, the working principle of the arbitrary wave generator predistortion method is as follows: Referring to Figure 6, firstly, the time-domain pulse step response signal output by the arbitrary wave generator is acquired. Then, a time-domain distortion model is established based on the time-domain pulse step response signal. Secondly, the time-domain distortion model is sequentially subjected to Laplace transform, predistortion transform, and bilinear transform to obtain a first-order IIR filter. The time-domain pulse step response signal and the time-domain distortion model are fitted to obtain the distortion model parameters of the first-order IIR filter. Finally, the distortion model parameters of the first-order IIR filter are input into the first-order IIR filter to compensate for the long-term overshoot distortion signal in the time-domain pulse step response signal. Therefore, it can be seen that long-term overshoot distortion can be corrected using a small number of IIR filters (such as first-order). Compared with existing technologies, the correction effect is better, the computational load is smaller, and it is easier to implement. Furthermore, obtaining the predistortion response by measuring the time-domain pulse step response signal of the arbitrary wave generator is a simpler measurement method compared to the traditional direct measurement of the system's frequency response.
[0112] Figure 12 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure. Optionally, based on the above embodiments, the method includes the following steps:
[0113] S410: Obtain the time-domain pulse step response signal output by the arbitrary wave generator.
[0114] S420. Establish a time-domain distortion model based on the time-domain pulse step response signal.
[0115] S430. Correct the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain the first pre-distortion signal.
[0116] S440. Perform differential operation on the first predistorted signal to obtain the impulse response.
[0117] The step waveform after processing the long-term overshoot distortion signal through the above steps, i.e., the first pre-distortion signal, will still retain some short-duration distortion. Figure 13 is a schematic diagram of the short-duration residual distortion signal after pre-distortion by a first-order IIR filter according to an embodiment of this disclosure, and Figure 14 is a partial enlarged view of Figure 13 according to an embodiment of this disclosure. As shown in Figures 13 and 14, the short-duration residual distortion signal after pre-distortion by a first-order IIR filter is generally caused by amplifier ringing, impedance mismatch, etc.
[0118] Specifically, the impulse response h(n) can be obtained by performing differential operations on the first predistorted signal after predistortion by a first-order IIR filter.
[0119] S450. Obtain the predistorted impulse response based on the impulse response and the unit impulse function.
[0120] For example, suppose the predistorted impulse response is h inv (n), then we have
[0121] h(n)*h inv (n)=δ(n)
[0122] Where δ(n) is the unit impulse function. Therefore, it is only necessary to obtain the predistorted impulse response h. inv (n) The predistortion response can be obtained by satisfying the above formula, that is, the short-term distortion signal in the first predistortion signal is corrected.
[0123] S460. The first predistorted signal is predistorted according to the predistorted impulse response to obtain the second predistorted signal.
[0124] Optionally, before performing differential operations on the first predistorted signal, the method further includes: normalizing the first predistorted signal.
[0125] In this embodiment, the working principle of the arbitrary wave generator predistortion method is as follows: Referring to Figure 12, firstly, the time-domain pulse step response signal output by the arbitrary wave generator is acquired. Then, a time-domain distortion model is established based on the time-domain pulse step response signal. Secondly, the long-term overshoot distortion signal in the time-domain pulse step response signal is corrected according to the time-domain distortion model to obtain a first predistortion signal. Differential operation is performed on the first predistortion signal to obtain the impulse response. The predistorted impulse response is obtained based on the impulse response and the unit impulse function. Finally, the first predistortion signal is predistorted according to the predistorted impulse response to obtain a second predistortion signal. Therefore, it can be seen that the correction of long-term overshoot distortion can be achieved using a small number of IIR filters (such as first-order filters). Compared with existing technologies, the correction effect is better, the computational load is smaller, and it is easier to implement. Furthermore, after predistorting the time-domain pulse step response signal using a first-order IIR filter, the predistorted impulse response is obtained by differential operation on the predistorted signal (i.e., the first distorted signal). Using this predistorted impulse response to further predistort the signal after the first-order IIR filter predistortion process allows for the processing of residual distortion signals with short durations. This improves the reliability of distortion signal processing in the arbitrary wave generator and enhances the accuracy of distortion signal correction. Moreover, obtaining the predistorted response by measuring the time-domain pulse step response signal of the arbitrary wave generator is a simpler measurement method compared to traditional direct measurement of the system's frequency response.
[0126] Figure 15 is a flowchart of another arbitrary wave generator predistortion method provided in an embodiment of this disclosure. Optionally, based on the above embodiments and referring to Figure 15, the method includes the following steps:
[0127] S510: Obtain the time-domain pulse step response signal output by the arbitrary wave generator.
[0128] S520. Establish a time-domain distortion model based on the time-domain pulse step response signal.
[0129] S530. Correct the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain the first pre-distortion signal.
[0130] S540. Perform differential operation on the first predistortion signal to obtain the impulse response.
[0131] S550. Based on the impulse response and the unit impulse function, the L2 regularization method is used to obtain the predistorted impulse response.
[0132] S560. The predistorted impulse response is used as the coefficient of the FIR filter, and the first predistorted signal is predistorted according to the FIR filter to obtain the second predistorted signal.
[0133] In practical applications, the measured impulse response h(n) often contains noise in addition to the useful signal, which leads to a predistorted impulse response h in the final calculation. inv (n) contains a large error. To reduce the influence of noise, the embodiments of this disclosure employ the L2 regularization method to solve the predistortion impulse response h. inv (n), as follows:
[0134] in, α is the regularization term, and α is the regularization coefficient.
[0135] Figure 16 is an image showing the effect of pre-distortion processing of the FIR filter provided in this embodiment. The obtained pre-distortion impulse response is used as the coefficient of the FIR digital filter, and the FIR digital filter is used to perform pre-distortion processing on the original waveform. The waveform after FIR pre-distortion processing is shown in Figure 16. Curve L1 is the waveform before FIR filter processing, and curve L2 is the waveform after FIR filter processing. As can be seen from Figure 16, the distorted signal is compensated, and the waveform after pre-distortion processing is close to the ideal waveform.
[0136] In this embodiment, the working principle of the arbitrary wave generator predistortion method is as follows: Referring to Figure 15, firstly, the time-domain pulse step response signal output by the arbitrary wave generator is acquired. Then, a time-domain distortion model is established based on the time-domain pulse step response signal. Secondly, the long-time overshoot distortion signal in the time-domain pulse step response signal is corrected according to the time-domain distortion model to obtain a first predistortion signal. The first predistortion signal is differentially processed to obtain an impulse response. The predistorted impulse response is obtained by using L2 regularization based on the impulse response and the unit impulse function. The predistorted impulse response is used as the coefficient of the FIR filter, and the first predistortion signal is predistorted according to the FIR filter to obtain a second predistortion signal. Therefore, it can be seen that long-time overshoot distortion can be corrected using a small number of IIR filters (such as first-order filters). Compared with existing technologies, the correction effect is better, the computational load is smaller, and it is easier to implement. Furthermore, after predistorting the time-domain pulse step response signal using a first-order IIR filter, the predistorted impulse response is obtained by differential operation on the predistorted signal (i.e., the first distorted signal). Using this predistorted impulse response to further predistort the signal after the first-order IIR filter predistortion allows for the processing of residual distortion signals with short durations. This improves the reliability of distortion signal processing in arbitrary wave generators and enhances the accuracy of distortion signal correction. Moreover, the use of L2 regularization in solving the predistorted impulse response reduces the impact of measurement noise. Therefore, obtaining the predistorted response by measuring the time-domain pulse step response signal of an arbitrary wave generator is simpler than the traditional method of directly measuring the system's frequency response. Establishing a time-domain distortion model to correct long-term overshoot distortion signals provides better correction results compared to existing techniques that directly compensate for the system's frequency response. Furthermore, pre-distortion processing of the time-domain impulse step response signal can be achieved using only a small number of IIR and FIR filters. This method can correct distortion signals of long overshoot, short-duration distortion signals, and noise signals. Compared to existing technologies that use FFT pre-distortion processing, this method requires less computation and is easier to implement on real-time processors such as FPGAs. In addition, by fitting the time-domain impulse step response signal to the time-domain distortion model and using L2 regularization to solve the pre-distortion impulse response, the influence of measurement noise can be reduced.
[0137] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this disclosure can be achieved, and this is not limited herein.
[0138] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure. Industrial applicability
[0139] The predistortion method for an arbitrary wave generator disclosed herein obtains the predistortion response by measuring the time-domain pulse step response signal of the arbitrary wave generator, making the measurement method simpler. By establishing a time-domain distortion model, the distortion signal can be compensated, realizing the predistortion processing of the arbitrary wave generator. Fitting the time-domain pulse step response signal and the time-domain distortion model can reduce the influence of measurement noise, thereby correcting distortion signals of long overshoot, short-duration distortion signals, and noise signals with good correction effect. The L2 regularization method can be used to solve the predistortion impulse response, which has a small computational load and is easy to implement with real-time processors such as FPGA. Thus, the predistortion processing of the arbitrary wave generator is realized, improving the reliability of the distortion signal processing of the arbitrary wave generator and improving the accuracy of distortion signal correction.
Claims
1. A predistortion method for an arbitrary wave generator, characterized in that, include: Obtain the time-domain pulse step response signal output by the arbitrary wave generator; A time-domain distortion model is established based on the time-domain pulse step response signal. The first pre-distortion signal is obtained by correcting the long-term overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model.
2. The predistortion method for an arbitrary wave generator according to claim 1, characterized in that, The step of correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model includes: A first-order IIR filter is obtained based on the time-domain distortion model. The distortion model parameters of the first-order IIR filter are obtained based on the time-domain pulse step response signal and the time-domain distortion model. The distortion model parameters of the first-order IIR filter are used to correct the long-term overshoot type distortion signal in the time-domain pulse step response signal.
3. The predistortion method for an arbitrary wave generator according to claim 2, characterized in that, The step of obtaining the first-order IIR filter based on the time-domain distortion model includes: The time-domain distortion model is sequentially subjected to Laplace transform, predistortion transform, and bilinear transform to obtain a first-order IIR filter.
4. The predistortion method for an arbitrary wave generator according to claim 2, characterized in that, The step of obtaining the distortion model parameters of the first-order IIR filter based on the time-domain pulse step response signal and the time-domain distortion model includes: The distortion model parameters of the first-order IIR filter are obtained by fitting the time-domain pulse step response signal and the time-domain distortion model.
5. The predistortion method for an arbitrary wave generator according to claim 4, characterized in that, The distortion model parameters of the first-order IIR filter include the amplitude of the distorted signal and the time constant.
6. The predistortion method for an arbitrary wave generator according to claim 2, characterized in that, The step of correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal based on the distortion model parameters of the first-order IIR filter and the first-order IIR filter includes: The distortion model parameters of the first-order IIR filter are input into the first-order IIR filter to compensate for the long-term overshoot type distortion signal in the time-domain pulse step response signal.
7. The predistortion method for an arbitrary wave generator according to claim 1, characterized in that, The time-domain distortion model is: y(t)=(1+Ae -t / τ )·u(t) Where y(t) is the voltage across the load, u(t) is the step response, A is the amplitude of the distorted signal, and τ is the time constant of the distorted signal.
8. The predistortion method for an arbitrary wave generator according to claim 1, characterized in that, After correcting the long-time overshoot type distortion signal in the time-domain pulse step response signal according to the time-domain distortion model to obtain the first pre-distortion signal, the method further includes: The impulse response is obtained by performing a differential operation on the first predistorted signal. The predistorted impulse response is obtained based on the impulse response and the unit impulse function; The first predistorted signal is predistorted based on the predistorted impulse response to obtain the second predistorted signal.
9. The predistortion method for an arbitrary wave generator according to claim 8, characterized in that, Before performing differential operations on the first predistorted signal, the method further includes: The first predistortion signal is normalized.
10. The predistortion method for an arbitrary wave generator according to claim 8, characterized in that, The step of obtaining the predistorted impulse response based on the impulse response and the unit impulse function includes: obtaining the predistorted impulse response using the L2 regularization method based on the impulse response and the unit impulse function; The step of predistorting the first predistorted signal according to the predistorted impulse response to obtain the second predistorted signal includes: using the predistorted impulse response as the coefficient of an FIR filter, and predistorting the first predistorted signal according to the FIR filter to obtain the second predistorted signal.
Citation Information
Patent Citations
Methods for compensating the frequency characteristics of arbitrary waveform generators
CN102291140A
Quantum bit control system and waveform calibration circuit
CN113760039A
Signal pre-compensation method and system of digital signal generator
CN116707492A
Predistortion method and system for a non-linear device-under-test
US11496166B1