High-resolution adc range extension method, system and apparatus based on programmable compensation

Abstract

The application provides a high-resolution ADC range expansion method, system and device based on program-controlled compensation, which comprises the following steps: based on the ADC output value y AD , judging whether the DAC compensation needs to be started according to a preset threshold value M; if y AD is greater than M, triggering the DAC to output a reverse compensation voltage; if y AD is less than -M, triggering the DAC to output a forward compensation voltage; the DAC dynamically generates a compensation voltage y DA according to the signal amplitude, ensuring that the superposition value of the input signal and the compensation voltage y DA is within the effective range of the ADC; and in the compensated ADC output voltage, the calculated step waveform is subtracted to obtain the final ADC output voltage. Through dynamic feedback compensation and digital step elimination, the application realizes the range expansion without reducing the resolution and interrupting the waveform, solves the inherent defects of the traditional range switching technology, and is especially suitable for the fields, such as seismic observation, which need high-resolution and wide-dynamic-range continuous acquisition.

Description

Technical Field

[0001] This invention relates to the field of data acquisition and signal processing technology, specifically providing a method, system, and device for extending the range of a high-resolution ADC based on programmable compensation. Background Technology

[0002] Dynamic range is one of the important indicators for evaluating the performance of data acquisition devices. Large dynamic range data acquisition systems are widely used in many engineering applications and scientific research. Achieving large dynamic range and wideband observation has always been an important goal in the field of seismic observation. Seismic observation signals have a large dynamic range; they can be weak ground pulsations, background noise signals, or even large earthquake signals of magnitude 8 or higher. For seismic waves at a distance of 10 km from the epicenter, the velocity amplitude can exceed 1 m / s, and the dynamic range of observable seismic waves exceeds 180 dB. The dynamic range of seismic observation is jointly determined by the seismometer and the data acquisition device. Currently, widely used broadband seismometers have a dynamic range between 145 dB and 155 dB, while seismic data acquisition devices have traditionally been limited by the ADC chips used, with a dynamic range of approximately 143 dB (0.02–20 Hz).

[0003] There are various 24-bit ADC chips available for seismic data acquisition. A traditional method to extend the ADC measurement range is range switching: using a small range for small signals and a large range for large signals. However, since seismic signal data acquisition is a continuous process, dynamic range switching not only causes changes in the quantization factor (grid value) but also introduces difficult-to-handle waveform discontinuities. A fatal flaw of the range switching method is that different ranges have different resolutions; the resolution decreases when switching from a small range to a large range. A small range can only achieve high-resolution data acquisition within a relatively small input voltage range close to the baseline; once this range is exceeded and a large range is used, the resolution decreases. This drawback is detrimental to observing local microseismic events triggered by distant surface waves. Summary of the Invention

[0004] This invention aims to overcome the aforementioned shortcomings of traditional range switching techniques and provides a high-resolution ADC range extension method and system based on programmable feedback compensation. This method can significantly extend the effective measurement range of the system while maintaining the original high resolution of the analog-to-digital converter, ensuring continuous and uninterrupted output waveforms, and simultaneously reducing system complexity and cost.

[0005] In a first aspect, the present invention provides a high-resolution ADC range extension method based on programmable compensation, comprising: based on the ADC output value y AD Determine whether DAC compensation needs to be activated based on a preset threshold M; If y AD If y is greater than M, the DAC outputs a reverse compensation voltage; if yAD If the value is less than -M, the DAC will output a positive compensation voltage. The DAC dynamically generates a compensation voltage y based on the signal amplitude. DA Ensure that the input signal matches the compensation voltage y DA The superimposed value is within the effective range of the ADC; The calculated stepped waveform is subtracted from the compensated ADC output voltage to obtain the final ADC output voltage.

[0006] Preferably, the compensated ADC output voltage : in, k A For the gain of the programmable amplifier, N DA The number of bits in the DAC. y DA Set the value to the known value.

[0007] Preferably, the method further includes a transfer function acquisition step: under the condition that the system input signal is 0, a unit step signal is output through the DAC, and the response waveform corresponding to the unit step signal is acquired by the ADC to obtain the transfer function of the ADC output to the DAC signal.

[0008] Preferably, the step waveform is calculated by multiplying the unit step signal response corresponding to the transfer function by the change amplitude of the DAC switching compensation voltage to obtain the step waveform.

[0009] Preferably, a unit step signal is output through a DAC, and the response corresponding to this unit step signal is acquired by an ADC and obtained through real-time online calibration testing. Specifically: When the system input is at the background noise level, a known DAC step signal is actively injected, and its response waveform is acquired by the ADC, and then normalized and denoised. Alternatively, it can be obtained through online learning correction, specifically by using actual DAC compensation events that occur during normal system operation as learning samples, and dynamically correcting the transfer function model through an adaptive algorithm to minimize the residual error at the splicing point.

[0010] Preferably, the ADC and DAC are chips with the same number of bits and the same precision, and they share the same reference source. This setting allows the data shift and splicing error to be controlled within ~1 LSB.

[0011] Preferably, the method further includes limiting the bandwidth and rate of change of the input signal through the low-pass filter to ensure the stitching accuracy of the acquired data.

[0012] Secondly, this invention provides a high-resolution ADC range extension system based on programmable compensation, comprising: a judgment module, used to determine the range extension based on the ADC output value y. AD Determine whether DAC compensation needs to be activated based on a preset threshold M; If y AD If y is greater than M, the DAC outputs a reverse compensation voltage; if y AD If the value is less than -M, the DAC will output a positive compensation voltage. The compensation voltage generation module is used by the DAC to dynamically generate a compensation voltage y based on the signal amplitude. DA Ensure that the input signal matches the compensation voltage y DA The superimposed value is within the effective range of the ADC; The calculation module is used to subtract the calculated stepped waveform from the compensated ADC output voltage to obtain the final ADC output voltage.

[0013] Thirdly, the present invention provides an earthquake data acquisition device that integrates the aforementioned high-resolution ADC range extension system based on programmable compensation.

[0014] The beneficial effects of this invention are as follows: 1. This solution achieves high-resolution ADC range expansion through programmable feedback compensation, avoiding the problem of reduced resolution when switching from a small range to a large range.

[0015] 2. By adopting the principle of programmable feedback compensation, the complex multi-range analog switch network is avoided, resulting in a simpler circuit structure, lower cost, and higher reliability.

[0016] 3. By using digital algorithms to compensate for the response caused by the DAC step jump in real time, seamless waveform splicing is achieved, and the output data stream is continuous, making it particularly suitable for fields such as earthquakes that require long-term continuous recording. Attached Figure Description

[0017] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. Furthermore, similar numbers in the drawings are used to denote similar components, wherein: Figure 1 This is a block diagram illustrating the dynamic range extension of a level shifting technique according to an embodiment of the present invention. Figure 2 This is a waveform example of a DAC output value change according to an embodiment of the present invention; Figure 3 This is a level shifting method for extending the measurement range according to an embodiment of the present invention; Figure 4This is a normalized response waveform diagram of the ADC output to the change of the DAC output according to an embodiment of the present invention. Figure 5 This is a basic rule diagram for level shifting via a DAC according to an embodiment of the present invention. Detailed Implementation

[0018] Some embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] Example 1 like Figure 1-5 As shown, this invention provides a high-resolution ADC range extension method based on programmable compensation, comprising: Based on ADC output value y AD Determine whether DAC compensation needs to be activated based on a preset threshold M; If y AD If y is greater than M, the DAC outputs a reverse compensation voltage; if y AD If the value is less than -M, the DAC will output a positive compensation voltage. The DAC dynamically generates a compensation voltage y based on the signal amplitude. DA Ensure that the input signal matches the compensation voltage y DA The superimposed value is within the effective range of the ADC; The calculated stepped waveform is subtracted from the compensated ADC output voltage to obtain the final ADC output voltage.

[0020] In this embodiment, the compensation voltage generated by the DAC circuit must be accurate to ensure high linearity of the analog-to-digital conversion; or its error must match the error of the high-resolution ADC to prevent a step in the final output signal caused by the DAC setting. During implementation, when the ADC output approaches its upper limit, the DAC is set so that the superposition of the DAC compensation voltage and the input signal is close to the center of the ADC measurement range, such as... Figure 2 As shown. To ensure that the ADC input voltage does not exceed its range, a reference value M for dynamic compensation activation needs to be set, meaning the voltage resulting from the superposition of the DAC output voltage and the input signal must not exceed M. The basic rules for achieving dynamic output compensation voltage in the DAC are as follows: Figure 5 As shown.

[0021] In one embodiment, the DAC output is a step signal. The ADC's response to the DAC's step signal can be obtained through offline calibration testing, or through online testing, normalized superposition, and other numerical calculation methods for model correction. Figure 2The step response shown can be compensated for numerically, thus eliminating it from the output data. This process allows the output data encoding to exceed the limitations of the ADC chip, while the input voltage of the ADC chip remains within its effective range, thereby extending the measurement range.

[0022] Specifically, a unit step signal is output through the DAC, and the response corresponding to this unit step signal is acquired by the ADC through real-time online calibration testing. This involves actively injecting a known unit step signal into the system and using the ADC to capture its output change to directly measure its transfer function; that is, the actual test record of the ADC's response to the DAC unit step signal. The specific steps are as follows: ① When the system monitors the input signal in real time and it is at a background noise level (small vibration), the ADC output value... or When this condition persists for a period of time, it is considered a "calibration window"; ② Control the DAC to generate a set of precise code value transformations. y DA_test A bidirectional step jump is used to eliminate the linear bias caused by signal drift (e.g., by first adding...). V step Further reduction V step ); ③ The ADC synchronously records the output sequence from the moment the DAC switches at a high sampling rate. y DA_test The record length L should cover the filter setup time; ④ Subtract the average value before the step jump from the collected sequence. y Baseline This yields the pure response waveform, which is then divided by the corresponding DAC switching amplitude (i.e., converted to the order of magnitude of the ADC LSB) to obtain the normalized unit step response. h [ n ] can be represented as ,in k A is the gain constant of the programmable amplifier from DAC to ADC.

[0023] ⑤ To suppress random noise (ground pulsation background noise), repeat the above process. N The number of times (e.g., 64 or 128) will yield the following results. h [ n Alignment and stacking are performed, and the average value is taken. Through stacking, irrelevant background noise will be... The ratio is reduced, thus obtaining a transfer function model with an extremely high signal-to-noise ratio.

[0024] ⑥ The final result h [ n The sequence is stored in a compensation register or lookup table (LUT) for subsequent real-time data deduction.

[0025] Alternatively, it can be obtained through online learning and correction: Model correction is performed using numerical calculation methods such as online testing and normalized superposition. Real range switching events occurring during normal system operation are used as "learning samples," and the response model is dynamically corrected using adaptive algorithms (such as normalized superposition or least mean square algorithm). The specific steps are as follows: ① The system loads an initial unit step response model calibrated in the laboratory (or at the factory) or pre-calculated. h init [ n ]; ② When signal X triggers normal DAC compensation action (ADC output value) or (Time), record the current DAC change. y DA_actual And the switching time t0, and extract the ADC output sequence before and after the switching point. y raw [ n ]; ③ Use the current model h [ n The predictable step waveform is: The compensated and spliced ​​waveform can be obtained as follows: If the model h [ n If it is perfect, then the spliced ​​waveform will be... It is continuous and smooth, if the model h [ n There is an error; a tiny "jump" will appear at the splicing point, which is the residual error e. ④ In each compensation step, the cross-correlation between the error e and the step input is calculated, and the model is updated. ,in To learn the step size, the model uses the least mean square algorithm and multiple compensation learning steps to minimize the residual error e. It continuously converges to the real physical transfer function.

[0026] ⑤ Or through a normalized superposition correction model h [ nIn earthquake observation, because seismic waves are random, residual waveform segments after multiple actual switching can be normalized and superimposed. Real signals (uncorrelated) will cancel each other out, while fixed biases (correlated) caused by model inaccuracies will become apparent. Numerical calculation methods (such as least squares fitting) can be used to correct the model. h [ n The parameters (such as cutoff frequency, gain, delay) are used to minimize the residual error after superposition.

[0027] In one embodiment, the scheme for extending the dynamic range is as follows: Figure 2 As shown in the figure, the low-pass filter is used to limit the input signal bandwidth and rate of change. The effective range of the high-resolution ADC is determined by the reference voltage. V REF The decision was made, and the system's full-scale range was set to... k A V REF The range of the input signal is much larger than that of the ADC. The function of the DAC is to compensate for the input signal, so that the superposition of the DAC output and the system input signal is within the effective range of the ADC. At this time, the output of the ADC can be expressed as: ,in, k A For the gain of the programmable amplifier, N AD The number of bits in the DAC. y AD Since the set value is known, the input value can be calculated using this formula. x The quantity value, input quantity x The range of variation is not less than k A V REF .

[0028] In one embodiment, to ensure accurate calibration of the compensation voltage and control the translation and data splicing errors within ~1 LSB, an ADC and DAC of the same precision (same number of bits) are selected to ensure that the data translation and splicing errors are controlled within ~1 LSB. At the same time, the DAC and ADC are designed to share a high-precision reference voltage source that matches the number of bits of the ADC, ensuring that the data splicing accuracy is not affected by the reference source.

[0029] In one embodiment, by setting a low-pass filter to limit the input signal bandwidth and increasing the sampling rate, the accuracy of the acquired data splicing can be guaranteed.

[0030] In one embodiment, the high-resolution ADC chip in this method always operates in a high-resolution state with a small range. When the input voltage is about to exceed the ADC's range, a compensation voltage is output through the DAC chip. The voltage resulting from the superposition of the DAC output voltage and the input signal approximates the baseline, thereby preventing the ADC chip from exceeding its range. Figure 3 As shown, a change in the DAC output voltage inevitably creates a step in the ADC output data, such as... Figure 2 As shown, using an algorithm to eliminate this step completes the waveform splicing.

[0031] Furthermore, with zero input, by using an ADC to acquire the step signal from the DAC output, the transfer function of the ADC output to the DAC signal can be obtained. The response of this transfer function to a unit step signal is as follows: Figure 4 As shown, multiplying the unit step signal response by the DAC change amplitude (i.e., the size of the ADC output step) yields the step-compensated waveform, thus effectively eliminating steps in the ADC output data and accurately stitching waveform data. This extended technique achieves balanced high resolution across the entire large range. Figure 5 The basic rules for output compensation voltage are shown.

[0032] This method achieves high-resolution ADC range extension through programmable feedback compensation, avoiding the resolution degradation that occurs when switching from a small to a large range. Traditional range switching techniques, such as multi-range and multi-core solutions, involve large circuit sizes. This solution reduces the circuit size and cost of range switching through programmable feedback compensation.

[0033] Example 2 like Figure 1 As shown, the present invention provides a high-resolution ADC range extension system based on programmable compensation, comprising: an analog-to-digital converter configured to operate at a fixed first range; a digital-to-analog converter for outputting the compensation voltage; and a controller connected to the output terminal of the analog-to-digital converter and the control terminal of the digital-to-analog converter, wherein the controller is configured to perform the steps of real-time monitoring, programmable digital-to-analog converter changing output, acquiring a step response model, and performing digital compensation based on the model.

[0034] In one embodiment, the system further includes a low-pass filter connected to the system input for limiting the bandwidth and rate of change of the input signal.

[0035] Example 3 This invention provides a seismic data acquisition device that integrates the aforementioned high-resolution ADC range extension system based on programmable compensation.

[0036] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the original technical features, and the technical solutions resulting from these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for extending the range of a high-resolution ADC based on programmable compensation, characterized in that, include: Based on ADC output value y AD Determine whether DAC compensation needs to be activated based on a preset threshold M; If y AD If y is greater than M, the DAC outputs a reverse compensation voltage; if y AD If the value is less than -M, the DAC will output a positive compensation voltage. The DAC dynamically generates a compensation voltage y based on the signal amplitude. DA Ensure that the input signal matches the compensation voltage y DA The superimposed value is within the effective range of the ADC; The calculated stepped waveform is subtracted from the compensated ADC output voltage to obtain the final ADC output voltage.

2. The method according to claim 1, characterized in that, Compensated ADC output voltage : in, k A For the gain of the programmable amplifier, N DA The number of bits in the DAC. y DA Set the value to the known value.

3. The method according to claim 1, characterized in that, The method further includes a transfer function acquisition step: under the condition that the system input signal is 0, a unit step signal is output through the DAC, and the response waveform corresponding to the unit step signal is acquired by the ADC to obtain the transfer function of the ADC output to the DAC signal.

4. The method according to claim 3, characterized in that, The step waveform is calculated by multiplying the unit step signal response corresponding to the transfer function by the change amplitude of the DAC switching compensation voltage to obtain the step waveform.

5. The method according to claim 4, characterized in that, The unit step signal is output by the DAC, and the response corresponding to the unit step signal is acquired by the ADC and obtained through real-time online calibration test. Specifically, when the system input is at the background noise level, a known DAC step signal is actively injected, and its response waveform is acquired by the ADC, and then obtained after normalization and noise reduction. Alternatively, it can be obtained through online learning correction, specifically by using actual DAC compensation events that occur during normal system operation as learning samples, and dynamically correcting the transfer function model through an adaptive algorithm to minimize the residual error at the splicing point.

6. The method according to claim 1, characterized in that, The ADC and DAC are chips with the same number of bits and the same precision, and they share the same reference source. This setting allows the data shifting and splicing errors to be controlled within ~1 LSB.

7. The method according to claim 1, characterized in that, The method further includes limiting the bandwidth and rate of change of the input signal through the low-pass filter to ensure the stitching accuracy of the acquired data.

8. A high-resolution ADC range extension system based on programmable compensation, characterized in that, include: The judgment module is used to determine the ADC output value y. AD Determine whether DAC compensation needs to be activated based on a preset threshold M; If y AD If y is greater than M, the DAC outputs a reverse compensation voltage; if y AD If the value is less than -M, the DAC will output a positive compensation voltage. The compensation voltage generation module is used by the DAC to dynamically generate a compensation voltage y based on the signal amplitude. DA Ensure that the input signal matches the compensation voltage y DA The superimposed value is within the effective range of the ADC; The calculation module is used to subtract the calculated stepped waveform from the compensated ADC output voltage to obtain the final ADC output voltage.

9. A seismic data acquisition device, characterized in that, It integrates the high-resolution ADC range extension system based on programmable compensation as described in claim 8.

Images