WO2014038056A1 - インターリーブa/d変換器 - Google Patents
インターリーブa/d変換器 Download PDFInfo
- Publication number
- WO2014038056A1 WO2014038056A1 PCT/JP2012/072837 JP2012072837W WO2014038056A1 WO 2014038056 A1 WO2014038056 A1 WO 2014038056A1 JP 2012072837 W JP2012072837 W JP 2012072837W WO 2014038056 A1 WO2014038056 A1 WO 2014038056A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- analog
- digital converter
- input
- unit
- adc
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/06—Continuously compensating for, or preventing, undesired influence of physical parameters
- H03M1/0617—Continuously compensating for, or preventing, undesired influence of physical parameters characterised by the use of methods or means not specific to a particular type of detrimental influence
- H03M1/0626—Continuously compensating for, or preventing, undesired influence of physical parameters characterised by the use of methods or means not specific to a particular type of detrimental influence by filtering
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/06—Continuously compensating for, or preventing, undesired influence of physical parameters
- H03M1/0617—Continuously compensating for, or preventing, undesired influence of physical parameters characterised by the use of methods or means not specific to a particular type of detrimental influence
- H03M1/0624—Continuously compensating for, or preventing, undesired influence of physical parameters characterised by the use of methods or means not specific to a particular type of detrimental influence by synchronisation
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/1205—Multiplexed conversion systems
- H03M1/121—Interleaved, i.e. using multiple converters or converter parts for one channel
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/1205—Multiplexed conversion systems
- H03M1/121—Interleaved, i.e. using multiple converters or converter parts for one channel
- H03M1/1215—Interleaved, i.e. using multiple converters or converter parts for one channel using time-division multiplexing
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/124—Sampling or signal conditioning arrangements specially adapted for A/D converters
- H03M1/1245—Details of sampling arrangements or methods
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/34—Analogue value compared with reference values
- H03M1/36—Analogue value compared with reference values simultaneously only, i.e. parallel type
Definitions
- the present invention relates to an interleaved A / D converter.
- ADC analog-to-digital converter
- ADCs used in measuring instruments such as inspection devices used in semiconductor manufacturing
- miniaturization advances high-speed, high-resolution medical diagnostic devices for more accurate diagnosis
- the rate at which a single ADC can sample while maintaining an effective resolution of 10 bits or more is considered to be limited to about 250 MHz because of the rate limiting derived from the frequency characteristics of the transistors constituting the ADC.
- time interleaving that parallelizes a plurality of ADCs, sets a time difference between the respective sampling timings, and increases the conversion speed as a whole has attracted attention.
- this technique also has a limited resolution that can be guaranteed due to mismatches such as DC offset, conversion gain variation, and sampling clock timing shift in each paralleled ADC.
- Patent Documents 1 and 2 are disclosed in Patent Documents 1 and 2 and Non-Patent Documents 2 and 3 to solve this problem. Heading to
- Non-Patent Document 1 shows one typical configuration example of an interleave ADC.
- 80 ADCs of 250 MS / s and 8 bits are arranged in parallel to realize an ADC having a sampling rate of 20 GS / s.
- the analog input signal conveyed to the ADC is received by a high-speed buffer and distributed to a parallel ADC sampling circuit (a circuit described as 80 T / H in Non-Patent Document 1) in the subsequent stage. .
- each ADC sampling circuit releases charge to the input at the sampling clock edge, and other parallel ADCs sample signals disturbed by this charge, thereby limiting the effective resolution.
- the problem that it will be done arises.
- the important point here is that the digital correction method cannot be effective against irregular factors, and directly leads to a decrease in effective resolution.
- the existing digital correction method cannot be applied because the charge released from the ADC input depends irregularly on the input signal. This problem will not be a problem if the effective resolution of 8 bits targeted in Non-Patent Document 1 is used, but it becomes a new problem that it becomes a barrier to realizing a resolution larger than this.
- Each ADC sampling circuit arranged in parallel in the interleave ADC performs charge discharge depending on the input signal at the time of sampling, and the other ADC arranged in parallel samples the disturbed input signal, thereby reducing the resolution of the interleave ADC.
- the problem to be solved is a problem to be solved by the invention.
- a common differential sampling clock signal is input to one set of ADCs that are sampled with a ⁇ phase shift of each ADC paralleled in the interleave ADC, and a separate buffer is provided in front of the input sampling circuit,
- the above problem can be improved by isolating the common analog input signal line and the ADC input terminal and digitally correcting characteristic deterioration due to buffer insertion.
- the sampling timing of a pair of ADCs having the highest possibility of interference due to the symmetry of the differential signal is accurately engraved with a ⁇ phase difference by a common sampling clock, so that the ideal ADC sampling timing is obtained.
- the interference between the ADC inputs due to an error from the value is suppressed, and further, the input signal is transmitted to the ADC sampling circuit of the subsequent stage by the buffer, but the charge discharged from the ADC sampling circuit is shielded by the buffer and not transmitted to the common analog input signal line.
- Other parallel ADCs do not receive interference with each other, and characteristic deterioration due to buffer insertion is suppressed by digital correction, so that a reduction in resolution can be prevented.
- FIG. 6 is a diagram illustrating Example 1;
- FIG. 3 is a diagram illustrating an embodiment of a digital correction unit in Example 1.
- FIG. 3 is a diagram illustrating an example of an embodiment of a delay device in Example 1.
- FIG. 6 is a diagram illustrating a second embodiment.
- FIG. 6 is a diagram illustrating an embodiment of a buffer unit in Example 2.
- FIG. 6 is a diagram illustrating Example 3;
- FIG. 6 is a diagram illustrating Example 4;
- FIG. 10 is a diagram for explaining a clock phase relationship in the sixth embodiment.
- FIG. 1 shows an embodiment of an interleave ADC according to the first embodiment.
- the input terminal of the reference ADC 104 is connected to the common analog input signal line.
- the ADC 101-k and the reference ADC 104 include an input sampling circuit.
- the digital output of each ADC 101-k is connected to the digital correction unit 102, and a direct current offset, a conversion gain variation, and a sampling timing shift generated in each AD conversion path based on a comparison between the output of the ADC 104 and the output of each ADC 101-k.
- the mismatch between each AD conversion path is corrected.
- the ideal sampling timing refers to a timing at which the sampling intervals between adjacent ADCs 101-k are equal for all sampling timings.
- FIG. 2 shows a specific configuration of the digital correction unit.
- the AD conversion result output from each ADC 101-k is connected to a gain correction unit 201-k (k is an integer of 1 to N) to correct the gain mismatch, and the corrected output is supplied to the gain correction unit 201-k.
- the offset is corrected by the connected offset correction unit 202-k, and the corrected output result is further corrected by the skew correction unit 203-k and output to the multiplexer 103.
- the skew is a deviation from the ideal sampling timing of the sampling in each ADC 101-k.
- the gain correction unit, the offset correction unit, and the skew correction unit make a difference between the results output from the reference ADC 104 and the skew correction unit by the subtractor 204-k, and reduce the difference value (hereinafter referred to as e (k)). Each correction value is determined.
- e (k) the difference value
- the AD conversion data (d0 (k, t)) output from each unit ADC 101-k is input to the gain correction unit 201-k.
- the gain correction unit 201-k the difference e (k) from the ideal AD conversion result obtained by the subtractor 204-k is converted by the multiplier 205-k into the AD conversion result d0 (k, t), Multiply by the gain correction loop factor ⁇ G.
- the output result of the multiplier 205-k is input to the adder 206-k, and is combined with the adder output one clock cycle before delayed by the delay unit 207-k.
- the output value ⁇ G (k, t) of the delay unit 207-k is a value obtained by integrating the output of the multiplier 205-k from the start of digital correction.
- the structure constituted by the adder 206-k and the delay unit 207-k is generally known as an integrator.
- ⁇ G (k, t) corresponds to a gain error between the reference ADC 104 and the unit ADC 101-k, and ⁇ G (k, t) is multiplied by the AD conversion data d0 (k, t) by the multiplier 208-k to obtain a gain. It is output as the output d1 (k, t) of the correction unit 201-k.
- the output result d1 (k, t) corrected by the gain correction unit 201-k is input to the offset correction unit 202-k.
- the difference value e (k) from the ideal conversion result and the offset correction loop coefficient ⁇ OFS are multiplied by the multiplier 209-k.
- the output result of the multiplier 209-k is input to the adder 210-k, and is combined with the adder output one clock cycle before delayed by the delayer 211-k.
- the output value ⁇ V (k, t) of the delay unit 211-k is a value obtained by integrating the output of the multiplier 209-k from the start of digital correction.
- ⁇ V (k, t) corresponds to an offset error between the reference ADC 104 and the unit ADC 101-k, and the input data d1 (k, t) is subtracted by ⁇ V (k, t) by the adder 212-k to correct the offset.
- the output is output as d2 (k, t) of the unit 202-k.
- the output result d2 (k, t) corrected by the offset correction unit 202-k is input to the skew correction unit 203-k. Further, a time differential value of d2 (k, t), d (d2 (k, t-XT)) / dt is also input to the skew correction unit 202-k.
- XT [s] is a delay time necessary for calculating a differential value by an FIR filter or the like
- T [s] is one clock cycle period.
- d (d2 (k, t ⁇ XT)) / dt can be calculated by an FIR filter or the like, and a specific method thereof will be described later.
- the difference e (k) from the ideal AD conversion result obtained by the subtractor 204-k is d (d2 (k, t-XT) by the multiplier 213-k. ) / Dt and the skew correction loop coefficient ⁇ SKEW.
- the output result of the multiplier 213-k is input to the adder 214-k, and is combined with the adder output one clock cycle before delayed by the delay unit 215-k.
- the output value ⁇ t (k, t) of the delay unit 215-k is a value obtained by integrating the output of the multiplier 214-k from the start of digital correction.
- ⁇ t (k, t) corresponds to a sampling timing error between the reference ADC 104 and the unit ADC 101-k
- ⁇ t (k, t) is obtained by the multiplier 216-k by d (d2 (k, t ⁇ XT)) / Multiply by dt.
- the result, ⁇ t (k, t) ⁇ d (d2 (k, t ⁇ XT)) / dt corresponds to the error of the AD conversion result caused by the sampling timing error.
- the input data d2 (k, t) is input to the delay unit 217-k, and the output d2 (k, t-XT) of the delay unit 217-k is the output of the multiplier 216-k in the adder 218-k.
- a certain ⁇ t (k, t) ⁇ d (d2 (k, t ⁇ XT)) / dt is subtracted and output as an output d3 (k, t) of the skew correction unit 203-k.
- FIG. 9A shows an example of an embodiment of an FIR filter for obtaining a differential value d (d2 (k, t ⁇ XT)) / dt.
- the FIR filter receives the offset correction unit output d2 (k, t) of FIG. 2, and is connected to each delay unit 901-n (n is an integer of 1 to K) connected to the input and to the subsequent stage of the delay unit, and the weight A tap unit 902-n for applying a coefficient and an adder 903 for adding the outputs multiplied by the weighting coefficient.
- different ADC outputs can be input. In this example, four ADCs are interleaved.
- the output of ADC 101-k is input to delay block 901-n or tap section 902-n, respectively, and the output of delay block 901-n is input to 902-n and multiplied by a weighting factor.
- the outputs multiplied by the respective weighting factors are added by an adder 903 and output.
- the differential value d (d2 (k, t ⁇ XT)) / dt can be obtained from the adder 903 by setting the weighting coefficient in each tap unit 902-n to an appropriate value.
- the AD conversion results of the ADCs 101-k corrected by the digital correction unit 102 are integrated by the multiplexer 103 and output as a series of time series data.
- Each ADC 101-l (l is an odd number from 1 to N) has a pair of ADCs 101-m (m is an even number from 1 to N), and the positive input clock terminal of each ADC 101-l is a pair of ADCs 101-l.
- the negative input clock terminal of each ADC 101-l is connected to the positive input clock terminal of the ADC 101-m paired with the negative input clock terminal of -m.
- the differential clock signal line is connected to the differential input clock terminals of the ADCs 101-1 and 101-m connected to each other via the delay unit 105-k, and the sampling clock signal having the frequency fs is input.
- the positive and negative two signal lines of the differential input signal line may be connected to either of the two signal lines connecting the clock terminals of the ADC 101-l and ADC 101-m, respectively.
- the value of the analog input signal Vin AD-converted at the sampling rate fs is output as a digital signal from each ADC 101-k, and they are complementary to each other.
- a digital output Dout obtained by AD conversion at an interval of t + 1 / (fs ⁇ N) is obtained from the multiplexer, and as a result, AD conversion is performed at a sampling rate of fs ⁇ N.
- each ADC 101-l is connected to the negative input clock terminal of the pair of ADC 101-m, and the negative input clock terminal of each ADC 101-l is connected to the positive input clock terminal of the pair of ADC 101-m. It may be the point that has been done. With this connection, the ADC 101-l and ADC 101-m that are paired share one type of sampling clock signal, but their clock input terminals are connected to each other in the positive and negative directions.
- the ADC 101-m operates assuming that sampling clocks having opposite phases are input to each other.
- the paired ADCs 101-k sample the input signals with a time difference of 1 / (fs ⁇ 2) and perform AD conversion.
- the ADC 101-k turns the switch on and off by an internal sampling mechanism at the rising and falling moments of the sampling clock input to the ADC 101-k (hereinafter referred to as clock edges).
- a kickback phenomenon is observed in which the charge accumulated in the sampling capacitor and the charge accumulated in the transistor constituting the switch are released from the input terminal of the ADC 101-k, and the input signal is disturbed.
- ADC 101-k samples the input at the falling edge of the sampling clock and shifts to the sampling period at the rising edge of the sampling clock.
- An ADC 101-k is most strongly affected by the input disturbance that occurs immediately before it.
- ADC 1 and ADC 2 connected in parallel to the common analog input terminal.
- sampling clock signals generated by different paths are input to these ADCs and each has a phase difference of 1 / (fs ⁇ 2)
- the sampling clock phase difference generation method and the influence of the wiring path Therefore, the timing of the clock edge is shifted between ADC 1 and ADC 2 and an event occurs in which ADC 2 disturbs the analog input signal at the clock edge input to ADC 2 just before ADC 1 samples.
- the kickback component is about 3 mV. If the ADC 1 samples the input signal as it is before the kickback component converges, it becomes less than 8 bits in terms of effective resolution.
- the sampling clock input to the ADC No. 2 is the same as the sampling clock input to the ADC No. 1 although the connection is different. Therefore, the clock edge timing is almost the same. Therefore, if the time for the disturbance of the input generated by the ADC 2 to reach the input of the ADC 1 through the signal line is taken into consideration, the input disturbance generated by the ADC 2 does not affect the sampling of the ADC 1. This is because if the time at which the ADC 1 generates a disturbance component with respect to the input signal line at the clock edge timing is t1, the ADC 2 samples at exactly the same time as t1.
- ADC 2 is not affected by the disturbance caused by ADC 1. This is also the case when ADC part 2 samples, and so on between the other ADC 101-k pairs.
- This embodiment avoids such a phenomenon that the input signal just disturbed due to the timing shift between the sampling clocks is sampled, and at least 1 / (fs is required for the disturbed input to converge. ⁇ The period of N) is secured. Further, this embodiment is characterized by further comprising a digital correction unit, and the reason and effect will be described below.
- a pair of ADC 101-l and ADC 101-m share one type of sampling clock signal, and the clock input terminals are connected to each other in the positive and negative directions, so that between ADC 101-l and ADC 101-m.
- the operation is performed when sampling clocks having inversion phases are input, and as a result, the paired ADCs 101-k sample the input signals with a time difference of 1 / (fs ⁇ 2) and perform AD conversion. As described above, this ensures that the sampling timing between the paired ADCs is exactly 1 / (fs ⁇ 2) time difference from the nature of the differential sampling clock signal.
- this sampling timing error is corrected using the reference ADC 104 and the digital correction unit 102, thereby solving the above-described problem that the sampling timing between ADCs deviates from the ideal value.
- an inductance element 300 and a capacitance element An LC filter delay device using 301 is shown in FIG.
- filter types such as a Butterworth filter and a Chebyshev filter.
- an LC filter called a Bessel filter with little change in delay characteristics in a wide frequency band will be described as an example.
- 3B shows element constants of a fifth-order ⁇ -type Bessel filter normalized to a cutoff frequency of 1 ⁇ 2 ⁇ [Hz] and an impedance of 1 [ ⁇ ].
- the impedance of the conversion destination is R [ ⁇ ]
- K R / 1
- K is applied to all inductance element values, and all capacitance element values are converted. Is divided by K.
- M fc ⁇ 2 ⁇ .
- the delay amount of the signal input to the filter depends on the frequency of the input signal, the cutoff frequency of the filter, and the order of the filter, an appropriate amount of delay occurs with respect to the sampling clock frequency fs using a circuit simulator or numerical calculation. What is necessary is just to calculate M and to design a filter.
- the capacitance elements 301 to a variable capacitor 303, it is possible to provide an adjustable width of the delay amount.
- FIG. 4 shows an embodiment of an interleave ADC according to the second embodiment.
- This example is shown as an example of a solution for further improving the problem that the kickback component of the ADC in Example 1 disturbs the input signal.
- the input terminal of the buffer unit 401-k is connected to the common analog input signal line.
- the input terminal of the buffer unit 400 is connected to the common input signal line, and the input terminal of the reference ADC 104 is connected to the output terminal of the buffer unit 400.
- the ADC 101-k and the reference ADC 104 include an input sampling circuit.
- each ADC 101-k is connected to the digital correction unit 102, and in the same way as in the first embodiment, the comparison with the output of the ADC 104 causes a direct current offset, conversion gain variation, sampling timing shift, etc. A mismatch between AD conversion paths is corrected.
- the AD conversion results of the ADCs 101-k corrected by the digital correction unit 102 are integrated by the multiplexer 103 and output as a series of time series data.
- Each ADC 101-l (l is an odd number from 1 to N) has a pair of ADCs 101-m (m is an even number from 1 to N), and the positive input clock terminal of each ADC 101-l is a pair of ADCs 101-l.
- the negative input clock terminal of each ADC 101-l is connected to the positive input clock terminal of the ADC 101-m paired with the negative input clock terminal of -m.
- the differential clock signal line is connected to the differential input clock terminals of the ADCs 101-1 and 101-m connected to each other via the delay unit 105-k, and the sampling clock signal having the frequency fs is input.
- the positive and negative two signal lines of the differential input signal line may be connected to either of the two signal lines connecting the clock terminals of the ADC 101-l and ADC 101-m, respectively.
- a feature of this embodiment is a buffer unit 401-k provided between the ADC 101-k and the common analog input signal line.
- FIG. 5A showing an embodiment of the buffer unit has a negative feedback configuration in which an analog signal input line is connected to the non-inverting input terminal of the operational amplifier 500 and the output terminal and the inverting input terminal of the operational amplifier 500 are connected.
- the non-inverting input terminal of the operational amplifier 500 is grounded via the capacitor 501.
- a differential operational amplifier 502 may be connected to the output of the operational amplifier 500 to differentiate the analog input signal as shown in FIG. 5B. It is possible, and it does not matter whether the analog input terminal of the unit ADC is single-ended or differential.
- the first embodiment by using a common differential clock signal as a sampling clock between ADCs in a pair, it is possible to prevent sampling of a disturbed signal due to a shift in sampling timing between ADCs. However, no countermeasure can be taken in the case where the influence of the disturbance caused by another ADC not paired remains.
- the buffer unit 401-k in the second embodiment has a low output impedance as a characteristic peculiar to the buffer element, and the non-inverting input terminal and the inverting input terminal of the operational amplifier 500 constituting the buffer unit are insulated. Only parasitic capacitance exists. Therefore, the signal disturbance generated on the output side of the buffer unit 401-k is hardly transmitted to the input side of the buffer unit, and the disturbance component transmitted through the parasitic capacitance is also a parasitic capacitance between the capacitance of the capacitor 501 and the input terminal of the operational amplifier 500. Since the voltage is divided according to the ratio of the magnitudes, the signal disturbance component transmitted to the input signal side is very small. As described above, in the present embodiment, by providing the buffer unit 401-k, it is possible to prevent the kickback component from being transmitted to the other ADC 101-k. It is possible to solve the back problem and secure an effective resolution.
- FIG. 5 shows an embodiment of an interleave ADC according to the third embodiment.
- the characteristics of the buffer section are digitally corrected in order to prevent the deterioration of the characteristics due to the provision of the buffer section disclosed in the second embodiment.
- N ADCs 101-k 1 to N natural numbers
- buffer units 401-k connected to the analog input terminals of the ADCs 101-k
- the ADCs 101-k are connected. It comprises a multiplexer 103 connected to the output of the digital correction unit 102, a buffer unit 400 whose input terminal is connected to the common input signal line, and a reference ADC 104 whose input terminal is connected to the output terminal of the buffer unit 400.
- the ADC 101-k and the reference ADC 104 include an input sampling circuit.
- Each buffer unit 401-k receives an analog signal Vin, and ADC 101-k receives a sampling clock signal having a frequency fs.
- the digital output of each ADC 101-k is connected to the digital correction unit 102, and a mismatch between the AD conversion paths such as a DC offset, conversion gain variation, and sampling timing shift generated in the AD conversion path from comparison with the output of the ADC 104 Correct.
- This correction can be performed in the same manner as described in the first embodiment.
- the digital correction unit 102 further corrects the voltage offset, voltage gain, and signal band variation of the buffer unit 401-k and the buffer unit 400.
- the AD conversion results of the respective ADCs 101-k corrected by the digital correction unit 102 are integrated by the multiplexer 103 and output as a series of time series data.
- the sampling clock signals input to each ADC 101-k have a phase difference with each other, and those having the closest phases have a phase difference of approximately 1 / (fs ⁇ N).
- the feature of this embodiment is that the digital correction unit 102 simultaneously corrects characteristic variations caused by the buffer unit 401-k, particularly signal band variations.
- the analog signal passes through a different buffer unit for each AD conversion path of the interleaved ADC, it is affected by variations in the signal band of the operational amplifier constituting the buffer unit 401-k.
- Such a signal band variation of the buffer unit causes a reduction in the resolution of the AD conversion result synthesized by the multiplexer 103 as well as other variations.
- the digital correction unit 102 periodically specifies the amount of mismatch between these paths by comparing the sampling of each ADC 101-k, the AD conversion result of the ADC 104, and the AD conversion result of each ADC 101-k, The digital output value of each ADC 101-k is recalculated so as to cancel these mismatches, and is output to the multiplexer.
- the configuration of the digital correction unit is the same as the configuration of FIG. 2, and the correction parameter of the signal band is reduced to the same parameter as the correction parameter of the sampling timing described in the first embodiment, thereby preventing a reduction in resolution due to mismatch between paths. be able to. The reason is as follows.
- the frequency characteristic of the buffer unit 401-k can be generally expressed as a transfer function of Formula 1.
- Equation 2 can be approximated in the form of Equation 2 in a frequency band lower than the frequency band of the buffer unit 401-k.
- the AD conversion result affected by ⁇ t can be expressed by a transfer function as shown in Equation 3.
- a component of ⁇ t ⁇ s + ( ⁇ t 2 ) / 2 ⁇ s 2 is an error caused by ⁇ t.
- the digital correction of the sampling timing described above in Example 1 Delta] t is the coefficient of the s, seeking a Delta] t 2/2 is the coefficient of s 2, performing a process of subtracting from the AD conversion result to calculate the error term ing.
- the influence due to the sampling timing error and the influence caused by the frequency characteristic of the buffer occur simultaneously, it can be expressed as a transfer function as shown in Expressions 2 and 3 to 4.
- Equation 4 it can be seen that the influence of the sampling timing error and the frequency characteristic error between the buffer units can be corrected by obtaining the s coefficient and the s 2 coefficient. It can be seen that it can be corrected by the method. In this embodiment, by obtaining this coefficient, the sampling timing mismatch and the frequency characteristic mismatch between the buffer units are corrected simultaneously. Further, the gain correction and the offset correction of the buffer unit can be performed with the configuration of FIG. 2 similarly to the first embodiment, and it is possible to further prevent the effective resolution from being lowered.
- FIG. 7 shows an embodiment of an interleave ADC according to the fourth embodiment.
- the simultaneous correction structure of the characteristic mismatch between the buffers and the sampling timing described in the third embodiment, and the delay device 700 capable of adjusting the phase of the clock input to the reference ADC 104. Is provided.
- each ADC 101 -k is connected to the multiplexer 103.
- Each ADC 101-l (l is an odd number from 1 to N) has a pair of ADCs 101-m (m is an even number from 1 to N), and the positive input clock terminal of each ADC 101-l is a pair of ADCs 101-m
- the negative input clock terminal of each ADC 101-l is connected to the positive input clock terminal of the ADC 101-m that forms a pair.
- a differential clock signal line is connected to the differential input clock terminals of the ADCs 101-1 and 101-m connected to each other, and a sampling clock signal having a frequency fs is input.
- the sampling clock signal input to each differential clock signal line is k / (fs ⁇ N) (k) generated by the delay unit 105-k from the reference differential clock signal (clk +, clk ⁇ ).
- N / 2 types of clocks having a phase of 0 to N / 2-1 (natural number).
- the differential clock signal (clk +, clk ⁇ ) may use a balanced-unbalanced converter or other elements that convert a single-ended signal into a differential signal.
- a differential signal directly from the clock generator. May be generated.
- the positive and negative two signal lines of the differential input signal line may be connected to either of the two signal lines connecting the clock terminals of the ADC 101-l and ADC 101-m, respectively.
- the reference ADC 104 connected in parallel to the analog input signal line, and the digital correction unit for correcting the output result of the ADC 101-k based on the output comparison result of the ADC 104 and each ADC 101-k. It has.
- the buffer units 401-k and 400 are further connected between the analog input terminals of the respective ADCs 101-k and the reference ADC 104 and the common input signal line. From the comparison value between the output of the ADC 104 and the output of the ADC 101-k, the digital correction unit simultaneously determines the delay error generated between the delay devices 105-k and the characteristic variation such as the input frequency band generated between the input buffers 401-k. The calculated output value of the ADC 101-k is obtained based on the calculation.
- a delay device 700 capable of adjusting the delay amount is provided on the clock signal line input to the reference ADC 104.
- phase relationship between the clocks input to the reference ADC 104 and each ADC 101-k is shown in FIG. clk1 and clk2 indicate arbitrary two types of sampling clock signals having the smallest phase difference among the sampling clock signals input to any of the ADCs 101-k.
- t1 and t2 indicate clock falling times of clk1 and clk2, respectively.
- ref_clk is a sampling clock input to the reference ADC 104, and is a reference clock for correcting the sampling timing of the ADC 101-k.
- Tref indicates the clock fall time of ref_clk, and t1 and t2 have a time difference of approximately 1 / (fs ⁇ N). This interval determines the sampling interval of the interleave ADC.
- the digital correction unit compares the value AD-converted by the ADC 101-k at time t1 with the value AD-converted by the reference ADC 104 at time tref, and calculates the correction value, thereby correcting the influence due to the shift in t1. If the time between tref and t1 is too far, a correct correction value cannot be calculated.
- clk2 since it is necessary to bring t1 closer to tref, it is necessary to increase the delay amount of clk1. However, since t1 and t2 need to be maintained at approximately 1 / (fs ⁇ N), clk2 must also increase the delay amount. In this way, all the clk delay amounts must be increased in a chained manner. However, when the delay amount is generally increased in the delay unit, the delay circuit is increased, which is not desirable in terms of mounting.
- the delay amount adjustment width of the delay device and the delay adjustment accuracy are generally in a trade-off, it is desirable to reduce the adjustment range of the delay amount.
- the ref_clk itself is delayed by the delay device 700 provided on the path of the ref_clk input to the reference ADC 104.
- tref and t2 can be brought closer by delaying tref.
- the phase relationship between the sampling clock signals input to each ADC 101-k is maintained at approximately 1 / (fs ⁇ N), and the falling edge of ref_clk is input to any one ADC 101-k at any timing. If it coincides with the falling edge of the sampling clock signal, the other falling edges of the ADC can be periodically synchronized with any falling edge of ref_clk.
- any falling edge of ref_clk can be periodically synchronized with the falling edge of the clock input of all ADCs 101-k.
- ADC 102 Digital correction unit 103: Multiplexer 104: ADC for reference 105-1 to 105-N: delay units 201-1 to 201-N: gain correction units 202-1 to 202-N: offset correction units 203-1 to 203-N: skew correction units 204-1 to 204-N : Subtracters 205-1 to 205-N: multipliers 206-1 to 206-N: adders 207-1 to 207-N: delay units 208-1 to 208-N: multipliers 209-1 to 209-N : Multipliers 210-1 to 210-N: adders 211-1 to 211-N: delay units 212-1 to 212-N: adders 213-1 to 213-N: multipliers 214-1 to 214-N : Adders 215-1 to 215-N: delay units 216-1 to 216-N: multipliers 217-1 to 217-N: delay units 218-1 to 218-N: adder 300: resistor 301: capacitor 302
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Analogue/Digital Conversion (AREA)
Abstract
Description
本構成では各々のADC101-kからはサンプリングレートfsでAD変換されたアナログ入力信号Vinの値がデジタル信号として出力され、それらは互いに補完関係にある。
102:デジタル補正部
103:マルチプレクサ
104:参照用ADC
105-1~105-N:遅延器
201-1~201-N:利得補正部
202-1~202-N:オフセット補正部
203-1~203-N:スキュー補正部
204-1~204-N:減算器
205-1~205-N:乗算器
206-1~206-N:加算器
207-1~207-N:遅延器
208-1~208-N:乗算器
209-1~209-N:乗算器
210-1~210-N:加算器
211-1~211-N:遅延器
212-1~212-N:加算器
213-1~213-N:乗算器
214-1~214-N:加算器
215-1~215-N:遅延器
216-1~216-N:乗算器
217-1~217-N:遅延器
218-1~218-N:加算器
300:抵抗
301:コンデンサ
302:インダクタンス
303:可変容量コンデンサ
400:バッファ部
401-1~401-N:バッファ部
500:オペアンプ
501:コンデンサ
502:差動出力オペアンプ
700:遅延器
901-1~901-K:遅延器
902-1~902-K:タップ部
903:加算器
Claims (17)
- 複数の単位アナログデジタル変換器によって構成される、タイムインターリーブ型アナログデジタル変換器において、2本で1組の差動クロック信号線の片側の第1のクロック信号線が、前記タイムインターリーブ型アナログデジタル変換器を構成する、第1の単位アナログデジタル変換器の正入力クロック端子と、第1の単位アナログデジタル変換器とひと組になってタイムインターリーブ型アナログデジタル変換器を構成する第2の単位アナログデジタル変換器の負入力クロック端子に入力され、前記差動クロック信号線対のうち、前記第1のクロック信号線とは異なる側のクロック信号線である第2のクロック信号線が、前記第1のアナログデジタル変換器の負入力クロック端子と、前記第2のアナログデジタル変換器の正入力クロック端子に入力されることを特徴とするタイムインターリーブ型アナログデジタル変換器。
- 各単位アナログデジタル変換器に入力される各サンプリングクロックの理想的なサンプリングタイミングからのずれを、各単位アナログデジタル変換器後に設けた各デジタル補正部において各単位アナログデジタル変換器出力の微分値に相当する値と理想的なサンプリングタイミングからのタイミング誤差に相当する値を乗算した値を各単位アナログデジタル変換器出力から差し引くことで補正することを特徴とする請求項1記載のタイムインターリーブ型アナログデジタル変換器。
- 各単位アナログデジタル変換器と共通のアナログ入力電圧をサンプリングする参照用アナログデジタル変換器の変換結果を用いてデジタル補正を行うことを特徴とする請求項2記載のタイムインターリーブ型アナログデジタル変換器。
- 共通アナログ入力信号線の後段に複数のアナログ入力バッファが接続され、前記各入力バッファの後段に前記単位アナログデジタル変換器が接続されることを特徴とする請求項1記載のタイムインターリーブ型アナログデジタル変換器。
- 共通アナログ入力信号線の後段に複数のアナログ入力バッファが接続され、前記各入力バッファの後段に前記単位アナログデジタル変換器が接続されることを特徴とする請求項2記載のタイムインターリーブ型アナログデジタル変換器。
- 共通アナログ入力信号線の後段に複数のアナログ入力バッファが接続され、前記各入力バッファの後段に前記単位アナログデジタル変換器が接続されることを特徴とする請求項3記載のタイムインターリーブ型アナログデジタル変換器。
- 複数の単位アナログデジタル変換器によって構成される、タイムインターリーブ型アナログデジタル変換器において、共通アナログ入力信号線の後段に複数のアナログ入力バッファが接続され、前記入力バッファの後段に前記単位アナログデジタル変換器が接続され、各単位アナログデジタル変換器直後に設けたデジタル補正部において、理想的なバッファ部利得に相当する値を各単位アナログデジタル変換器出力に乗算することによるバッファ利得の補正、または、理想的な直流オフセット電圧値に相当する値を各単位アナログデジタル変換器出力から差し引くことによる直流オフセットの補正、または、各単位アナログデジタル変換器出力の微分値に相当する値と理想的なバッファ周波数帯域からの誤差に相当する値を乗算した値を各単位アナログデジタル変換器出力から差し引くことによるバッファ間の信号帯域補正のいずれかひとつを実施することを特徴としたタイムインターリーブ型アナログデジタル変換器。
- 共通のアナログ入力電圧をサンプリングする参照用アナログデジタル変換器の変換結果を用いて、デジタル補正を行うことを特徴とする請求項7記載のタイムインターリーブ型アナログデジタル変換器。
- サンプリングクロックの理想時間からのずれを、各単位アナログデジタル変換器後に設けたデジタル補正部で補正することを特徴とする請求項8記載のタイムインターリーブ型アナログデジタル変換器。
- 共通アナログ入力信号線の後段に複数のアナログ入力バッファが接続され、前記入力バッファの後段に前記単位アナログデジタル変換器が接続され、各単位アナログデジタル変換器直後に設けたデジタル補正部において、理想的なバッファ部利得に相当する値を各単位アナログデジタル変換器出力に乗算することによるバッファ利得の補正、または、理想的な直流オフセット電圧値に相当する値を各単位アナログデジタル変換器出力から差し引くことによる直流オフセットの補正、または、各単位アナログデジタル変換器出力の微分値に相当する値と理想的なバッファ周波数帯域からの誤差に相当する値を乗算した値を各単位アナログデジタル変換器出力から差し引くことによるバッファ間の信号帯域補正のいずれかひとつを実施することを特徴とする請求項1記載のタイムインターリーブ型アナログデジタル変換器。
- サンプリングクロックの理想時間からのずれを、前記各単位アナログデジタル変換器後に設けたデジタル補正部で補正することを特徴とする請求項10記載のタイムインターリーブ型アナログデジタル変換器。
- 共通のアナログ入力電圧をサンプリングする参照用アナログデジタル変換器の変換結果を用いて、デジタル補正を行うことを特徴とする請求項10記載のタイムインターリーブ型アナログデジタル変換器。
- サンプリングクロックの理想時間からのずれを、前記各単位アナログデジタル変換器後に設けたデジタル補正部で補正することを特徴とする請求項12記載のタイムインターリーブ型アナログデジタル変換器。
- 前記参照用アナログデジタル変換器に入力するサンプリングクロック信号の位相を調整可能な位相遅延回路を持つことを特徴としたタイムインターリーブ型アナログデジタル変換器。
- 請求項13に記載のアナログデジタル変換器において、前記参照用アナログデジタル変換器に入力するサンプリングクロック信号の位相を調整可能な位相遅延回路を持つことを特徴とする請求項13記載のタイムインターリーブ型アナログデジタル変換器。
- サンプリングクロックの位相差をLCフィルタにより作り出すことを特徴とする請求項1記載のタイムインターリーブ型アナログデジタル変換器。
- ナログデジタル変換器に入力されるサンプリングクロックの少なくとも1つの遅延をLCフィルタにより作り出すことを特徴とする請求項15記載のタイムインターリーブ型アナログデジタル変換器。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2012/072837 WO2014038056A1 (ja) | 2012-09-07 | 2012-09-07 | インターリーブa/d変換器 |
US14/421,512 US9337853B2 (en) | 2012-09-07 | 2012-09-07 | Interleaved A/D converter |
JP2014534119A JP5836493B2 (ja) | 2012-09-07 | 2012-09-07 | インターリーブa/d変換器 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2012/072837 WO2014038056A1 (ja) | 2012-09-07 | 2012-09-07 | インターリーブa/d変換器 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2014038056A1 true WO2014038056A1 (ja) | 2014-03-13 |
Family
ID=50236705
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/072837 WO2014038056A1 (ja) | 2012-09-07 | 2012-09-07 | インターリーブa/d変換器 |
Country Status (3)
Country | Link |
---|---|
US (1) | US9337853B2 (ja) |
JP (1) | JP5836493B2 (ja) |
WO (1) | WO2014038056A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017047110A (ja) * | 2015-09-04 | 2017-03-09 | キヤノン株式会社 | 撮像装置 |
CN110445481A (zh) * | 2019-08-06 | 2019-11-12 | 中国科学院近代物理研究所 | 一种采样间隔不均匀性修正电路及方法 |
JP2023152240A (ja) * | 2022-03-31 | 2023-10-16 | 創意電子股▲ふん▼有限公司 | アナログ-デジタル変換装置及びオフセット補正方法 |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6481307B2 (ja) * | 2014-09-24 | 2019-03-13 | 株式会社ソシオネクスト | アナログデジタル変換器、半導体集積回路、及びアナログデジタル変換方法 |
EP3304744A1 (en) * | 2015-05-29 | 2018-04-11 | Telefonaktiebolaget LM Ericsson (publ) | Analog-to-digital converter system |
US9503115B1 (en) * | 2016-02-19 | 2016-11-22 | Xilinx, Inc. | Circuit for and method of implementing a time-interleaved analog-to-digital converter |
US10393861B2 (en) * | 2016-04-05 | 2019-08-27 | Mitsubishi Electric Corporation | Frequency modulation circuit, FM-CW radar, and high-speed modulation radar |
CN111193959B (zh) * | 2018-11-15 | 2022-01-07 | 西安诺瓦星云科技股份有限公司 | 模拟视频信号处理方法和模拟视频处理装置 |
CN113037283A (zh) * | 2019-12-09 | 2021-06-25 | 中兴通讯股份有限公司 | 时间交织逐次逼近型模数转换器及其校准方法 |
CN111277267A (zh) * | 2020-03-05 | 2020-06-12 | 中国人民解放军国防科技大学 | 双通道tiadc的时延失配补偿方法、装置及电子设备 |
CN112067868B (zh) * | 2020-09-07 | 2023-04-21 | 中电科思仪科技股份有限公司 | 一种具有自动校准功能的数字示波器多路adc交叉采样电路及其校准方法 |
JP2022146460A (ja) | 2021-03-22 | 2022-10-05 | キオクシア株式会社 | 半導体回路、受信装置及びメモリシステム |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009159534A (ja) * | 2007-12-27 | 2009-07-16 | Advantest Corp | アナログデジタル変換装置、アナログデジタル変換方法、制御装置及びプログラム |
JP2009272915A (ja) * | 2008-05-08 | 2009-11-19 | Semiconductor Technology Academic Research Center | サンプルホールド回路及びa/d変換装置 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3745962B2 (ja) | 2001-01-24 | 2006-02-15 | 株式会社アドバンテスト | インターリーブad変換方式波形ディジタイザ装置、及び試験装置 |
US7541950B2 (en) * | 2006-07-20 | 2009-06-02 | Samplify Systems, Inc. | Enhanced time-interleaved A/D conversion using compression |
WO2009090514A1 (en) * | 2008-01-14 | 2009-07-23 | Nxp B.V. | Time interleaved analog-to-digital-converter |
WO2010095232A1 (ja) * | 2009-02-19 | 2010-08-26 | 株式会社日立製作所 | アナログデジタル変換器および半導体集積回路装置 |
US8581634B2 (en) * | 2010-02-24 | 2013-11-12 | Texas Instruments Incorporated | Source follower input buffer |
JP5880565B2 (ja) * | 2011-09-30 | 2016-03-09 | 日本電気株式会社 | アナログデジタル変換回路及びアナログデジタル変換方法 |
US8487895B1 (en) * | 2012-06-26 | 2013-07-16 | Google Inc. | Systems and methods for enhancing touch event processing performance |
-
2012
- 2012-09-07 US US14/421,512 patent/US9337853B2/en active Active
- 2012-09-07 WO PCT/JP2012/072837 patent/WO2014038056A1/ja active Application Filing
- 2012-09-07 JP JP2014534119A patent/JP5836493B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009159534A (ja) * | 2007-12-27 | 2009-07-16 | Advantest Corp | アナログデジタル変換装置、アナログデジタル変換方法、制御装置及びプログラム |
JP2009272915A (ja) * | 2008-05-08 | 2009-11-19 | Semiconductor Technology Academic Research Center | サンプルホールド回路及びa/d変換装置 |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2017047110A (ja) * | 2015-09-04 | 2017-03-09 | キヤノン株式会社 | 撮像装置 |
US10537242B2 (en) | 2015-09-04 | 2020-01-21 | Canon Kabushiki Kaisha | Imaging apparatus |
CN110445481A (zh) * | 2019-08-06 | 2019-11-12 | 中国科学院近代物理研究所 | 一种采样间隔不均匀性修正电路及方法 |
JP2023152240A (ja) * | 2022-03-31 | 2023-10-16 | 創意電子股▲ふん▼有限公司 | アナログ-デジタル変換装置及びオフセット補正方法 |
Also Published As
Publication number | Publication date |
---|---|
US20150341044A1 (en) | 2015-11-26 |
JPWO2014038056A1 (ja) | 2016-08-08 |
US9337853B2 (en) | 2016-05-10 |
JP5836493B2 (ja) | 2015-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5836493B2 (ja) | インターリーブa/d変換器 | |
JP5095007B2 (ja) | アナログデジタル変換器および半導体集積回路装置 | |
US9608652B2 (en) | Methods and apparatus for reducing timing-skew errors in time-interleaved analog-to-digital converters | |
Matsuno et al. | All-digital background calibration technique for time-interleaved ADC using pseudo aliasing signal | |
CN106685424B (zh) | 用于模数转换器的微处理器辅助校准 | |
TWI539756B (zh) | 類比至數位轉換器裝置、用於校準類比至數位轉換器的方法、及用於儲存執行該方法之指令的電腦可讀取媒體 | |
US8890739B2 (en) | Time interleaving analog-to-digital converter | |
US7301486B2 (en) | Time-interleaved analog-to-digital converter having timing calibration | |
US7629905B2 (en) | Analog correction of a phase-mismatch in high-sample rate time-interleaved analog-to-digital converters | |
JP5288003B2 (ja) | A/d変換装置とその補正制御方法 | |
US8564462B2 (en) | Digital correction techniques for data converters | |
JP5142342B2 (ja) | Ad変換回路 | |
JP2016213826A (ja) | タイムインターリーブ型ad変換器 | |
US20140152478A1 (en) | Randomized time-interleaved sample-and-hold system | |
JP2008011189A (ja) | タイム・インターリーブa/d変換装置 | |
US7808407B2 (en) | Sub-channel distortion mitigation in parallel digital systems | |
EP2041873A1 (en) | Time- interleaved analog-to-digital converter system | |
JP5286420B2 (ja) | アナログデジタル変換器およびそれを用いた半導体集積回路装置 | |
KR101691367B1 (ko) | M채널 TI-ADCs에서 미스매치에 대한 디지털 후면 교정 방법 및 그 장치 | |
Nam et al. | A 12.8-Gbaud ADC-based NRZ/PAM4 receiver with embedded tunable IIR equalization filter achieving 2.43-pJ/b in 65nm CMOS | |
Luo et al. | Coordinated blind calibration for time interleaved ADCS | |
Huiqing et al. | Adaptive digital calibration of timing mismatch for TIADCs using correlation | |
EP2503696B1 (en) | Time-interleaved analog-to-digital converter system | |
Pillai et al. | Efficient signal reconstruction scheme for M-channel time-interleaved ADCs | |
US20240007120A1 (en) | Time-Interleaved ADC |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12884043 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014534119 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 14421512 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12884043 Country of ref document: EP Kind code of ref document: A1 |