CN104980153B - Analog-digital Converter module, transfer function compensation method and device - Google Patents
Analog-digital Converter module, transfer function compensation method and device Download PDFInfo
- Publication number
- CN104980153B CN104980153B CN201410128551.6A CN201410128551A CN104980153B CN 104980153 B CN104980153 B CN 104980153B CN 201410128551 A CN201410128551 A CN 201410128551A CN 104980153 B CN104980153 B CN 104980153B
- Authority
- CN
- China
- Prior art keywords
- function
- signal
- analog
- digital converter
- output signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The embodiment of the present invention provides a kind of Analog-digital Converter module, transfer function compensation device and method, and the transfer function compensation method to nonlinear effect existing for Analog-digital Converter module compensating.First, multiple penalty functions corresponding to multiple input ranges are provided, wherein described penalty function is real output signal caused by the input signal as the corresponding input range to be compensated, to reduce the signal difference value between the real output signal and desired output signal.Then, the input range of the input signal is determined.Then, using the penalty function corresponding to the input range of decision to as actual transmissions function be based on the input signal caused by the real output signal compensate.By above-mentioned transfer function compensation method, the real output signal of Analog-digital Converter module can be approximately or equal to desired output signal.
Description
Technical field
The present invention relates to a kind of Analog-digital Converter (analog-digital conversion) module, and particularly use
In transfer function (transfer function) compensation method of Analog-digital Converter module and device.
Background technology
Analog-digital Converter module is a kind of primary element being common in electronic equipment, and has transfer function, with logical
Transfer function is crossed by analog signal conversion digital signal.In general, the actual transmissions function of Analog-digital Converter module can be because
For amplifier nonlinearity gain with ideal transfer functions difference.
Refer to Fig. 1, Fig. 1 be conventional tube formula (pipelined) Analog-digital Converter module actual transmissions function with
The schematic diagram of ideal transfer functions.In Fig. 1, trunnion axis is to represent the unit strength of input signal, and vertical axis is to table
Show the unit strength of corresponding output signal.Curve C11 is representing the actual transmissions of conventional tube formula Analog-digital Converter module
Function, and curve C12 is representing the ideal transfer functions of conventional tube formula Analog-digital Converter module.
By curve C12 it is known that when the unit strength of input signal falls in -1 to 0.4 region and 0.4 to 1 area
Between when, ideal transfer functions can cause the unit strength of output signal to be equal to the unit strength of input signal, but pass through curve C11
It can be found that actual transmissions function can't cause the unit strength of output signal to be equal to the unit strength of input signal.Change speech
It, conventional tube formula Analog-digital Converter module is there is nonlinear effect, but because gain is linear, so non-linear effect
Should and it can not be corrected separately through gain compensation.
The content of the invention
The embodiment of the present invention provides a kind of compensating to nonlinear effect existing for Analog-digital Converter module
Transfer function compensation method.First, there is provided corresponding to multiple penalty functions of multiple input ranges, wherein the simulation numeral turns
Input signal is converted to real output signal by mold changing block using actual transmissions function, and the real output signal exports with preferable
There is signal difference value, the penalty function is to by by the input signal institute of the corresponding input range between signal
The caused real output signal compensates, and to reduce the signal difference value, and the desired output signal is by institute
The ideal transfer functions for stating input/output module are based on the input signal and produced.Then, the institute of the input signal is determined
State input range.The person of connecing, using the penalty function corresponding to the input range of decision to by the actual transmissions function
Compensated based on the real output signal caused by the input signal.
The embodiment of the present invention provides a kind of compensating to nonlinear effect existing for Analog-digital Converter module
Transfer function compensation device, and the transfer function compensation device includes multiple compensating units, input range detector and multiplexing
Device, wherein the multiplexer is coupled to the multiple compensating unit and the input range detector.The multiple compensating unit
There is penalty function respectively, wherein input signal is converted to reality by the Analog-digital Converter module using actual transmissions function
Output signal, has signal difference value between the real output signal and desired output signal, the compensating unit uses benefit
Function pair real output signal as caused by the input signal of the corresponding input range is repaid to compensate, with
The signal difference value is reduced, and the desired output signal is to be based on institute by the ideal transfer functions of the input/output module
State input signal and produce.The input range detector is to determine the input range of the input signal, to produce
Selection signal.The institute that the multiplexer is compensated according to one of the multiple compensating unit of selection signal selection output
State real output signal.
The embodiment of the present invention provides a kind of Analog-digital Converter module, and word modules are stated in the simulation includes Analog-digital Converter
Device and transfer function compensation device, wherein transfer function compensation device are coupled to analog-digital converter.Analog-digital converter
Input signal is converted into real output signal using actual transmissions function, wherein the real output signal is believed with preferable output
There is signal difference value, and the desired output signal is the ideal transfer functions base by the analog-digital converter between number
Produced in the input signal.Transfer function compensation device include multiple compensating units, input range detector, multiplexer with
Drill circuit, wherein the multiplexer is coupled to the multiple compensating unit and the input range detector, and the training
Circuit is coupled to the multiplexer.The multiple compensating unit has penalty function respectively, wherein described in compensating unit use
Penalty function compensates to the real output signal caused by the input signal as the corresponding input range,
To reduce the signal difference value.The input range detector to determine the input range of the input signal, with
Produce selection signal.The multiplexer exports one of the multiple compensating unit according to selection signal selection and compensated
The real output signal.It is described to according to described caused by the input signal based on the corresponding input range
Signal difference value performs training program and obtains multiple coefficients of the penalty function.
In summary, compared to conventional transmission function compensation method, the transfer function compensation that the embodiment of the present invention is provided
The design of the circuit or algorithm of method and apparatus is more simple, and uncomplicated, and has preferable compensation effect.Therefore, use
The conversion accuracy of the Analog-digital Converter module of above-mentioned transfer function compensation method and device is higher, and with relatively low
Cost and circuit complexity.
For the enabled feature and technology contents for being further understood that the present invention, refer to below in connection with the present invention specifically
Bright and accompanying drawing, but this explanation is only used for illustrating the present invention with accompanying drawing, not scope of the presently claimed invention is made any
Limitation.
Brief description of the drawings
Fig. 1 is the actual transmissions function of conventional tube formula Analog-digital Converter module and the schematic diagram of ideal transfer functions.
Fig. 2A is the block diagram of the Analog-digital Converter module of the embodiment of the present invention.
Fig. 2 B are the block diagrams of the Analog-digital Converter module of another embodiment of the present invention.
Fig. 2 C are the block diagrams of the Analog-digital Converter module of another embodiment of the present invention.
Fig. 3 is the block diagram of the pipeline type analog-digital converter of the embodiment of the present invention.
Fig. 4 is that the compensation of the ideal transfer functions and actual transmissions function of the analog-digital converter of the embodiment of the present invention is shown
It is intended to.
Fig. 5 is the flow chart for the transfer function compensation method that the embodiment of the present invention is provided.
Fig. 6 is the block diagram of the transfer function compensation device of the embodiment of the present invention.
Fig. 7 is the ideal transfer functions of the Analog-digital Converter module of the embodiment of the present invention, actual transmissions function and compensation
The schematic diagram of actual transmissions function afterwards.
【Symbol description】
21~23：Analog-digital Converter module
211、221、231：With reference to analog-digital converter
212、222、223、232、233：Analog-digital converter
213、224、225、235、6：Transfer function compensation device
226：Parallel/serial converter
234：Multiplexer
3：Pipeline type analog-digital converter
31~33：Analog digital converting unit
61：Input range detector
62：Drill circuit
631~63n：Activate grid
641~64n：Compensating unit
65：Multiplexer
Coefficient vector
C11, C12, C41, C42, C71~C73：Curve
D1~D3：Quantized value
Res1~res3：Residual voltage
R1~R3：Input range
Selection signal
S1~Sn：Switching signal
S51~S53：Steps flow chart
x0~x3：The intensity level of input signal
Y (x)=f1 (y ' (x)), y (x)=f2(y ' (x)), y (x)=f3(y’(x))：Real output signal after compensation
Embodiment
Various exemplary embodiments will be more fully described referring to alterations below, shown in alterations
Exemplary embodiments.However, concept of the present invention may embody in many different forms, and it should not be construed as limited by institute herein
The exemplary embodiments of elaboration.Specifically, there is provided these exemplary embodiments make it that the present invention will be detailed and complete, and will
Idea of the invention category is fully passed on to those skilled in the art.In all schemas, the big of Ceng Ji areas is lavished praise on oneself in order to clear
Small and relative size.Similar numeral indicates similar component all the time.It should be understood that term "or" used herein should regard actual feelings
What condition may include being associated lists any one of project or all combinations of more persons.
The embodiment of the present invention provides a kind of for improving Analog-digital Converter module (for example, the simulation numeral of pipeline type turns
Change the mold block, but be not limited) transfer function there is transfer function compensation method and the device of nonlinear characteristic, wherein described
Transfer function compensation method can be implemented in Analog-digital Converter module, or, transfer function compensation device is arranged at simulation
Among data-converting block.
The transfer function compensation method defines multiple input ranges of input signal with device, and according to the multiple defeated
Enter scope and the actual transmissions function corresponding to real output signal is divided into some actual transmissions subfunctions.Then, obtain every
The penalty function of the multiple actual transmissions subfunction of one correspondence, wherein the multiple penalty function is to for each corresponding
Actual transmissions subfunction compensate calculating so that whole actual transmissions function levels off to an ideal transfer functions；That is, pass through
The signal difference value between real output signal and desired output signal after the compensation of overcompensation function (takes absolute value, not considered
It is positive and negative) predetermined threshold value can be less than (for example, 10-6Unit strength).
In other words, the transfer function compensation method provides multiple compensation letters corresponding to multiple input ranges with device
Number, and because nonlinear effect has signal difference value between real output signal and desired output signal.Penalty function to
Real output signal caused by input signal as corresponding input range is compensated, to reduce the signal difference
Value.
Then, the transfer function compensation method and device detection input signal, depending on input signal input range.So
Afterwards, the transfer function compensation method uses corresponding penalty function to actual transmissions letter with device according to the input range determined
Base compensates in real output signal caused by input signal.
In other embodiments of the invention, each of the multiple penalty function can be selectively linear compensation
Function or nonlinear compensation function (for example, the nonlinear compensation function of three ranks, but the present invention does not limit nonlinear compensation function
Type), namely whole penalty functions can be linear compensation function or nonlinear compensation function, or, the compensation letter of part
Number is linear compensation function, and partial penalty function is nonlinear compensation function.Furthermore multiple coefficients of penalty function pass through
Training program is based on the signal difference value and produced, and the training program can be minimum side's value (least mean
Square, LMS) or minimize mean square deviation (minimum mean square error, MMSE) training program, but the present invention
It is not intended to limit the type of training program.Person of an ordinary skill in the technical field should be understood that above-mentioned training program can be based on it
He seeks the algorithm of optimization or suboptimization to realize.
Fig. 2A is refer to, Fig. 2A is the block diagram of the Analog-digital Converter module of the embodiment of the present invention.Analog-digital Converter
Module 21 is included with reference to analog-digital converter 211, analog-digital converter 212 and transfer function compensation device 213.Transmit letter
Number compensation device 213 is coupled to reference to analog-digital converter 211.
With reference to analog-digital converter 211 and analog-digital converter 212, to receive input signal, (it believes for simulation
Number), input signal is converted to real output signal by analog-digital converter 212 according to its actual transfer function, with reference to simulation
Digital quantizer 211, which can be one, has higher bit (bit) number and the analog-digital converter of pinpoint accuracy, its actual transmissions
It can be equal to the ideal transfer functions of analog-digital converter 212 on function is equivalent, therefore input signal can be converted to two
The desired output signal of analog-digital converter 212.Furthermore, when analog-digital converter 212 operates in high-speed
When, the output signal of analog-digital converter 212 easily produces drift so as to produce nonlinear error, therefore for producing with reference to letter
Although number reference analog-digital converter 211 can be high accuracy, do not increasing cost and using same performance simulation
On the premise of data-converting block, with reference to analog-digital converter 211 operation rate than other analog-digital converter 212
Speed it is low, in this way, the output signal with reference to analog-digital converter 211 just has more reference value.
Transfer function compensation device 213 provide corresponding to multiple input ranges multiple penalty functions, penalty function to
Real output signal caused by input signal as corresponding input range is compensated, with reduce real output signal with
Signal difference value between desired output signal.Transfer function compensation device 213 can detect input signal, to determine input signal
Input range, namely determine that the intensity of input signal is located at the whichever of above-mentioned multiple input ranges.Then, transfer function compensates
Device 213 is using penalty function corresponding to the input range determined to real caused by being based on input signal as actual transmissions function
Border output signal compensates.
It should be appreciated that the quantity of above-mentioned multiple input ranges can be adjusted according to actual conditions, and it is general next
Say, the quantity of multiple input ranges is bigger, namely the section of each input range is smaller, then transfer function compensation device 213
The effect for compensating the nonlinear effect of Analog-digital Converter module 21 is better.
Still further aspect, penalty function can represent f corresponding to i-th of input rangei(), the then reality after compensating are defeated
F can be expressed as by going out signali(y ' (x)), wherein y ' (x) represent real output signal, and x represents input signal.In of the invention real
To apply in example, the type of penalty function has no any limitation, namely it can be linear compensation function or nonlinear compensation function,
Such as can be fi(y ' (x))=gi×y’(x)+oiOr fi(y ' (x))=ai+bi×y’(x)+ci×y’2(x)+di×y’3
(x), wherein coefficient gi、oiOr coefficient ai、bi、ci、diIt is to be obtained by training program.
Furthermore, coefficient g, o or coefficient a, b, c, d have initial value.In training program, transfer function is mended
Device 213 is repaid according to the real output signal after current compensation and the simulation number with reference to the equivalent upper offer of analog-digital converter 21
The desired output signal of word converter 212 is repeatedly updated to coefficient g, o or coefficient a, b, c, d numerical value, to mend
Real output signal after repaying can be close to desired output signal.Training program can be by asking optimization or seeking the algorithm of suboptimization
To realize, such as training program may, for example, be based on minimum side's value or minimize the training program of mean square deviation.
Fig. 2 B are refer to, Fig. 2 B are the block diagrams of the Analog-digital Converter module of another embodiment of the present invention.Compared to figure
2A Analog-digital Converter module 21, Analog-digital Converter module 22 extraly have more one group of analog-digital converter 223 with passing
Defeated function compensation device 225, and due to having had more the pass of one group of analog-digital converter 223 and transfer function compensation device 225
System, must design parallel/serial converter (a parallel to serial converter) 226 in Analog-digital Converter mould
In block 22.In addition, transfer function compensation device 225 is coupled to reference to analog-digital converter 221 and analog-digital converter
223.Parallel/serial converter 226 is coupled to transfer function compensation device 224,225.
With reference to analog-digital converter 221 function be same as Fig. 2A reference analog-digital converter 211 function, mould
The function of plan digital quantizer 222,223 is same as the function of Fig. 2A analog-digital converter 212, and transfer function compensation fills
Put 224,225 function be same as Fig. 2A transfer function compensation device 213 function.Briefly, parallel/serial converter
226 carry out parallel/serial place will come from the real output signal after transfer function compensation device 224,225 compensates
Reason.Briefly, transfer function compensation device 224,225 uses penalty function corresponding to the input range determined respectively to by mould
The actual transmissions function for intending digital quantizer 222,223 is compensated based on real output signal caused by input signal.
It should also be noted that by Fig. 2A and Fig. 2 B embodiment it is known that the present invention is not intended to limit Analog-digital Converter mould
Analog-digital converter possessed by block and the group number of transfer function compensation device.In other words, Analog-digital Converter module can be with
The analog-digital converter and the group number of transfer function compensation device expanded according to actual use situation.
Fig. 2 C are refer to, Fig. 2 C are the block diagrams of the Analog-digital Converter module of another embodiment of the present invention.Compared to figure
2B Analog-digital Converter module 22, Analog-digital Converter module 23 have lacked a transfer function compensation device, but same
With two analog-digital converters 232,233, and a multiplexer 234 is extraly devised, so that analog-digital converter
232nd, 233 transfer function compensation device 235 is coupled by multiplexer 234.
With reference to analog-digital converter 231 function be same as Fig. 2A reference analog-digital converter 211 function, mould
The function of plan digital quantizer 232,233 is same as the function of Fig. 2A analog-digital converter 212, and transfer function compensation fills
Put 235 function be same as Fig. 2A transfer function compensation device 213 function.Multiplexer 234 can receive a control letter
Number, to allow one of them the caused real output signal of analog-digital converter 232,233 to be transmitted function compensation device 235
Compensated.In other words, splitter 235 can be used in conjunction with analog-digital converter 232,233 by way of dividing the work and being multiplexed
Same transfer function compensation device 235, and the cost of Analog-digital Converter module 23 can be further reduced.
It should also be noted that the present invention is not intended to limit analog-digital converter possessed by Analog-digital Converter module and transmission
The quantity of function compensation device, the quantity of multiplexer is not limited yet.In other words, Analog-digital Converter module can be according to actually making
Expand the quantity of analog-digital converter with situation, and accordingly can make at least a portion mould using at least one multiplexer
Intend digital quantizer and share at least one transfer function compensation device.For example, Analog-digital Converter module can have four moulds
Intend digital quantizer, and by a multiplexer share a transfer function compensation device, or can by two multiplexers,
So that two of which Analog-digital Converter module shares a transfer function compensation device and turned with other two of which simulation numeral
Mold changing block shares two Analog-digital Converter modules of a transfer function compensation device and shares another transfer function compensation device.
Then, Fig. 3 is refer to, Fig. 3 is the block diagram of the pipeline type analog-digital converter of the embodiment of the present invention.In above-mentioned
In Fig. 2A~Fig. 2 C embodiment, analog-digital converter 212,222,223,232 and 233 can be Fig. 3 pipeline type simulation
Digital quantizer 3.Pipeline type analog-digital converter 3 includes multiple analog digital converting units 31~33 being one another in series.Mould
Intend digital conversion unit 31 and receive input signal, and output quantization value D1 (for example, quantized value of 2 bits), and by residual voltage
Res1 sends next analog digital converting unit 32 to.The reception residual voltage of analog digital converting unit 32, output quantization value D2,
And send residual voltage res2 to next analog digital converting units 33.The effect of analog digital converting unit 33 can basis
Analog digital converting unit 31 is learnt with 32, therefore is not repeated.It can learn what pipeline type analog-digital converter 3 was exported from above
Output signal is multiple quantized value D1~D3 combination.
Although it should also be noted that above-described embodiment with Analog-digital Converter module 2 for two-way pipeline type Analog-digital Converter
Module is illustrated, but the Analog-digital Converter module of the present invention is not limited with two-way pipeline type Analog-digital Converter module.
For example, Analog-digital Converter module can be single channel pipeline type Analog-digital Converter module or other kinds of simulation number
Word modular converter.In other words, above-mentioned function compensation device can be arranged at other not by person of ordinary skill in the field
Among congener Analog-digital Converter module, to compensate the nonlinear effect of Analog-digital Converter module in itself.
Fig. 4 is refer to, Fig. 4 is the ideal transfer functions and actual transmissions letter of the analog-digital converter of the embodiment of the present invention
Several compensation schematic diagrames.In Fig. 4, trunnion axis is to represent the unit strength of input signal, and vertical axis is representing corresponding
The unit strength of output signal.Curve C41 to represent the ideal transfer functions of analog-digital converter, and curve C42 to
Represent the actual transmissions function of analog-digital converter.
By curve C41 and C42 it is known that when input signal x intensity is located in input range R1 section (that is,
The intensity of input signal is more than or equal to x0And it is less than x1), then pass through penalty function f1(), the real output signal f after compensation1
(y ' (x)) is by convergence desired output signal y (x).When input signal x intensity is located in input range R2 section (that is, defeated
The intensity for entering signal is more than or equal to x1And it is less than x2), then pass through penalty function f2 (), the real output signal f after compensation2(y’
(x)) by convergence desired output signal y (x).When input signal x intensity is located in input range R3 section (that is, input
The intensity of signal is more than or equal to x2And it is less than x3), then pass through penalty function f3(), the real output signal f after compensation3(y’
(x)) by convergence desired output signal y (x).
Then, Fig. 5 is refer to, Fig. 5 is the flow chart for the transfer function compensation method that the embodiment of the present invention is provided.It is above-mentioned
Transfer function compensation method can be used for compensating the nonlinear effect of Analog-digital Converter module, and can be implemented in simulation numeral
In module, such as the transfer function compensation device in above-mentioned simulation numeral module.Person of ordinary skill in the field Ying Keli
Solution, transfer function compensation method can be realized by way of hardware, firmware or software.
First, in step S51, there is provided corresponding to multiple penalty functions of multiple input ranges of input signal, wherein defeated
Entering multiple input ranges of signal can determine according to actual conditions.It should also be noted that in general, multiple input ranges
Quantity is more, then better for the compensation effect of nonlinear effect.Still further aspect, the spacing of each input range can be with one
Mould is the same, can also be not all the same.Person of ordinary skill in the field it would be appreciated that, the definition of multiple input ranges also may be used
To be defined according to the difference condition between actual transmissions function and ideal transfer functions.It is sum it up, of the invention and unlimited
Make the definition mode of multiple input ranges.
Then, in step S52, input signal is detected, and determine the input range of input signal.Then, in step S53
In, reality output caused by input signal is based on to actual transmissions function using penalty function corresponding to the input range of decision
Signal compensates, so that real output signal is approximately or equal to desired output signal, so as to which nonlinear effect be reduced or avoided
Influence.
Then, for Fig. 2A~2C transfer function compensation device 213,214,215 and 235 implementation further
Illustrate.Fig. 6 is refer to, Fig. 6 is the block diagram of the transfer function compensation device of the embodiment of the present invention.Transfer function compensation dress
Put 6 include input range detector 61, drill circuit 62, activation (enable) grid 631~63n, 641~65n of compensating unit with
Multiplexer 65.Input range detector 61, which is coupled to, activates grid (gate) 631~63n and multiplexer 65, and compensating unit 641~
631~63n of 64n coupling activation grid, drill circuit 62 and multiplexer 65.
Input range detector 61 to determine the input range of input signal, namely determines defeated to detect input signal
The intensity for entering signal is located at the whichever of multiple input ranges.Input range detector 61 produces according to the input range of decision to be included
Multiple switching signal S1~SnSelection signalPlurality of switching signal S1~SnIt is respectively delivered to
Activate 631~63n of grid, and selection signalIt is transmitted to multiplexer 65.
631~63n of grid is activated respectively according to switching signal S1~SnIt is activated, and the real output signal for allowing it to receive
By wherein switching signal S1~SnOne of 631~63n of activation grid can be only allowed to be activated every time.Drill circuit 62 receives
Desired output signal and real output signal, and based on ask optimization or seek the algorithm of suboptimization according to desired output signal and reality
The coefficient vector of multiple coefficients of compensating unit 64i penalty functions corresponding to output signal generationWherein i be 1 to n it is whole
Number.
641~64n of compensating unit is according to the coefficient vector receivedIts corresponding penalty function is produced, wherein mending
It can be, for example, f to repay functioni(y ' (x))=gi×y’(x)+oiOr fi(y ' (x))=ai+bi×y’(x)+ci×y’2(x)+di×
y’3(x), coefficient vectorCoefficient g can be includedi、oiOr coefficient ai、bi、ci、di.641~64n of compensating unit is mended using it
Repay function pair its receive real output signal compensate, with produce compensation after real output signal.Then, multiplexer 65
According to selection signalReality output letter after multiple compensation that selection 641~64n of compensating unit is exported
Number one of which.
It should also be noted that in the embodiment of the present invention, the using of 631~63n of activation grid can cause compensating unit 641~
64n (n-1) person is without extra computing, to avoid extra power consumption from losing.Sum it up, the transfer function compensation of the present invention
Device is not using above-described embodiment as limitation, and other can realize the circuit or module group of above-mentioned transfer function compensation method such as
Close, can all be considered as the change of above-mentioned transfer function compensation device.
Then, Fig. 7 is refer to, Fig. 7 is the ideal transfer functions of the Analog-digital Converter module of the embodiment of the present invention, reality
The schematic diagram of transfer function and the actual transmissions function after compensation.In Fig. 7, trunnion axis is strong to represent the unit of input signal
Degree, and vertical axis is representing the unit strength of corresponding output signal.In addition, in this embodiment, mould data-converting block is
Pipeline type Analog-digital Converter block.To represent the ideal transfer functions of Analog-digital Converter module, curve C72 is used curve C71
To represent the actual transmissions function of Analog-digital Converter module, and curve C73 is representing that Analog-digital Converter module is compensated
Actual transmissions function afterwards, that is, input signal and its it is compensated after the transfer function that is formed of real output signal.
By Fig. 7 it is known that by the way that the transfer function compensation method of the embodiment of the present invention and device are applied into simulation number
In word modular converter, the nonlinear effect of Analog-digital Converter module can be compensated effectively, and cause the reality after compensation
Border output signal is equivalent or is similar to desired output signal.
In summary described, the transfer function compensation method that the embodiment of the present invention is provided is with device for numeric field
Real output signal compensates, and not the input signal of analog domain is compensated as the compensation method of conventional transmission function,
Therefore the design of its circuit or algorithm is more simple, and uncomplicated.In addition, the transfer function compensation method and device are foundations
Real output signal corresponding to the input range selection penalty function of input signal compensates, therefore its compensation effect is preferable,
Namely the real output signal after compensation is very close to or equal to desired output signal.In addition, the penalty function is not single
It is only third-order non-linear penalty function purely, it can be designed according to actual conditions as linear compensation function or other are multistage
Nonlinear compensation function, therefore its compensation effect is preferable.Therefore, using above-mentioned transfer function compensation method and the simulation number of device
The conversion accuracy of word modular converter is higher, and has relatively low cost and circuit complexity.
Described above, only currently preferred specific embodiment, the feature of the right present invention is not limited thereto, Ren Heben
The those of ordinary skill in field in the field of the invention, can think easily and change or modification, can cover following hair
Bright the scope of the claims.
Claims (14)
1. a kind of transfer function compensation method, to be compensated to nonlinear effect existing for Analog-digital Converter module, its
It is characterised by, the transfer function compensation method includes：
Multiple penalty functions corresponding to multiple input ranges are provided, wherein the Analog-digital Converter module uses actual transmissions
Input signal is converted to real output signal by function, has signal difference between the real output signal and desired output signal
Different value, the penalty function is to by the reality output caused by the input signal as the corresponding input range
Signal compensates, and to reduce the signal difference value, and the desired output signal is by the reason of the input/output module
Think that transfer function is based on the input signal and produced；
Determine the input range of the input signal；And
Using the penalty function corresponding to the input range of decision to being based on the input by the actual transmissions function
The real output signal compensates caused by signal.
2. transfer function compensation method according to claim 1, it is characterised in that the multiple penalty function it is all
Multiple linear compensation functions or multiple nonlinear compensation functions, or, the part of the multiple penalty function is the multiple line
Property penalty function, and the part of the multiple penalty function is the multiple nonlinear compensation function.
3. transfer function compensation method according to claim 2, it is characterised in that the nonlinear compensation function is three ranks
Nonlinear compensation function.
4. transfer function compensation method according to claim 1, it is characterised in that multiple coefficients of the penalty function are
The signal difference value performs training program and produced caused by the input signal based on the corresponding input range.
5. transfer function compensation method according to claim 4, it is characterised in that the training program is based on minimum side
Value or the training program for minimizing mean square deviation.
6. transfer function compensation method according to claim 5, it is characterised in that the Analog-digital Converter module is pipe
The Analog-digital Converter module of road formula.
7. a kind of transfer function compensation device, to be compensated to nonlinear effect existing for Analog-digital Converter module, its
It is characterised by, the transfer function compensation device includes：
Multiple compensating units, respectively with penalty function, wherein the Analog-digital Converter module will using actual transmissions function
Input signal is converted to real output signal, has signal difference value between the real output signal and desired output signal,
The compensating unit is using the penalty function to the reality caused by the input signal as corresponding input range
Output signal compensates, and to reduce the signal difference value, and the desired output signal is the reason by input/output module
Think that transfer function is based on the input signal and produced；
Input range detector, to determine the input range of the input signal, to produce selection signal；And
Multiplexer, the multiple compensating unit and the input range detector are coupled to, are selected according to the selection signal defeated
Go out the real output signal that one of the multiple compensating unit is compensated.
8. transfer function compensation device according to claim 7, it is characterised in that the transfer function compensation device is also wrapped
Include：
Multiple activation grid, the multiple compensating unit and the input range detector are coupled to, wherein the multiple activation grid
Multiple switching signals of the selection signal are controlled by, and the real output signal is inputted to described more
One of individual compensating unit.
9. transfer function compensation device according to claim 7, it is characterised in that the multiple penalty function it is all
Multiple linear compensation functions or multiple nonlinear compensation functions, or, the part of the multiple penalty function is or the multiple
Linear compensation function, and the part of the multiple penalty function is the multiple nonlinear compensation function.
10. transfer function compensation device according to claim 9, it is characterised in that the nonlinear compensation function is three
The nonlinear compensation function of rank.
11. transfer function compensation device according to claim 7, it is characterised in that the transfer function compensation device is also
Including：
Drill circuit, the multiplexer is coupled to, to according to the input signal production based on the corresponding input range
The raw signal difference value performs training program and obtains multiple coefficients of the penalty function.
12. transfer function compensation device according to claim 11, it is characterised in that the training program is based on minimum
Variance or the training program for minimizing mean square deviation.
13. transfer function compensation device according to claim 7, it is characterised in that the Analog-digital Converter module is
The Analog-digital Converter module of pipeline type.
14. a kind of Analog-digital Converter module, it is characterised in that the Analog-digital Converter module includes：
Analog-digital converter, input signal is converted into real output signal using actual transmissions function, wherein the reality
There is signal difference value, and the desired output signal is turned by the simulation numeral between output signal and desired output signal
The ideal transfer functions of parallel operation are based on the input signal and produced；And
Transfer function compensation device, couples analog-digital converter, and the transfer function compensation device includes：
Multiple compensating units, respectively with penalty function, wherein the compensating unit with the penalty function to by corresponding defeated
Enter the real output signal caused by the input signal of scope to compensate, to reduce the signal difference value；
Input range detector, to determine the input range of the input signal, to produce selection signal；
Multiplexer, the multiple compensating unit and the input range detector are coupled to, are selected according to the selection signal defeated
Go out the real output signal that one of the multiple compensating unit is compensated；And
Drill circuit, the multiplexer is coupled to, to according to the input signal production based on the corresponding input range
The raw signal difference value performs training program and obtains multiple coefficients of the penalty function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410128551.6A CN104980153B (en) | 2014-04-01 | 2014-04-01 | Analog-digital Converter module, transfer function compensation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410128551.6A CN104980153B (en) | 2014-04-01 | 2014-04-01 | Analog-digital Converter module, transfer function compensation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104980153A CN104980153A (en) | 2015-10-14 |
| CN104980153B true CN104980153B (en) | 2018-02-16 |
Family
ID=54276312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410128551.6A Active CN104980153B (en) | 2014-04-01 | 2014-04-01 | Analog-digital Converter module, transfer function compensation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104980153B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101729067A (en) * | 2008-10-22 | 2010-06-09 | 财团法人工业技术研究院 | Calibration device and method thereof for pipelined analog-to-digital converter |
| US7952501B2 (en) * | 2007-05-11 | 2011-05-31 | Sung Wan Kim | Demodulator capable of compensating offset voltage of RF signal and method thereof |
| CN102386917A (en) * | 2011-08-25 | 2012-03-21 | 北京昆腾微电子有限公司 | Analogue digital conversion method, analogue digital conversion device and digital microphone |
| CN102594348A (en) * | 2011-01-10 | 2012-07-18 | 英特尔移动通信有限公司 | Calibration circuit and method for calibrating capacitive compensation in digital-to-analog converters |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4827627B2 (en) * | 2006-06-16 | 2011-11-30 | キヤノン株式会社 | Imaging apparatus and processing method thereof |
-
2014
- 2014-04-01 CN CN201410128551.6A patent/CN104980153B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7952501B2 (en) * | 2007-05-11 | 2011-05-31 | Sung Wan Kim | Demodulator capable of compensating offset voltage of RF signal and method thereof |
| CN101729067A (en) * | 2008-10-22 | 2010-06-09 | 财团法人工业技术研究院 | Calibration device and method thereof for pipelined analog-to-digital converter |
| CN102594348A (en) * | 2011-01-10 | 2012-07-18 | 英特尔移动通信有限公司 | Calibration circuit and method for calibrating capacitive compensation in digital-to-analog converters |
| CN102386917A (en) * | 2011-08-25 | 2012-03-21 | 北京昆腾微电子有限公司 | Analogue digital conversion method, analogue digital conversion device and digital microphone |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104980153A (en) | 2015-10-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Synchronization of stochastic hybrid coupled systems with multi-weights and mixed delays via aperiodically adaptive intermittent control | |
| CN106341130B (en) | Digital analog converter | |
| Wang et al. | Cluster synchronization in nonlinearly coupled delayed networks of non-identical dynamic systems | |
| CN103900610B (en) | MEMS gyro random error Forecasting Methodology based on Lycoperdon polymorphum Vitt wavelet neural network | |
| Mason et al. | On linear copositive Lyapunov functions and the stability of switched positive linear systems | |
| Yu et al. | Neural-network-based output-feedback adaptive dynamic surface control for a class of stochastic nonlinear time-delay systems with unknown control directions | |
| CN106330194B (en) | A kind of D/A converting circuit, display panel and display device | |
| CN106788442A (en) | D/A converting circuit and method, source electrode driver and display device | |
| CN103888391B (en) | Signal blind detection method based on double Sigmoid chaotic neural network | |
| CN107765549A (en) | A kind of New Iterative learning control method of batch industrial process | |
| CN106998206A (en) | Charge scaling Continuous Approximation formula analog-digital converter and its control method | |
| CN104135289B (en) | The method and device of many reference voltage monocline ADC of calibration row level | |
| CN104980153B (en) | Analog-digital Converter module, transfer function compensation method and device | |
| Bu | Actor-critic reinforcement learning control of non-strict feedback nonaffine dynamic systems | |
| CN103973306B (en) | A kind of backstage digital calibration device and method of pipeline ADC | |
| CN110175627A (en) | A kind of complicated classification method of knowledge based recombination | |
| Parandin | Numerical solution of fuzzy differential equations of 2nd-order by Runge-Kutta method | |
| Zhao et al. | Vector L2-gain and stability of feedback switched systems | |
| NZ508930A (en) | N-Tuple or ram based neural network classification system having the output score functions and/or decision rules are determined based on the information of the determined column vector cells | |
| CN104467847A (en) | Calibration method for redundancy self-adaptation charge redistribution analog-digital converter | |
| CN102970037A (en) | Self-calibrating circuit of current source | |
| CN102025375A (en) | Analogue-to-digital converter and digital calibration circuit thereof | |
| Tao et al. | Filtering of two-dimensional periodic Roesser systems subject to dissipativity | |
| Cheng et al. | Error estimation of perturbations under CRI | |
| DE69802372T4 (en) | Classification system and method with N-tuple or RAM-based neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |