CN1983832A - Method and system for determining filter coefficient - Google Patents
Method and system for determining filter coefficient Download PDFInfo
- Publication number
- CN1983832A CN1983832A CN 200610060649 CN200610060649A CN1983832A CN 1983832 A CN1983832 A CN 1983832A CN 200610060649 CN200610060649 CN 200610060649 CN 200610060649 A CN200610060649 A CN 200610060649A CN 1983832 A CN1983832 A CN 1983832A
- Authority
- CN
- China
- Prior art keywords
- filter coefficient
- waveform
- chip
- combination
- pulse 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.)
- Pending
Links
Images
Landscapes
- Dc Digital Transmission (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
Abstract
The invention is concerned with the method and the system for confirm the coefficient of the filter, the system includes: the sending chip, the receiving chip, the system unit that connects with the sending chip and the receiving chip by the system unit interface, the method is: computes the channel characteristic according to the outputting waveform data of the sending chip and the receiving chip; analog creates the sending pulse signal that is same as the driving waveform edge of the sending chip; traverses the filter coefficient combination, based on each filter coefficient combination, gets the pulse response or eye pattern that is corresponding with the filter coefficient combination by combine the channel characteristic and the sending pulse signal; compares the eye height extent of the eye pattern to confirm the filter coefficient; or compares the waveform extent of the pulse respond and the coding crosstalk target to confirm the filter coefficient, solves the problem in the existing technology.
Description
Technical field
The present invention relates to communications field Signal Processing technology, especially a kind of method and system of definite filter coefficient.
Background technology
In communication process, digital signal is transmitted in communication channel and is had a more and more higher transmission rate, because the passage Insertion Loss characteristic of communication channel can exert an influence to signal, in case will influence quality of signals so communication channel produces loss, deterioration when signal is transmitted in communication channel is more and more obvious, in order to guarantee signal high-quality and transmission fast, the passage loss of communication channels such as printed circuit board and cable can not be left in the basket.
The passage loss comprises conductor losses and dielectric loss.In low-frequency range, conductor losses is generally big than dielectric loss; Exceed certain frequency range, dielectric loss is bigger than conductor losses.Under the more and more higher situation of signal rate, the passage loss has become the key factor of Signal Degrade, and the Insertion Loss characteristic of passage as shown in Figure 1.
At present, in order to overcome the deterioration effect of passage loss to signal, adopted signal is increased the weight of or balanced technology, use is at the transmission chip internal usually to increase the weight of technology, and balancing technique is used the receiving terminal at the inner or close receiving chip of receiving chip usually.Signal is increased the weight of or equilibrium treatment, see and Figure 2 shows that single order increases the weight of and balanced schematic diagram, can compensate the passage loss, offset to a certain extent because the deterioration that the passage loss causes signal, increase the weight of or balanced to the channel compensation characteristic as shown in Figure 3, solid line is a passage Insertion Loss characteristic, and dotted line is for increasing the weight of or equalizer curve, and dotted line is to increase the weight of or balanced characteristic curve after passage Insertion Loss characteristic is compensated.No matter signal is adopted and increase the weight of still equilibrium treatment, when the filter coefficient of input is different,, see Fig. 4, for single order increases the weight of or equalizing coefficient is the output waveform of 0.3 o'clock filter the degree of compensation difference of passage loss.Therefore, coefficient that need to determine filter comes to passage loss to compensate with the deterioration of offseting signal or the deterioration of signal is reduced to minimum degree as far as possible.
In the prior art, if employing increases the weight of technology and determines filter coefficient, at first need manually to travel through all accentuation filter coefficients, the receiving terminal eye pattern that manual measurement is corresponding with each accentuation filter coefficient then is by comparing the accentuation filter coefficient that eye pattern is searched required accentuation filter coefficient or determined to make Signal Degrade degree minimum.Traversal accentuation filter coefficient needs manually to be configured by computer program, needs special-purpose communication interface between computer and the chip.Simultaneously, at the receiving chip place, need the eye pattern of test received signal, this receiving chip need be connected with the oscillography display, shows eye pattern at the oscilloscope end.By the different accentuation filter coefficient of artificial traversal, and the quality of the eye pattern on the artificial judgment oscilloscope, choosing in required accentuation filter coefficient or all the accentuation filter coefficients to make Signal Degrade be reduced to minimum accentuation filter coefficient, then the accentuation filter coefficient of selecting is write chip, see Fig. 5.
Determine in the prior art that the concrete steps that increase the weight of the accentuation filter coefficient under the technology comprise:
1, a conduct in the combination of taking-up accentuation filter coefficient is when the accentuation filter coefficient of Pretesting;
2, by computer configuration, the accentuation filter coefficient is write the transmission chip;
3, send chip and send waveform to receiving chip, receiving chip is tested eye pattern at receiving terminal, and sends to display;
4, display acceptance test result and showing;
5, take out one group of new accentuation filter coefficient combination, repeat above-mentioned steps 2,3 and 4,, finish the eye pattern test of all filter coefficients up to having traveled through all filter coefficients;
6, all test results of artificial judgment demonstration are selected the accentuation filter coefficient, and the accentuation filter coefficient of determining is write chip.
Determine filter coefficient if adopt balancing technique, when equilibrium treatment when the receiving terminal, its method is identical with increasing the weight of technology, if it is balanced in receiving chip inside, the eye pattern of then measuring after the equilibrium is unpractical, must determine at the inner coefficient of equalizing wave filter of realizing of receiving chip by adaptive equalization algorithm.
In the prior art, if adopt balancing technique to determine filter coefficient, its equilibrium treatment is again in receiving chip inside, and traditional method is:
1, after receiving chip receives the waveform that sends chip, Wave data is carried out equilibrium treatment by equalizer;
2, the waveform after the equilibrium treatment is passed through decision device, obtain adjudicating the back waveform.It is poor that error circuit is responsible for waveform before the judgement and judgement back waveform, obtains the error of actual waveform and expectation waveform;
3, this error is passed to the adaptive algorithm device, according to this control information, the adaptive algorithm device can calculate one group of new equalizer coefficients, then this coefficient is write equalizer;
4,, finally this circuit can obtain a required equalizer coefficients by continuous error ratio.
Though described scheme of the prior art can satisfy the requirement that prevents Signal Degrade to a certain extent, but still has following shortcoming:
1, need manually to travel through filter coefficient in the prior art scheme, every traversal once needs the eye pattern of manual measurement receiving terminal, and traversal and the step measured cause complex operation step by manually repeating, and during operating cost.
2, the prior art scheme requires to be subjected to the restriction of physical condition by oscilloscope measurement receiving terminal eye pattern, and it is very difficult receiving waveform in the test of receiving chip place, or even impossible.
3, to measure the position of eye pattern generally be that package outside at receiving chip is the receiving terminal place to the prior art scheme, because Chip Packaging and junction capacitance influence, there is difference in the eye pattern that the eye pattern of receiving chip inside and package outside are measured, this differences affects the accuracy of definite filter coefficient.
In the prior art scheme for determining that filter coefficient need adopt complicated adaptive equalization technique in that receiving chip is inner, increased the design difficulty and the complexity of chip, increase chip cost and power consumption simultaneously.
Summary of the invention
The object of the present invention is to provide a kind of method and system of determining filter coefficient automatically.For achieving the above object, technical scheme provided by the invention is as follows:
A kind of method of definite filter coefficient, described method comprises step:
Calculate channel characteristic according to the waveform data that sends the output of chip and receiving chip;
Simulation produces the identical transmission pulse signal in edge with the drive waveforms that sends chip;
Traversal filter coefficient combination is based on each filter coefficient combination, in conjunction with described channel characteristic and described transmission pulse signal obtains and the combination of described filter coefficient is corresponding impulse response or eye pattern;
The eye high-amplitude of more described eye pattern is determined filter coefficient; Or the wave-shape amplitude of more described impulse response and intersymbol interference index are determined filter coefficient.
Described waveform data according to transmission chip and receiving chip output is calculated channel characteristic and specifically be may further comprise the steps:
Send chip and send alignment of waveforms to receiving chip;
Receiving chip is sampled to the described alignment of waveforms that receives, and obtains sample waveform;
Obtain first frequency spectrum of described alignment of waveforms and second frequency spectrum of described sample waveform respectively, and calculate channel characteristic according to described first, second frequency spectrum.
Described filter coefficient is combined as accentuation filter coefficient combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
According to described accentuation filter coefficient combination described transmission pulse signal is carried out filtering operation and obtain increasing the weight of waveform;
The described channel characteristic of inverse Fourier transform is the passage time domain waveform, and described passage time domain waveform and the described waveform that increases the weight of are carried out convolution algorithm acquisition reception waveform;
To the described reception waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
Described filter coefficient is combined as coefficient of equalizing wave filter combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
The described channel characteristic of inverse Fourier transform be the passage time domain waveform and with described passage time domain waveform and described transmission pulse signal carry out convolution algorithm obtain balanced before waveform;
According to described coefficient of equalizing wave filter combination waveform before the described equilibrium is carried out filtering operation and obtain balanced back waveform;
To the described balanced back waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
Described filter coefficient is combined as accentuation filter coefficient combination or coefficient of equalizing wave filter combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
According to described each group filter coefficient combination, to the frequency spectrum of described transmission pulse signal, the combination of described filter coefficient corresponding increasing the weight of or equilibrium treatment after filter spectrum and the frequency spectrum of described channel characteristic carry out computing, obtain the receiving terminal frequency spectrum;
The described receiving terminal frequency spectrum of inverse Fourier transform obtains receiving waveform;
To the described reception waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
A kind of system of definite filter coefficient, described system comprises:
Send chip, described transmission chip is used to send the verification waveform;
Receiving chip, described receiving chip are used to receive the verification waveform to sending chip, and send sample waveform;
System unit, described system unit is connected with described receiving chip with described transmission chip by the system unit interface, and described system unit comprises:
The waveform processing module, described waveform processing module is used to receive verification waveform and sample waveform, and calculates channel characteristic and produce the transmission pulse signal;
Memory module, described memory module are used for the combination of memory filter coefficient;
Calculation process module, described calculation process module are used to extract described filter coefficient combination, and in conjunction with coming channel characteristic and transmission pulse signal to the waveform processing module to determine filter coefficient.
Described calculation process module comprises:
Data extracting unit, described data extracting unit is connected with described memory module, is used for extracting successively from described memory module each group filter coefficient combination of storage;
Signal processing unit, in order to receive filter coefficient combination from this data extracting unit, utilize the combination of described filter coefficient and from the channel characteristic of described waveform processing module with send pulse signal and obtain eye pattern or pulse signal, and definite filter coefficient.
Described waveform processing module also comprises start unit, notifies described transmission chip and receiving chip to send Wave data in order to send starting command.
The beneficial effect that the present invention adopts this technical scheme to bring is as follows:
1, the test operation of the traversal of filter coefficient combination and eye pattern becomes coefficient elements by manual type and handles automatically.Automatically calculate eye pattern or impulse response, and all groups of filter coefficients of one-time write merge storage, after finishing, an eye pattern test extracts new filter coefficient combination automatically, reduced the loaded down with trivial details degree of operation, make step simplify, and saved the time, can finish determining of filter coefficient accurately, fast;
2, need not connect test and the demonstration that oscilloscope is realized eye pattern at the receiving chip place, make that the test of eye pattern is more convenient and can realize, and the filter coefficient of determining is more reasonable;
3, the eye pattern of coefficient arithmetic element acquisition has comprised the influence of Chip Packaging to signal, makes that the filter coefficient of determining is more accurate;
4, need not determine equalizing coefficient, realize determining of equalizing coefficient, can reduce the design difficulty and the complexity of chip, reduce chip cost and power consumption in chip sunset fore-telling portion by the adaptive algorithm device of complexity.
Description of drawings
Fig. 1 is the Insertion Loss characteristic schematic diagram of passive channel;
Fig. 2 increases the weight of for single order and balanced schematic diagram;
Fig. 3 is for increasing the weight of or balanced to channel compensation characteristic schematic diagram;
Fig. 4 increases the weight of for single order or equalizing coefficient is 0.3 filter output waveform;
Fig. 5 is prior art system figure;
Fig. 6 is for determining the schematic diagram of filter coefficient in the inner employing of receiving chip balancing technique in the prior art;
Fig. 7 is a system diagram of the present invention;
Fig. 8 is a transmission chip schematic diagram of the present invention;
Fig. 9 is a receiving chip schematic diagram of the present invention;
Figure 10 is a system of the present invention arithmetic element schematic diagram
Figure 11 is the basic principle figure of the inventive method.
Figure 12 is the flow chart of the inventive method.
Embodiment
Be explained in detail below in conjunction with the method and system of each accompanying drawing a kind of definite filter coefficient of the present invention.
Be illustrated in figure 7 as the schematic diagram of system of the present invention, in conjunction with this figure, system implementation mode of the present invention is as follows:
The system of described a kind of definite filter coefficient, comprise and send chip and receiving chip, described transmission chip is connected by communication channel with receiving chip, described system also comprise the system unit interface with by the system unit interface respectively with send the system unit that chip links to each other with receiving chip.
Wherein, send chip and comprise first interface circuit and first treatment circuit, see shown in the accompanying drawing 8.
First interface circuit is connected with the system unit interface.
First treatment circuit sends the verification waveform to system unit and receiving chip respectively by first interface circuit and described passage.
Receiving chip receives to the verification waveform that sends chip, start the waveform acquisition circuit and carry out waveform sampling, receiving chip also comprises second treatment circuit, and the waveform after described second treatment circuit will be sampled is sent to system unit by second interface circuit that is connected with the system unit interface.See shown in the accompanying drawing 9.
Described waveform acquisition circuit can pass through the A/D sample circuit, also can realize the sampling of waveform by decision device.
Definite process of filter coefficient is finished in described system unit control.Definite required parameters of filter coefficient is searched and calculated to system unit automatically, all groups of filter coefficients by automatic traversal storage merge in conjunction with parameters, obtain eye pattern or impulse response, required filter coefficient is determined in more described eye pattern or impulse response.
Described system unit comprises the waveform processing module, the waveform processing module receives to the verification waveform that sends chip and the sample waveform of receiving chip, obtain first frequency spectrum of described alignment of waveforms and second frequency spectrum of described sample waveform by described waveform, and according to described first, second frequency spectrum calculating channel characteristic, in addition, the waveform processing module also produces the transmission pulse signal identical with the edge of waveform according to the drive waveforms that sends chip;
Described system unit also comprises the memory module that is used for the combination of memory filter coefficient, and what be connected with described memory module is the calculation process module of system unit.
Described calculation process module is used to receive the channel characteristic of waveform processing module and sends pulse signal, and from memory module, extract each group filter coefficient successively and make up, obtain and corresponding eye pattern or the impulse response of each group filter coefficient combination in conjunction with channel characteristic and transmission pulse signal, determine the filter coefficient that system is required by the eye high-amplitude of more described eye pattern or the wave-shape amplitude and the intersymbol interference index of impulse response.
Described calculation process module comprises data extracting unit, and described data extracting unit is connected with described memory module, is used for extracting successively from described memory module each group filter coefficient combination of storage;
The signal processing unit that also has the calculation process module that is connected with described data extracting unit, signal processing unit is used for sending coefficient to described data extracting unit and extracts order, described data extracting unit receives the order of described signal processing unit, and the filter coefficient combination is sent to signal processing unit;
Described signal processing unit receives the output of described waveform processing module and described data extracting unit and calculates and corresponding eye pattern or the response impulse of described each group filter coefficient combination, the eye high-amplitude of more described eye pattern or the wave-shape amplitude of described impulse response and intersymbol interference index are determined filter coefficient.
In the waveform processing module of described system unit, can also comprise the order start unit, send chip and receiving arrival in the starting command of ordering start unit, explain the control command of sending from system unit by first treatment circuit, and beginning sends the verification waveform to system unit and receiving chip, accordingly, receiving chip receives to the starting command of ordering start unit, explain the control command of sending from system unit by second treatment circuit, and begin to start the waveform acquisition circuit to sampling to the verification waveform that sends chip, and the Wave data after will sampling sends to system unit and handles.
The system unit here can be CPU in the system master unit or the DSP in the system master unit, also can be the computer of independent operating.
Be the basic principle figure of the inventive method as shown in figure 11, the step of the inventive method is as follows:
Calculate channel characteristic according to the waveform data that sends the output of chip and receiving chip;
Simulation produces the identical transmission pulse signal in edge with the drive waveforms that sends chip;
The combination of traversal filter coefficient is obtained impulse response corresponding with described filter combination or eye pattern based on each filter coefficient combination, described channel characteristic and described transmission pulse signal;
The eye high-amplitude of more described eye pattern is determined filter coefficient; Or the wave-shape amplitude of more described impulse response and intersymbol interference index are determined filter coefficient.
Be the inventive method flow chart as shown in figure 12, in conjunction with this figure, the execution mode of the inventive method is as follows:
Step 1: system unit sends starting command to sending chip and receiving chip, send chip and receive described order back to system unit and receiving chip transmission verification waveform, receiving chip will send to system unit to the sample waveform behind the verification waveform sampling after receiving described order.
Step 2: system unit receives to the sample waveform of verification waveform that sends chip and receiving chip, produces the transmission pulse signal identical with the edge of the drive waveforms that sends chip and calculates channel characteristic according to above-mentioned Wave data simulation.
Waveform processing module in the system unit receives to the verification waveform that sends chip and simulates the identical transmission pulse signal in drive waveforms edge that produces and send chip, time domain convolution theorem in while basis signal and the system, be that signal and system carry out in time domain that convolution equals signal and system multiplies each other at frequency domain, obtain first frequency spectrum of described alignment of waveforms and second frequency spectrum of described sample waveform respectively, and calculate channel characteristic according to described first, second frequency spectrum.Its computational process is specially: according to hypothesis passage frequency spectrum is H (w), and the transmission signal spectrum is X (w), so the frequency spectrum R (w) of received signal=X (w) * H (w).The frequency spectrum of present known transmission signal spectrum and received signal, i.e. the Wave data of system unit reception: verification waveform and sample waveform, passage spectrum H (w)=R (w)/X (w) just can calculate the Insertion Loss characteristic of passive channel, i.e. channel characteristic so.
Step 3: the traversal filter coefficient, obtain eye pattern or pulse signal in conjunction with the described channel characteristic that calculates with the transmission pulse signal.The detailed process of obtaining eye pattern or pulse signal is as follows:
01, data extracting unit is extracted a filter combination from the filter coefficient combination of memory module storage;
02, adopt the processing method of time domain or frequency domain to obtain eye pattern or impulse response according to described filter combination in conjunction with channel characteristic and transmission pulse signal;
03, preserves described eye pattern or impulse response, extract a new filter coefficient combination again and adopt above-mentioned processing method to obtain and corresponding eye pattern or the impulse response of described filter coefficient combination;
04, traversal filter coefficient combination is made up up to having extracted all filter coefficients, obtains one group and makes up corresponding eye pattern or impulse response with each filter coefficient;
In above-mentioned 01 step, all coefficient combinations of filter are stored in the memory module of described system unit.With the firstorder filter coefficient is example, is output as Y (n)=(1+b) * X (n)-b*x (n-1).Obtain different filter coefficient combinations according to different b.For example: when the b excursion is: 0.1-0.4, step-length 0.1, the filter coefficient combination has 4 kinds so: 1.1 ,-0.1; 1.2 ,-0.2; 1.3 ,-0.3; 1.4 ,-0.4, described filter combination one-time write memory module, storage and uniform, traversal has reduced the loaded down with trivial details degree of operating automatically, makes step simplify, and has saved the time.
In above-mentioned steps 02, the method that time domain is calculated eye pattern or pulse signal promptly adopts the convolution algorithm method.
When employing increased the weight of technology and determines filter coefficient, described filter coefficient was and increases the weight of coefficient.For increasing the weight of determining of coefficient, at first increase the weight of to handle, promptly according to the filter coefficient combination of extracting, carry out filtering operation to sending pulse signal, obtain increasing the weight of the back waveform, and then trying to achieve channel characteristic is carried out inverse Fourier transform obtain the passage time domain waveform, and with the passage time domain waveform with increase the weight of the back waveform and carry out convolution algorithm, calculate the reception response wave shape.
When adopting balancing technique to determine filter coefficient, described filter coefficient is an equalizing coefficient.For increasing the weight of determining of coefficient, at first trying to achieve channel characteristic is carried out inverse Fourier transform and obtain the passage time domain waveform, and with described passage time domain waveform with send pulse signal and carry out waveform before the convolution algorithm equilibrium.The filter coefficient that equilibrium treatment is described extraction carries out filtering operation, calculates balanced back waveform.
The method of frequency-domain calculations receiving terminal response signal is at one group of filter coefficient combination, to send signal spectrum, increase the weight of or balanced filter spectrum characteristic and passage spectral characteristic three are multiplied each other and obtained receiving frequency domain data, the reception frequency domain data that obtains then carries out the reception response wave shape that inverse Fourier transform obtains and the combination of described filter coefficient is corresponding.
For the balanced back waveform corresponding with receive response wave shape and superpose by the bit wide of each bit and obtain eye pattern, preserve described eye pattern with each group filter coefficient.
System unit also can carry out the impulse response that convolution algorithm obtains passage by single-bit pulse and passage impulse response, and channel characteristic herein and passage impulse response are reciprocity, and the single-bit pulse is a special case that sends pulse.
Step 4: more described eye pattern or pulse signal, according to the eye high-amplitude of described eye pattern, perhaps the wave-shape amplitude of described impulse response and intersymbol interference index are determined required filter coefficient combination;
Require down for the coefficient of determining to choose eye high-amplitude maximum in the eye pattern of filter coefficient or satisfying receiving chip decision level and certain voltage margin, choose a coefficient of high-amplitude minimum farthest reduces signal as the optimum filter coefficient deterioration, after perhaps deducting the intersymbol interference amplitude by wave-shape amplitude in impulse response, the filter coefficient of correspondence was as the optimum filter coefficient when its result still can satisfy the requirement of receiving terminal receiving margin.
Choose the filter coefficient that suitable signal that its needs obtain transmits for the demand of determining also to answer the gauger of filter coefficient according to eye pattern or impulse response in channel.
In order to prevent that power down from losing, coefficient can write flash memory, does not lose after the power down.After finishing coefficient and being provided with, chip can call this coefficient repeatedly, and passive channel is compensated.
The system unit here can be CPU in the system master unit or the DSP in the system master unit, also can be the computer of independent operating.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.
Claims (8)
1, a kind of method of definite filter coefficient is characterized in that, described method comprises step:
Calculate channel characteristic according to the waveform data that sends the output of chip and receiving chip;
Simulation produces the identical transmission pulse signal in edge with the drive waveforms that sends chip;
Traversal filter coefficient combination is based on each filter coefficient combination, in conjunction with described channel characteristic and described transmission pulse signal obtains and the combination of described filter coefficient is corresponding impulse response or eye pattern;
The eye high-amplitude of more described eye pattern is determined filter coefficient; Or the wave-shape amplitude of more described impulse response and intersymbol interference index are determined filter coefficient.
2. method according to claim 1 is characterized in that, described waveform data according to transmission chip and receiving chip output is calculated channel characteristic and specifically be may further comprise the steps:
Send chip and send alignment of waveforms to receiving chip;
Receiving chip is sampled to the described alignment of waveforms that receives, and obtains sample waveform;
Obtain first frequency spectrum of described alignment of waveforms and second frequency spectrum of described sample waveform respectively, and calculate channel characteristic according to described first, second frequency spectrum.
3, method according to claim 1, it is characterized in that, described filter coefficient is combined as accentuation filter coefficient combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
According to described accentuation filter coefficient combination described transmission pulse signal is carried out filtering operation and obtain increasing the weight of waveform;
The described channel characteristic of inverse Fourier transform is the passage time domain waveform, and described passage time domain waveform and the described waveform that increases the weight of are carried out convolution algorithm acquisition reception waveform;
To the described reception waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
4, method according to claim 1, it is characterized in that, described filter coefficient is combined as coefficient of equalizing wave filter combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
The described channel characteristic of inverse Fourier transform be the passage time domain waveform and with described passage time domain waveform and described transmission pulse signal carry out convolution algorithm obtain balanced before waveform;
According to described coefficient of equalizing wave filter combination waveform before the described equilibrium is carried out filtering operation and obtain balanced back waveform;
To the described balanced back waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
5, method according to claim 1, it is characterized in that, described filter coefficient is combined as accentuation filter coefficient combination or coefficient of equalizing wave filter combination, described based on each filter coefficient combination, obtain impulse response or the eye pattern corresponding in conjunction with described channel characteristic and described transmission pulse signal and specifically comprise with described filter coefficient combination:
According to described each group filter coefficient combination, to the frequency spectrum of described transmission pulse signal, the combination of described filter coefficient corresponding increasing the weight of or equilibrium treatment after filter spectrum and the frequency spectrum of described channel characteristic carry out computing, obtain the receiving terminal frequency spectrum;
The described receiving terminal frequency spectrum of inverse Fourier transform obtains receiving waveform;
To the described reception waveform acquisition eye pattern that superposes, perhaps,
Carry out convolution algorithm and obtain impulse response sending pulse signal and channel characteristic.
6, a kind of system of definite filter coefficient is characterized in that, described system comprises:
Send chip, described transmission chip is used to send the verification waveform;
Receiving chip, described receiving chip are used to receive the verification waveform to sending chip, and send sample waveform;
System unit, described system unit is connected with described receiving chip with described transmission chip by the system unit interface, and described system unit comprises:
The waveform processing module, described waveform processing module is used to receive verification waveform and sample waveform, and calculates channel characteristic and produce the transmission pulse signal;
Memory module, described memory module are used for the combination of memory filter coefficient;
Calculation process module, described calculation process module are used to extract described filter coefficient combination, and in conjunction with coming channel characteristic and transmission pulse signal to the waveform processing module to determine filter coefficient.
7, system according to claim 6 is characterized in that, described calculation process module comprises:
Data extracting unit, described data extracting unit is connected with described memory module, is used for extracting successively from described memory module each group filter coefficient combination of storage;
Signal processing unit, in order to receive filter coefficient combination from this data extracting unit, utilize the combination of described filter coefficient and from the channel characteristic of described waveform processing module with send pulse signal and obtain eye pattern or pulse signal, and definite filter coefficient.
8, system according to claim 6 is characterized in that,
Described waveform processing module also comprises start unit, notifies described transmission chip and receiving chip to send Wave data in order to send starting command.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610060649 CN1983832A (en) | 2006-05-01 | 2006-05-01 | Method and system for determining filter coefficient |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610060649 CN1983832A (en) | 2006-05-01 | 2006-05-01 | Method and system for determining filter coefficient |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1983832A true CN1983832A (en) | 2007-06-20 |
Family
ID=38166123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200610060649 Pending CN1983832A (en) | 2006-05-01 | 2006-05-01 | Method and system for determining filter coefficient |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1983832A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010081340A1 (en) * | 2009-01-14 | 2010-07-22 | 华为终端有限公司 | Method and apparatus for determining filter coefficient |
| CN103200040A (en) * | 2012-11-08 | 2013-07-10 | 盛科网络（苏州）有限公司 | Interface signal quality testing method and system based on special chip |
| CN103885907A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Equalization Coefficient Search Algorithm |
-
2006
- 2006-05-01 CN CN 200610060649 patent/CN1983832A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010081340A1 (en) * | 2009-01-14 | 2010-07-22 | 华为终端有限公司 | Method and apparatus for determining filter coefficient |
| CN103200040A (en) * | 2012-11-08 | 2013-07-10 | 盛科网络（苏州）有限公司 | Interface signal quality testing method and system based on special chip |
| CN103200040B (en) * | 2012-11-08 | 2016-03-16 | 盛科网络（苏州）有限公司 | Based on method of testing and the system of the interface signal quality of special chip |
| CN103885907A (en) * | 2012-12-20 | 2014-06-25 | 辉达公司 | Equalization Coefficient Search Algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7295642B2 (en) | Jitter compensation and generation in testing communication devices | |
| EP2680452B1 (en) | Re-Sampling S-Parameters for Serial Data Link Analysis | |
| JP5306445B2 (en) | Method for transmitting a waveform having controllable attenuation and propagation velocity | |
| US7940056B2 (en) | Time-domain reflectometry | |
| US20100176815A1 (en) | Apparatus, Method, and Computer Program for Obtaining a Time-Domain-Reflection Response-Information | |
| US20070223571A1 (en) | Decision-feedback equalizer simulator | |
| US20120320964A1 (en) | Methods and systems for providing optimum decision feedback equalization of high-speed serial data links | |
| AU2020311281B2 (en) | Adaptive hopping equalizer | |
| CN116048897B (en) | High-speed serial signal receiving end pressure eye diagram construction and testing method and system | |
| CN1983832A (en) | Method and system for determining filter coefficient | |
| EP3176591B1 (en) | Measurement apparatus | |
| US6760372B1 (en) | Adaptive signal processor using an eye-diagram metric | |
| EP3179259B1 (en) | Group delay based averaging | |
| US7852912B2 (en) | Direct determination equalizer system | |
| US10523341B1 (en) | Methods and systems for in-situ crosstalk measurements in communication systems | |
| CN101729097B (en) | Signal processing method, control module, interference cancellation device and communication system | |
| US10185020B1 (en) | Method of compensating loss and dispersion of transmission line for time domain reflectometry | |
| CN103560809B (en) | Communication device and noise cancellation method | |
| NZ779952B2 (en) | Adaptive hopping equalizer | |
| CN115695098A (en) | Apparatus and method for generating an estimate of a channel frequency response | |
| Moon et al. | Efficient Built-In Test and Calibration of High Speed Serial I/O Systems Using Monobit Signal Acquisition | |
| CN116566481A (en) | Consistency testing method and related device | |
| CN116805887A (en) | Digital optical module adjustment and measurement method and system for communication | |
| CN117728886A (en) | TDEC (time domain reflectometry) testing method and related equipment | |
| Nakamura et al. | Novel crosstalk evaluation method for high-density signal traces using clock waveform conversion technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20070620 |