CN105281792B - The method of adjustment and device at adaptive equalizer coefficients center - Google Patents
The method of adjustment and device at adaptive equalizer coefficients center Download PDFInfo
- Publication number
- CN105281792B CN105281792B CN201410360122.1A CN201410360122A CN105281792B CN 105281792 B CN105281792 B CN 105281792B CN 201410360122 A CN201410360122 A CN 201410360122A CN 105281792 B CN105281792 B CN 105281792B
- Authority
- CN
- China
- Prior art keywords
- coefficient
- input
- center
- adaptive equalizer
- input coefficient
- 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
Landscapes
- Optical Communication System (AREA)
Abstract
The invention discloses a kind of methods of adjustment at adaptive equalizer coefficients center, applied in high-speed optical transmission system, the method of adjustment at the adaptive equalizer coefficients center is the following steps are included: input coefficient needed for obtaining adaptive equalizer, whether the coefficient center for monitoring the input coefficient shifts relative to tap coefficient center；If shifting, the input coefficient is adjusted；The corresponding input data of input coefficient adjusted is filtered, filtered output data insert and deletes value processing.The invention also discloses a kind of adjustment devices at adaptive equalizer coefficients center, the present invention can be realized effective control to adaptive equalizer input coefficient, stablize the coefficient center of input coefficient in coefficient tap coefficient center, system stability is improved, and reduces the bit error rate of system.
Description
Technical field
The present invention relates to high-speed optical transmission technical field more particularly to a kind of adjustment sides at adaptive equalizer coefficients center
Method and device.
Background technique
In high-speed optical transmission system, coherent receiver needs to compensate the various damages in optical channel, such as polarizing coating dispersion
(Polarization Mode Dispersion, PMD), chromatic dispersion (Chromatic Dispersion, CD) and polarization phase
Close loss (Polarization Dependent Loss, PDL) etc..Adaptive equalizer passes through Digital Signal Processing
(Digital Signal Processing, DSP) adaptive tracing characteristic of channel, to compensate the various damages in optical channel.
As shown in Figure 1, coherent receiver mainly include dispersion compensation, clock recovery, adaptive equalization and offset estimation,
The modules such as skew estimation (wherein, H, V indicate each road input).The function of dispersion compensation is the chromatic dispersion compensated in channel, when
The function that clock restores is the clock synchronization issue solved between transmitter and receiver, and the function of adaptive equalization is compensation
The function of PMD, remnants CD, PDL and polarization mode demultiplexing, offset estimation and skew estimation is that correction emitting laser and local oscillator swash
Between light device there is frequency and phase deviation.
Adaptive equalizer plays the role of very important in coherent receiver.Adaptive equalizer is by several limited punchings
It hits response (Finite impulse response, FIR) filter and the blind equalization for generating coefficient needed for FIR is calculated is calculated
Method unit, as constant modulus algorithm (Common Modulus Algorithm, CMA) unit forms.The adaptive convergence process of coefficient
It plays a key effect for the steady operation of adaptive equalizer.But since channel impairments are time-varying, it may result in FIR
Coefficient off center position go at index boundaries, or even go to outside coefficient tap, therefore system error code increases suddenly, or even need
Adaptive equalizer is restrained again, and the coefficient of adaptive equalizer is unable to get effective control, high-speed optical transmission system work
It is unstable.
The center of gravity that a kind of usage factor exists in the prior art comes the method for adjustment of control coefrficient, including feedback signal to slotting
Value module carries out the fine tuning of coefficient and when coefficient monitors the heart and is greater than certain thresholding at a distance from tap coefficient center again, by coefficient
Mobile integral multiple sampling point.Since coefficient matrix is the myopia solution for meeting orthogonal property after adaptive equalizer convergence, coefficient
Be on Focus Theory it is metastable, not will lead to X/Y polarization state when coefficient fine tuning and deviated together toward the same direction, therefore not
The coefficient center of X/Y polarization state can be all adjusted to tap coefficient center, and carrying out the inclined of coefficient integral multiple sampling point
When shifting, in the prior art in order to not influence the synchronization of high-speed optical transmission system, 0 is directly filled in filter output data, this
It is higher that sample will lead to the bit error rate.It therefore, still cannot be effectively by coefficient adjustment to tap coefficient center, system stability
It is poor.
Summary of the invention
It is a primary object of the present invention to solve the coefficient center of adaptive equalizer cannot effectively be adjusted to tap system
The technical problem for counting center and causing system unstable.
To achieve the above object, the present invention provides a kind of method of adjustment at adaptive equalizer coefficients center, is applied to height
In fast optical transmission system, the method for adjustment at the adaptive equalizer coefficients center the following steps are included:
When detecting adaptive equalizer has data input, input coefficient needed for obtaining adaptive equalizer, monitoring
Whether the coefficient center of the input coefficient shifts relative to tap coefficient center；
If shifting, the input coefficient is adjusted；
It is filtered according to data of the input coefficient adjusted to input, filtered output data insert and deletes value
Processing.
Preferably, input coefficient needed for the acquisition adaptive equalizer, monitors the coefficient center of the input coefficient
The step of whether shifting relative to tap coefficient center include:
Input coefficient needed for obtaining the adaptive equalizer, and calculate the characteristic parameter of the input coefficient；
Whether occurred according to the coefficient center that the characteristic parameter monitors the input coefficient relative to tap coefficient center
Offset.
Preferably, input coefficient needed for the acquisition adaptive equalizer, monitors the coefficient center of the input coefficient
The step of whether shifting relative to tap coefficient center include:
Input coefficient needed for obtaining the adaptive equalizer, and calculate the input coefficient in preset time period
Characteristic parameter；
Obtain the corresponding average value of the characteristic parameter；
Whether occurred partially according to the coefficient center that the average value monitors the input coefficient relative to tap coefficient center
It moves.
Preferably, input coefficient needed for the acquisition adaptive equalizer, monitors the coefficient center of the input coefficient
The step of whether shifting relative to tap coefficient center include:
Input coefficient needed for obtaining the adaptive equalizer, calculates the characteristic parameter of the input coefficient, and obtains
The characterization parameter of the polarizing coating dispersion of the adaptive equalizer；
The distance of the input coefficient peak value is calculated according to the baud rate of the characterization parameter and the adaptive equalizer；
The coefficient center phase of the input coefficient is monitored according to the distance of the input coefficient peak value and the characteristic parameter
Whether shift for tap coefficient center.
Preferably, which is characterized in that if described shift, the step of being adjusted to the input coefficient includes:
The offset direction that the coefficient center of the input coefficient shifts is obtained, according to the offset direction to described defeated
Enter coefficient to be adjusted.
In addition, to achieve the above object, the present invention also provides a kind of adjustment device at adaptive equalizer coefficients center,
It is characterized in that, the adjustment device includes:
Monitoring modular, for obtaining needed for adaptive equalizer when detecting adaptive equalizer has data input
Whether input coefficient, the coefficient center for monitoring the input coefficient shift relative to tap coefficient center；
Module is adjusted, if being adjusted for shifting to the input coefficient；
Processing module, for being filtered according to data of the input coefficient adjusted to input, to filtered output
Data, which insert, deletes value processing.
Preferably, the monitoring modular includes:
First computing unit for input coefficient needed for obtaining the adaptive equalizer, and calculates the input system
Several characteristic parameters；
First monitoring unit, for monitoring the coefficient center of the input coefficient according to the characteristic parameter relative to tap
Whether coefficient center shifts.
Preferably, the monitoring modular includes:
Second computing unit for input coefficient needed for obtaining the adaptive equalizer, and calculates preset time period
The characteristic parameter of the interior input coefficient；
First acquisition unit, for obtaining the corresponding average value of the characteristic parameter；
Second monitoring unit, for monitoring the coefficient center of the input coefficient according to the average value relative to tap system
Whether number center shifts.
Preferably, the monitoring modular includes:
Second acquisition unit calculates the input coefficient for input coefficient needed for obtaining the adaptive equalizer
Characteristic parameter, and obtain the characterization parameter of the polarizing coating dispersion of the adaptive equalizer；
Third computing unit, it is described defeated for being calculated according to the baud rate of the characterization parameter and the adaptive equalizer
Enter the distance of coefficient peak value；
Third monitoring unit, for monitoring the input according to the distance of the input coefficient peak value and the characteristic parameter
Whether the coefficient center of coefficient shifts relative to tap coefficient center.
Preferably, the adjustment module is specifically used for obtaining the offset side that the coefficient center of the input coefficient shifts
To being adjusted according to the offset direction to the input coefficient.
A kind of method of adjustment and dress applied to the adaptive equalizer coefficients center in high-speed optical transmission system of the present invention
It sets, then input coefficient needed for calculating FIR filter by adaptive blind equalization algorithm is filtered, while to blind equal
The coefficient center for the input coefficient that the method for accounting acquires is monitored, if coefficient center occurs relative to tap coefficient center
Offset, then be adjusted input coefficient, then insert to filtered output data and deletes value processing, the present invention is realized to certainly
Effective control of adaptive equalizer input coefficient stablizes the coefficient center of input coefficient in coefficient tap coefficient centre bit
It sets, improves system stability, and reduce the bit error rate of system.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the coherent receiver of high-speed optical transmission system in the prior art；
Fig. 2 is the flow diagram of one embodiment of method of adjustment at adaptive equalizer coefficients center of the present invention；
Fig. 3 is the exemplary diagram of adaptive equalizer in Fig. 2；
Fig. 4 is the adjustment of input coefficient and the slotting flow example figure for deleting value processing in Fig. 2；
Fig. 5 is input coefficient needed for obtaining adaptive equalizer in Fig. 2, monitors the coefficient center phase of the input coefficient
Whether shift the refinement flow diagram of an embodiment for tap coefficient center；
Fig. 6 is the flow example figure that the judgement of coefficient off-centring is carried out in Fig. 5；
Fig. 7 is input coefficient needed for obtaining adaptive equalizer in Fig. 2, monitors the coefficient center phase of the input coefficient
Whether shift the refinement flow diagram of another embodiment for tap coefficient center；
Fig. 8 is the flow example figure that the judgement of coefficient off-centring is carried out in Fig. 7；
Fig. 9 is input coefficient needed for obtaining adaptive equalizer in Fig. 2, monitors the coefficient center phase of the input coefficient
Whether shift the refinement flow diagram of another embodiment for tap coefficient center；
Figure 10 is the flow example figure that the judgement of coefficient off-centring is carried out in Fig. 9；
Figure 11 is the functional block diagram of adjustment one embodiment of device at adaptive equalizer coefficients center of the present invention；
Figure 12 is the refinement the functional block diagram of one embodiment of monitoring modular in Figure 11；
Figure 13 is the refinement the functional block diagram of another embodiment of monitoring modular in Figure 11；
Figure 14 is the refinement the functional block diagram of monitoring modular another embodiment in Figure 11.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The present invention provides a kind of method of adjustment applied to the adaptive equalizer coefficients center in high-speed optical transmission system,
Referring to Fig. 2, in one embodiment, the method for adjustment at the adaptive equalizer coefficients center includes:
Step S101, when detecting adaptive equalizer has data input, input needed for obtaining adaptive equalizer
Whether coefficient, the coefficient center for monitoring the input coefficient shift relative to tap coefficient center；
The present embodiment method hardware running environment is related to adaptive equalizer in high-speed optical transmission system, based on to adaptive
Effective control of balanced device input coefficient is stablized the coefficient center of input coefficient in tap coefficient center, is improved
System stability.
Specifically, the present embodiment adaptive equalization implement body schematic diagram is as shown in Figure 3.Wherein, it is calculated by adaptive blind equalization
Input coefficient needed for method (by taking CMA algorithm as an example) calculates FIR filter, is then filtered input data.Meanwhile it is right
The output data of FIR filter, which insert, deletes value processing, farthest to reduce the bit error rate.
Wherein, FIR filter can perhaps frequency domain realizes blind equalization algorithm or in time domain or frequency domain reality in time domain
It is existing.
To describe the present embodiment method, the FIR filter mathematic(al) representation in adaptive equalizer makees following agreement:
Wherein, M is the tap number of FIR filter.Input coefficient cxh(m)、cxv(m)、cyh(m) and cyv(m) by such as
The blind equalization algorithms such as CMA generate.cxh(m) refer to that input h is mapped to the coefficient in x-polarisation state, cxv(m) refer to that output v is mapped to
Coefficient in x-polarisation state, cyh(m) refer to that input h is mapped to the coefficient in y-polarisation state, cyv(m) it is inclined to refer to that output v is mapped to y
Coefficient in polarization state.
The technical solution that the present embodiment uses is consistent the processing mode of polarization state X or polarization state Y, between polarization state
Keep independent, in full text by taking polarization state X as an example.
In the present embodiment, the characteristic parameter of input coefficient is further calculated, characteristic parameter includes the range value of input coefficient
And its corresponding serial number, the number of the energy of boundary input coefficient and boundary input coefficient, the coefficient for characterizing input coefficient
Center is at a distance from the relationship of tap coefficient center etc..
In the present embodiment, by judging by characteristic parameter formation condition, so as to monitor input coefficient center
Whether shift.
Step S102 is adjusted the input coefficient if shifting；
It is right if the coefficient center of input coefficient shifts relative to tap coefficient center in the present embodiment
Input coefficient is adjusted, and specifically, can be adjusted according to offset direction to input coefficient or be adjusted to the right to the left, if not having
It shifts, input coefficient does not adjust.
In the present embodiment, it is assumed that the movement of samp sampling point need to be carried out to input coefficient.Divide following situations processing:
When input coefficient adjusts to the left,
Wherein, cxh'(m)、cxv' (m) be through input coefficient adjusted to the left.
When input coefficient adjusts to the right,
Wherein, cxh'(m)、cxv' (m) be through input coefficient adjusted to the right.
Step S103 is filtered, to filtered output data according to data of the input coefficient adjusted to input
Insert and deletes value processing.
In the present embodiment, due to being adjusted to input coefficient, the output data for inevitably resulting in FIR filter generates one
Fixed influence causes the generation of the bit error rate, and therefore, it is necessary to insert to filtered output data deleting value processing.
In the present embodiment, defining the FIR filter output data length in adaptive equalizer is N.Please refer to figure
4, the output data of FIR filter insert and deletes value point following steps:
Currently step a carries out group number according to [samp+ (M-1)+N] and inputs, i.e., when carrying out adaptive equalization FIR filtering
The FIR data N of timeticks adds (M-1) a data of previous timeticks tail portion, adds samp sampling point mobile data,
Then FIR filtering is carried out jointly, and FIR filter exports (N+samp) a sampling point.
Step b insert to the output data of FIR filtering and is deleted value processing, as shown in figure 4, in the following several ways:
Step b1: when not being adjusted to input coefficient, adaptive equalizer exports the rear N point data of (N+samp)；
Step b2: when input coefficient is adjusted to the right, adaptive equalizer exports the rear N-samp point of (N+samp)
Data；
Step b3: when input coefficient is adjusted to the left, adaptive equalizer exports (N+samp) point data.
Step c inserts step b and deletes value treated that data carry out butterfly shaping, the data of output fixation N length.
Compared with prior art, the present embodiment is calculated in high-speed optical transmission system by adaptive blind equalization algorithm
Input coefficient needed for FIR filter, is then filtered, while the coefficient center of the input coefficient acquired to blind equalization algorithm
Position is monitored, if coefficient center shifts relative to tap coefficient center, is adjusted to input coefficient, then
Filtered output data insert and deletes value processing, the present embodiment realizes effective control to adaptive equalizer input coefficient
System stablizes the coefficient center of input coefficient in coefficient tap coefficient center, improves system stability, and reduce and be
The bit error rate of system.
In a preferred embodiment, as shown in figure 5, on the basis of the embodiment of above-mentioned Fig. 2, above-mentioned steps S101 packet
It includes:
Step S1011, input coefficient needed for obtaining the adaptive equalizer, and calculate the feature of the input coefficient
Parameter；
Step S1012 monitors the coefficient center of the input coefficient relative in tap coefficient according to the characteristic parameter
Whether the heart shifts.
In the present embodiment, for X polarization state, the characteristic parameter of input coefficient is calculated, comprising:
Step 1, range value s1 (m), the s2 (m) of input coefficient are calculated:
S1 (m)=(real (cxh(m)))2+(imag(cxh(m)))2；
S2 (m)=(real (cxv(m)))2+(imag(cxv(m)))2；
Wherein, [1, M] m ∈, real are to take real, and imag is the imaginary part for taking plural number.
Step 2, maximum value m1, m2 of calculating input coefficient range value and corresponding serial number i1, i2, for judging input
The main peak of coefficient energy:
M1=max (s1 (m)), i1 are serial number of the m1 in s1 (m)；
M2=max (s2 (m)), i2 are serial number of the m2 in s2 (m).
Step 3, the energy of border coefficient is calculated, parameter P1 and P2 are set, the leftmost number of input coefficient is respectively represented
With the number of coefficient rightmost, calculate respectively with cxh(m)、cxv(m) corresponding right boundary energy sl1, sl2 and sr1, sr2,
For preventing input coefficient from stablizing on boundary or going to outside index boundaries:
With
Step 4, following formula s3 (m), s4 (m), s5 (m), s6 (m) and s5 (m), maximum value m3, m4 of s6 (m) and institute are calculated
Corresponding serial number i3, i4 highlights the peak value of input coefficient energy by related operation in conjunction with the two-way coefficient of X polarization state:
S3 (m)=m2 × s1 (m), s4 (m)=m1 × s2 (m)；
S5 (m)=| s3 (m)+s4 (m) |, s6 (m)=| s3 (m)-s4 (m) |；
M3=max (s5 (m)), i3 are serial number of the m3 in s5 (m)；
M4=max (s6 (m)), i4 are serial number of the m4 in s6 (m).
Step 5, the minimum value mvmin and the maximum value imax of i1, i2, i3, i4, minimum value of m1, m2, m3, m4 are calculated
Imin, the peak value for judgement factor energy.
Step 6, relationship gap d1, the d2 for being used for judgement factor center and tap coefficient center are calculated:
D2=imax-imin.
So far, calculation of characteristic parameters finishes.
In the present embodiment, in conjunction with refering to Fig. 6, all " Y " indicates that condition meets in Fig. 6, and all " N " is discontented for condition
Foot.According to the above-mentioned characteristic parameter being calculated, the judgement of input coefficient off-centring is carried out, coefficient center is obtained and shifts
Offset direction, a point following situations is adjusted to input coefficient according to offset direction and is handled:
Condition 1:(i1≤P3&&i2 >=P4) | | (i2≤P3&&i1 >=P4), i.e., as i1≤P3 and i2 >=P4, or
When person, i2≤P3 and i1 >=P4, input coefficient is not adjusted；Parameter P3, P4 represents the serial number of coefficient.
Condition 2:(m1 > G1&&i1≤P3) | | (m2 > G1&&i2≤P3) works as m1 > G1 and i1≤P3, alternatively, m2 >
G1 and when i2≤P3, input coefficient is adjusted to the right.Parameter G1 is input coefficient energy threshold.
Condition 3:(m1 > G1&&i1 >=P4) | | (m2 > G1&&I2 >=P4) works as m1 > G1 and i1 >=P4, alternatively, m2 >
G1 and when i2 >=P4, input coefficient is adjusted to the left.
Condition 4:(sl1 > G2 | | sl2 > G2) && (m1 > G1&&i1 >=P5&&i1≤P6) && (m2 > G1&&i2 >=P5&&
I2≤P6), that is, work as sl1 > G2 or sl2 > G2, and m1 > G1, i1 >=P5, i1≤P6, and m2 > G1, i2 >=P5, i2≤
When P6, input coefficient is adjusted to the right.Parameter P5, P6 represents the serial number of coefficient.
Condition 5:(sr1 > G2 | | sr2 > G2) && (m1 > G1&&i1 >=P7&&i1≤P8) && (m2 > G1&&i2 >=P7&&
I2≤P8), that is, work as sr1 > G2 or sr2 > G2, and m1 > G1, i1 >=P7, i1≤P8, and m2 > G1, i2 >=P7, i2≤
When P8, input coefficient adjusts to the left.Parameter P7, P8 represents the serial number of coefficient.
Condition 6:m1 > G2&&m2 > G2&&mvmin > G3&& (d1 >=P9&&d2 >=P10&&imin >=P11), that is, work as m1 >
G2 and m2 > G2 and mvmin > G3, and d1 >=P9 and d2 >=P10 and when imin >=P11, input coefficient adjusts to the left.
Condition 7:m1 > G2&&m2 > G2&&mvmin > G3&& (d1≤- P9&&d2 >=P10&&imax >=P12), that is, work as m1
> G2 and m2 > G2 and mvmin > G3, and d1≤- P9 and d2 >=P10 and when imax >=P12, input coefficient adjusts to the right.Parameter
P9, P10 and P11 represent the serial number of coefficient.
Be unsatisfactory for above-mentioned condition it is other under the conditions of, input coefficient without adjustment.
In an additional preferred embodiment, as shown in fig. 7, on the basis of the embodiment of above-mentioned Fig. 2, above-mentioned steps S101
Can also include:
Step S1013, input coefficient needed for obtaining the adaptive equalizer, and calculate described in preset time period
The characteristic parameter of input coefficient；
Step S1014 obtains the corresponding average value of the characteristic parameter；
Step S1015 monitors the coefficient center of the input coefficient relative to tap coefficient center according to the average value
Whether shift.
In the present embodiment, since the coefficient center of adaptive equalizer input coefficient can be according to the spy of polarizing coating dispersion PMD
Property change, the present embodiment by the characterisitic parameter to input coefficient averagely come improve to coefficient center estimate it is accurate
Property, to reach the adjustment to input coefficient more accurately.
By taking the FIR tap number of adaptive equalizer is 16 as an example (m=16), the input coefficient number for needing to adjust is 2,
FIR output data length is 128, is described by taking X polarization state as an example:
Step 1, the characteristic parameter of input coefficient is counted, divides following steps:
Step 11, range value s1 (m), the s2 (m) of input coefficient are calculated:
S1 (m)=(real (cxh(m)))2+(imag(cxh(m)))2；
S2 (m)=(real (cxv(m)))2+(imag(cxv(m)))2；
Wherein, [1,16] m ∈；Real is to take real, and imag is the imaginary part for taking plural number.
Step 12, maximum value m1, m2 of calculating input coefficient range value and corresponding serial number i1, i2, it is defeated for judging
Enter the main peak of coefficient energy, circular is as follows:
M1=max (s1 (m)), i1 are serial number of the m1 in s1 (m)；
M2=max (s2 (m)), i2 are serial number of the m2 in s2 (m).
Step 13, the energy of border coefficient is calculated, boundary energy sl1, sl2 and sr1, sr2 are calculated, for preventing coefficient steady
It is scheduled on boundary or goes to outside index boundaries, circular is as follows:
Sl1=s1 (1)+s1 (2), sl2=s2 (1)+s2 (2)；
Sr1=s1 (15)+s1 (16), sr2=s2 (15)+s2 (16)；
Step 14, calculate following formula s3 (m), s4 (m), s5 (m), s6 (m) and s5 (m), s6 (m) maximum value m3, m4 and
Corresponding serial number i3, i4.The peak value of coefficient energy is highlighted by related operation in conjunction with the two-way coefficient of X polarization state, specifically
Calculation method is as follows:
S3 (m)=m2 × s1 (m), s4 (m)=m1 × s2 (m)；
S5 (m)=| s3 (m)+s4 (m) |, s6 (m)=| s3 (m)-s4 (m) |；
M3=max (s5 (m)), i3 are serial number of the m3 in s5 (m)；
M4=max (s6 (m)), i4 are serial number of the m4 in s6 (m).
Step 15, the minimum value mvmin of m1, m2, m3, m4 and the maximum value imax and minimum value of i1, i2, i3, i4 are calculated
Imin, the peak value for judgement factor energy.
Step 16, distance d1, d2 is calculated, for the center of judgement factor and the relationship of tap coefficient center: d1=
(imax+imin)/2-8, d2=imax-imin.
Step 2, according to the characteristic of system PMD, the spy that the step 1 in preset time period static_time is previously mentioned is found out
The mean value for levying parameter, obtains the statistical property of adaptive equalizer coefficients.D1_ave by taking characteristic parameter d1 as an example, after being averaged
Calculation method is as follows:
It is other it is average after statistical parameter do not list one by one, be corresponding flat in preset time period static_time
Mean value.
Step 3, according to the statistical property of step 2, the judgement of coefficient off-centring is carried out, is adjusted in the way of Fig. 8 defeated
Enter coefficient, all " Y " indicates that condition meets in Fig. 8, and all " N " is unsatisfactory for for condition:
Condition 1:(i1≤3&&i2 >=14) | | (i2≤3&&i1 >=14), i.e., when i1_ave≤3 and i2_ave >=
14, alternatively, i2_ave≤3 and i1_ave >=14 when, do not adjust input coefficient.
Condition 2:(m1 > G1&&i1≤3) | | (m2 > G1&&i2≤3) work as m1_ave > G1 and i1_ave≤3, or
When person, m2_ave > G1 and i2_ave≤3, input coefficient is adjusted to the right.
Condition 3:(m1 > G1&&i1 >=14) | | (m2 > G1&&i 2 >=14), that is, work as m1_ave > G1 and i1_ave >=14,
Alternatively, m2_ave > G1 and i2_ave >=14 when, input coefficient is adjusted to the left.
Condition 4:(sl1>G2 | | sl2>G2) && (m1>G1&&i1>) of=4&&i1≤10 && (m2>G1&&i2>=4&&i2<
=10), that is, work as sl1_ave > G2 or sl2_ave > G2, and m1_ave > G1 and i1_ave >=4 and i1_ave≤10, and
M2_ave > G1 and when i2_ave >=4 and i2_ave≤11, input coefficient adjusts to the right.
Condition 5:(sr1>G2 | | sr2>G2) && (m1>G1&&i1>) of=7&&i1≤13 && (m2>G1&&i2>=7&&i2<
=13), that is, work as sr1_ave > G2 or sr2_ave > G2, and m1_ave > G1 and i1_ave >=7 and i1_ave≤13, and
M2_ave > G1 and when i2_ave >=7 and i2_ave≤13, input coefficient adjusts to the left.
Condition 6:m1>G2&&m2>G2&&mvmin>G3&& (d1>=2&&d2>=3&&imin>=4), even m1_ave<
G2 and m2_ave<G2, and mvmin_ave>G3 and d1_ave>=2 and d2_ave>=3 and imin_ave>=4 when, input system
Number adjusts to the left.
Condition 7:m1 > G2&&m2 > G2&&mvmin > G3&& (d1≤- 2&&d2 >=3&&imax >=13), even m1_ave
<G2 and m2_ave<G2, and mvmin_ave>G3 and d1_ave≤- 2 and d2_ave>=3 and imax_ave>=13 when, it is defeated
Enter coefficient to adjust to the right.
Be unsatisfactory for above-mentioned condition it is other under the conditions of, the adjustment without input coefficient.
Step 4, input coefficient adjustment is carried out according to the offset direction of step 3 coefficient, a point following situations is handled:
Step 41, when input coefficient adjusts to the left,
Step 42, when input coefficient adjusts to the right,
Step 5, according to the adjustment of step 3 and 4 pairs of input coefficients, FIR filter output data insert and deletes value.Definition
Adaptive equalization FIR output data length is N.Divide following steps:
Step 51, FIR filtering is carried out to 145 sampling point data, exports 130 sampling point data.
Step 52, FIR filtering output data insert and delete value processing, in the following several ways:
Step 521: when not being adjusted to input coefficient, this 130 sampling point data are removed in adaptive equalizer output
Two sampling point data of head；
Step 522: when input coefficient is adjusted to the right, adaptive equalizer exports all 130 sampling point data；
Step 523: when input coefficient is adjusted to the left, this 130 sampling point data are removed in adaptive equalizer output
Four sampling point data of head.
Step 53, the data after deleting value are inserted to step 52 and carries out butterfly buffer shaping, 128 data are fixed in output.
In another preferred embodiment, as shown in figure 9, on the basis of the embodiment of above-mentioned Fig. 2, above-mentioned steps S101
Can also include:
Step S1016, input coefficient needed for obtaining the adaptive equalizer calculate the feature ginseng of the input coefficient
Number, and obtain the characterization parameter of the polarizing coating dispersion of the adaptive equalizer；
Step S1017 calculates the input coefficient according to the baud rate of the characterization parameter and the adaptive equalizer
The distance of peak value；
Step S1018 monitors the input coefficient according to the distance of the input coefficient peak value and the characteristic parameter
Whether coefficient center shifts relative to tap coefficient center.
In the present embodiment, high-speed optical transmission system needs to monitor the operating status of system, the ginseng such as size including CD, DGD
Number.DGD (i.e. Differential Group Delay between two polarizing coatings) is the important parameter for characterizing PMD size, it is by adaptive
The input coefficient that blind equalization algorithm is found out carries out what certain mathematic(al) manipulation obtained.In the course of work of adaptive equalizer,
The adjustment of input coefficient can be more accurately carried out in conjunction with the characterization parameter DGD of PMD polarizing coating dispersion.
Input coefficient needed for the present embodiment obtains adaptive equalizer and calculating characteristic parameter are with above-mentioned Fig. 7's and Fig. 9
Embodiment is identical, and details are not described herein again.
The difference is that the present embodiment also needs to obtain the characterization parameter DGD of the polarizing coating dispersion of adaptive equalizer, root
The distance d_est of input coefficient peak value is calculated according to the baud rate of characterization parameter DGD and adaptive equalizer.
As shown in Figure 10, by taking the FIR tap number of adaptive equalizer is 17 as an example (m=17), the input system for needing to adjust
Several numbers are 1, and FIR output data length is 96, and by taking X polarization state as an example, all " Y " indicates that condition meets in Figure 10, institute
" N " having is unsatisfactory for for condition, according to above-mentioned characteristic parameter and the distance d_est of input coefficient peak value, judgement factor center
Whether deviate, and adjust input coefficient, comprising:
Condition 1:(i1≤3&&i2 >=15) | | (i2≤3&&i1 >=15), i.e., when i1≤3 and i2 >=15, alternatively,
I2≤3 and i1 >=15 when, do not adjust input coefficient.
Condition 2:(m1 > G1&&i1≤3) | | (m2 > G1&&i2≤3) work as m1 > G1 and i1≤3, alternatively, m2 > G1 and
When i2≤3, input coefficient is adjusted to the right.
Condition 3:(m1 > G1&&i1 >=15) | | (m2 > G1&&I2 >=15), that is, work as m1 > G1 and i1 >=15, alternatively, m2 >
G1 and i2 >=15 when, input coefficient is adjusted to the left.
Condition 4:(sl1>G2 | | sl2>G2) && (m1>G1&&i1>) of=4&&i1≤11 && (m2>G1&&i2>=4&&i2<
=11), that is, work as sl1 > G2 or sl2 > G2, and m1 > G1 and i1 >=4 and i1≤11, and m2 > G1 and i2 >=4 and i2≤
When 11, input coefficient is adjusted to the right.
Condition 5:(sr1>G2 | | sr2>G2) && (m1>G1&&i1>) of=7&&i1≤14 && (m2>G1&&i2>=7&&i2<
=14), that is, work as sr1 > G2 or sr2 > G2, and m1 > G1 and i1 >=7 and i1≤14, and m2 > G1 and i2 >=7 and i2≤
When 14, input coefficient adjusts to the left.
Condition 6:m1 > G2&&m2 > G2&&mvmin > G3&& (d1 >=2&&d2 >=3&&imin >=4), that is, work as m1 > G2 and
M2 > G2, and mvmin > G3 and d1 >=2 and d2 >=3 and imin >=4, and when d1 is matched with d_est, input coefficient is to the left
Adjustment.
Condition 7:m1 > G2&&m2 > G2&&mvmin > G3&& (d1≤- 2&&d2 >=3&&imax >=14), that is, work as m1 > G2
And m2 > G2, and mvmin > G3 and d1≤- 2 and d2 >=3 and imax >=14, and when d1 is matched with d_est, input coefficient
It adjusts to the right.
Be unsatisfactory for above-mentioned condition it is other under the conditions of, the adjustment without input coefficient.
The present invention also provides a kind of adjustment devices at adaptive equalizer coefficients center, as shown in figure 11, in an embodiment
In, the adjustment device at the adaptive equalizer coefficients center includes:
Monitoring modular 101, for obtaining needed for adaptive equalizer when detecting adaptive equalizer has data input
Input coefficient, whether the coefficient center for monitoring the input coefficient shifts relative to tap coefficient center；
The present embodiment hardware running environment is related to adaptive equalizer in high-speed optical transmission system, based on to adaptive equalization
Effective control of device input coefficient stablizes the coefficient center of input coefficient in tap coefficient center, improves system
Stability.
Specifically, the present embodiment adaptive equalization implement body is as shown in Figure 3 above.Wherein, by adaptive blind equalization algorithm
Input coefficient needed for (by taking CMA algorithm as an example) calculates FIR filter, is then filtered input data.Meanwhile it is right
The output data of FIR filter, which insert, deletes value processing, farthest to reduce the bit error rate.
Wherein, FIR filter can perhaps frequency domain realizes blind equalization algorithm or in time domain or frequency domain reality in time domain
It is existing.
To describe the present embodiment method, the FIR filter mathematic(al) representation in adaptive equalizer makees following agreement:
Wherein, M is the tap number of FIR filter.Input coefficient cxh(m)、cxv(m)、cyh(m) and cyv(m) by such as
The blind equalization algorithms such as CMA generate.cxh(m) refer to that input h is mapped to the coefficient in x-polarisation state, cxv(m) refer to that output v is mapped to
Coefficient in x-polarisation state, cyh(m) refer to that input h is mapped to the coefficient in y-polarisation state, cyv(m) it is inclined to refer to that output v is mapped to y
Coefficient in polarization state.
The technical solution that the present embodiment uses is consistent the processing mode of polarization state X or polarization state Y, between polarization state
Keep independent, in full text by taking polarization state X as an example.
In the present embodiment, the characteristic parameter of input coefficient is further calculated, characteristic parameter includes the range value of input coefficient
And its corresponding serial number, the number of the energy of boundary input coefficient and boundary input coefficient, the coefficient for characterizing input coefficient
Center is at a distance from the relationship of tap coefficient center etc..
In the present embodiment, by judging by characteristic parameter formation condition, so as to monitor input coefficient center
Whether shift.
Module 102 is adjusted, if being adjusted for shifting to the input coefficient；
It is right if the coefficient center of input coefficient shifts relative to tap coefficient center in the present embodiment
Input coefficient is adjusted, and specifically, can be adjusted according to offset direction to input coefficient or be adjusted to the right to the left, if not having
It shifts, input coefficient does not adjust.
In the present embodiment, it is assumed that the movement of samp sampling point need to be carried out to input coefficient.Divide following situations processing:
When input coefficient adjusts to the left,
Wherein, cxh'(m)、cxv' (m) be through input coefficient adjusted to the left.
When input coefficient adjusts to the right,
Wherein, cxh'(m)、cxv' (m) be through input coefficient adjusted to the right.
Processing module 103, for being filtered according to data of the input coefficient adjusted to input, to filtered defeated
Data, which insert, out deletes value processing.
In the present embodiment, due to being adjusted to input coefficient, the output data for inevitably resulting in FIR filter generates one
Fixed influence causes the generation of the bit error rate, and therefore, it is necessary to insert to filtered output data deleting value processing.
In the present embodiment, defining the FIR filter output data length in adaptive equalizer is N.Such as above-mentioned Fig. 4 institute
Show, processing module 103 insert to the output data of FIR filter and deletes value and include:
When carrying out adaptive equalization FIR filtering, group number is carried out according to [samp+ (M-1)+N] and is inputted, is i.e. present clock section
The FIR data N of bat adds (M-1) a data of previous timeticks tail portion, adds samp sampling point mobile data, then altogether
With FIR filtering is carried out, FIR filter exports (N+samp) a sampling point.
The output data of FIR filtering insert and deletes value processing, as shown in Figure 4 above, in the following several ways:
When not being adjusted to input coefficient, adaptive equalizer exports the rear N point data of (N+samp)；
When input coefficient is adjusted to the right, adaptive equalizer exports the rear N-samp point data of (N+samp)；
When input coefficient is adjusted to the left, adaptive equalizer exports (N+samp) point data.
Value is deleted treated that data carry out butterfly shaping, the data of output fixation N length to inserting.
In a preferred embodiment, as shown in figure 12, on the basis of the embodiment of above-mentioned Figure 11, above-mentioned monitoring modular
101 include:
First computing unit 1011 for input coefficient needed for obtaining the adaptive equalizer, and calculates described defeated
Enter the characteristic parameter of coefficient；
First monitoring unit 1012, for monitored according to the characteristic parameter coefficient center of the input coefficient relative to
Whether tap coefficient center shifts.
In the present embodiment, for X polarization state, the first computing unit 1011 calculates the characteristic parameter of input coefficient, comprising:
Calculate range value s1 (m), the s2 (m) of input coefficient:
S1 (m)=(real (cxh(m)))2+(imag(cxh(m)))2；
S2 (m)=(real (cxv(m)))2+(imag(cxv(m)))2；
Wherein, [1, M] m ∈, real are to take real, and imag is the imaginary part for taking plural number.
Maximum value m1, m2 of calculating input coefficient range value and corresponding serial number i1, i2, for judging input coefficient energy
The main peak of amount:
M1=max (s1 (m)), i1 are serial number of the m1 in s1 (m)；
M2=max (s2 (m)), i2 are serial number of the m2 in s2 (m).
The energy of border coefficient is calculated, parameter P1 and P2 are set, the leftmost number of input coefficient and coefficient are respectively represented
The number of rightmost, calculate respectively with cxh(m)、cxv(m) corresponding right boundary energy sl1, sl2 and sr1, sr2, for preventing
Only input coefficient is stablized on boundary or is gone to outside index boundaries:
With
Calculate following formula s3 (m), s4 (m), s5 (m), s6 (m) and s5 (m), maximum value m3, m4 of s6 (m) and corresponding
Serial number i3, i4 highlights the peak value of input coefficient energy by related operation in conjunction with the two-way coefficient of X polarization state:
S3 (m)=m2 × s1 (m), s4 (m)=m1 × s2 (m)；
S5 (m)=| s3 (m)+s4 (m) |, s6 (m)=| s3 (m)-s4 (m) |；
M3=max (s5 (m)), i3 are serial number of the m3 in s5 (m)；
M4=max (s6 (m)), i4 are serial number of the m4 in s6 (m).
The minimum value mvmin of m1, m2, m3, m4 and maximum value imax, the minimum value imin of i1, i2, i3, i4 are calculated, is used for
The peak value of judgement factor energy.
Calculate relationship gap d1, the d2 for being used for judgement factor center and tap coefficient center:
D2=imax-imin.
So far, calculation of characteristic parameters finishes.
Wherein, according to the above-mentioned characteristic parameter being calculated, the judgement of input coefficient off-centring is carried out, is obtained in coefficient
The offset direction that the heart shifts, being adjusted according to offset direction to input coefficient can refer to above-mentioned Fig. 6 and related text is retouched
Part is stated, details are not described herein again.
In an additional preferred embodiment, as shown in figure 13, on the basis of the embodiment of Figure 11, monitoring modular 101 is also
May include:
Second computing unit 1013, for input coefficient needed for obtaining the adaptive equalizer, and when calculating default
Between the input coefficient in section characteristic parameter；
First acquisition unit 1014, for obtaining the corresponding average value of the characteristic parameter；
Second monitoring unit 1015, for monitoring the coefficient center of the input coefficient according to the average value relative to pumping
Whether head coefficient center shifts.
In the present embodiment, since the coefficient center of adaptive equalizer input coefficient can be according to the spy of polarizing coating dispersion PMD
Property change, the present embodiment by the characterisitic parameter to input coefficient averagely come improve to coefficient center estimate it is accurate
Property, to reach the adjustment to input coefficient more accurately.
By taking the FIR tap number of adaptive equalizer is 16 as an example (m=16), the input coefficient number for needing to adjust is 2,
FIR output data length is 128, and the present embodiment is described by taking X polarization state as an example:
The characteristic parameter of input coefficient is counted, comprising:
Calculate range value s1 (m), the s2 (m) of input coefficient:
S1 (m)=(real (cxh(m)))2+(imag(cxh(m)))2；
S2 (m)=(real (cxv(m)))2+(imag(cxv(m)))2；
Wherein, [1,16] m ∈；Real is to take real, and imag is the imaginary part for taking plural number.
Maximum value m1, m2 of calculating input coefficient range value and corresponding serial number i1, i2, for judging input coefficient energy
The main peak of amount, circular are as follows:
M1=max (s1 (m)), i1 are serial number of the m1 in s1 (m)；
M2=max (s2 (m)), i2 are serial number of the m2 in s2 (m).
The energy of border coefficient is calculated, boundary energy sl1, sl2 and sr1, sr2 are calculated, for preventing coefficient stabilization on side
Boundary is gone to outside index boundaries, and circular is as follows:
Sl1=s1 (1)+s1 (2), sl2=s2 (1)+s2 (2)；
Sr1=s1 (15)+s1 (16), sr2=s2 (15)+s2 (16)；
Calculate following formula s3 (m), s4 (m), s5 (m), s6 (m) and s5 (m), maximum value m3, m4 of s6 (m) and corresponding
Serial number i3, i4.The peak value of coefficient energy, circular are highlighted by related operation in conjunction with the two-way coefficient of X polarization state
It is as follows:
S3 (m)=m2 × s1 (m), s4 (m)=m1 × s2 (m)；
S5 (m)=| s3 (m)+s4 (m) |, s6 (m)=| s3 (m)-s4 (m) |；
M3=max (s5 (m)), i3 are serial number of the m3 in s5 (m)；
M4=max (s6 (m)), i4 are serial number of the m4 in s6 (m).
The minimum value mvmin of m1, m2, m3, m4 and the maximum value imax and minimum value imin of i1, i2, i3, i4 are calculated, is used
In the peak value of judgement factor energy.
Distance d1, d2 is calculated, for the center of judgement factor and the relationship of tap coefficient center: d1=(imax+
Imin)/2-8, d2=imax-imin.
According to the characteristic of system PMD, the mean value of the characteristic parameter in preset time period static_time is found out, is obtained certainly
The statistical property of adaptive equalizer coefficient.By taking characteristic parameter d1 as an example, the d1_ave calculation method after being averaged is as follows:
It is other it is average after statistical parameter do not list one by one, be corresponding flat in preset time period static_time
Mean value.
According to above-mentioned statistical property, the judgement of coefficient off-centring is carried out, describes portion according to above-mentioned Fig. 8 and related text
Divide adjustment input coefficient, details are not described herein again.
In another preferred embodiment, as shown in figure 14, on the basis of the embodiment of Figure 11, monitoring modular 101 is also
May include:
Second acquisition unit 1016 calculates the input for input coefficient needed for obtaining the adaptive equalizer
The characteristic parameter of coefficient, and obtain the characterization parameter of the polarizing coating dispersion of the adaptive equalizer；
Third computing unit 1017, for calculating institute according to the baud rate of the characterization parameter and the adaptive equalizer
State the distance of input coefficient peak value；
Third monitoring unit 1018, for according to the distance of the input coefficient peak value and characteristic parameter monitoring
Whether the coefficient center of input coefficient shifts relative to tap coefficient center.
In the present embodiment, high-speed optical transmission system needs to monitor the operating status of system, the ginseng such as size including CD, DGD
Number.DGD (i.e. Differential Group Delay between two polarizing coatings) is the important parameter for characterizing PMD size, it is by adaptive
The input coefficient that blind equalization algorithm is found out carries out what certain mathematic(al) manipulation obtained.In the course of work of adaptive equalizer,
The adjustment of input coefficient can be more accurately carried out in conjunction with the characterization parameter DGD of PMD polarizing coating dispersion.
Input coefficient needed for the present embodiment obtains adaptive equalizer and calculating characteristic parameter and above-mentioned Figure 12 and Figure 13
Embodiment it is identical, details are not described herein again.
The difference is that the present embodiment also needs to obtain the characterization parameter DGD of the polarizing coating dispersion of adaptive equalizer, root
The distance d_est of input coefficient peak value is calculated according to the baud rate of characterization parameter DGD and adaptive equalizer.
In addition, by taking the FIR tap number of adaptive equalizer is 17 as an example (m=17), the input coefficient number for needing to adjust
Be 1, FIR output data length is 96, by taking X polarization state as an example, according to above-mentioned characteristic parameter and input coefficient peak value away from
From d_est, whether judgement factor center is deviated, and adjusts input coefficient, can refer to above-mentioned Figure 10 i.e. related text description section,
When adjusting input coefficient in condition 6 and condition 7, sentenced in conjunction with the distance d_est of the input coefficient peak value in the present embodiment
It is disconnected, then adjust again.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of method of adjustment at adaptive equalizer coefficients center is applied in high-speed optical transmission system, which is characterized in that institute
State the method for adjustment at adaptive equalizer coefficients center the following steps are included:
When detecting adaptive equalizer has data input, input coefficient needed for obtaining adaptive equalizer, described in monitoring
Whether the coefficient center of input coefficient shifts relative to tap coefficient center, wherein the input coefficient is mapped to two
The processing mode of different polarization states, described two different polarization states is consistent and described two different polarization states between keep only
It is vertical；
If shifting, the input coefficient is adjusted；
It is filtered according to data of the input coefficient adjusted to input, filtered output data insert and is deleted at value
Reason.
2. the method for adjustment at adaptive equalizer coefficients center as described in claim 1, which is characterized in that the acquisition is adaptive
Whether input coefficient needed for answering balanced device, the coefficient center for monitoring the input coefficient occur partially relative to tap coefficient center
The step of shifting includes:
Input coefficient needed for obtaining the adaptive equalizer, and calculate the characteristic parameter of the input coefficient；
Whether shifted according to the coefficient center that the characteristic parameter monitors the input coefficient relative to tap coefficient center.
3. the method for adjustment at adaptive equalizer coefficients center as described in claim 1, which is characterized in that the acquisition is adaptive
Whether input coefficient needed for answering balanced device, the coefficient center for monitoring the input coefficient occur partially relative to tap coefficient center
The step of shifting includes:
Input coefficient needed for obtaining the adaptive equalizer, and calculate the feature of the input coefficient in preset time period
Parameter；
Obtain the corresponding average value of the characteristic parameter；
Whether shifted according to the coefficient center that the average value monitors the input coefficient relative to tap coefficient center.
4. the method for adjustment at adaptive equalizer coefficients center as described in claim 1, which is characterized in that the acquisition is adaptive
Whether input coefficient needed for answering balanced device, the coefficient center for monitoring the input coefficient occur partially relative to tap coefficient center
The step of shifting includes:
Input coefficient needed for obtaining the adaptive equalizer calculates the characteristic parameter of the input coefficient, and described in acquisition
The characterization parameter of the polarizing coating dispersion of adaptive equalizer；
The distance of the input coefficient peak value is calculated according to the baud rate of the characterization parameter and the adaptive equalizer；
According to the distance of the input coefficient peak value and the characteristic parameter monitor the coefficient center of the input coefficient relative to
Whether tap coefficient center shifts.
5. such as the method for adjustment at the described in any item adaptive equalizer coefficients centers of Claims 1-4, which is characterized in that institute
The step of shifting if stating, being adjusted to the input coefficient include:
The offset direction that the coefficient center of the input coefficient shifts is obtained, is to the input according to the offset direction
Number is adjusted.
6. a kind of adjustment device at adaptive equalizer coefficients center, which is characterized in that the adjustment device includes:
Monitoring modular, for when detecting adaptive equalizer has data input, input needed for obtaining adaptive equalizer
Whether coefficient, the coefficient center for monitoring the input coefficient shift relative to tap coefficient center, wherein the input system
Number is mapped to two different polarization states, and the processing mode of described two different polarization states is consistent and described two different polarizations
It is kept between state independent；
Module is adjusted, if being adjusted for shifting to the input coefficient；
Processing module, for being filtered according to data of the input coefficient adjusted to input, to filtered output data
Insert and deletes value processing.
7. adjustment device as claimed in claim 6, which is characterized in that the monitoring modular includes:
First computing unit for input coefficient needed for obtaining the adaptive equalizer, and calculates the input coefficient
Characteristic parameter；
First monitoring unit, for monitoring the coefficient center of the input coefficient according to the characteristic parameter relative to tap coefficient
Whether center shifts.
8. adjustment device as claimed in claim 6, which is characterized in that the monitoring modular includes:
Second computing unit for input coefficient needed for obtaining the adaptive equalizer, and calculates in preset time period
The characteristic parameter of the input coefficient；
First acquisition unit, for obtaining the corresponding average value of the characteristic parameter；
Second monitoring unit, for monitoring the coefficient center of the input coefficient according to the average value relative in tap coefficient
Whether the heart shifts.
9. adjustment device as claimed in claim 6, which is characterized in that the monitoring modular includes:
Second acquisition unit calculates the spy of the input coefficient for input coefficient needed for obtaining the adaptive equalizer
Parameter is levied, and obtains the characterization parameter of the polarizing coating dispersion of the adaptive equalizer；
Third computing unit, for calculating the input system according to the baud rate of the characterization parameter and the adaptive equalizer
The distance of number peak value；
Third monitoring unit, for monitoring the input coefficient according to the distance of the input coefficient peak value and the characteristic parameter
Coefficient center whether shift relative to tap coefficient center.
10. such as the described in any item adjustment devices of claim 6 to 9, which is characterized in that the adjustment module is specifically used for obtaining
The input coefficient is adjusted according to the offset direction in the offset direction that the coefficient center of the input coefficient shifts
It is whole.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360122.1A CN105281792B (en) | 2014-07-25 | 2014-07-25 | The method of adjustment and device at adaptive equalizer coefficients center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360122.1A CN105281792B (en) | 2014-07-25 | 2014-07-25 | The method of adjustment and device at adaptive equalizer coefficients center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105281792A CN105281792A (en) | 2016-01-27 |
| CN105281792B true CN105281792B (en) | 2019-03-12 |
Family
ID=55150213
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410360122.1A Active CN105281792B (en) | 2014-07-25 | 2014-07-25 | The method of adjustment and device at adaptive equalizer coefficients center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105281792B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106330320B (en) * | 2015-06-24 | 2019-05-24 | 中兴通讯股份有限公司 | A kind of adaptive equalizer and its method for realizing adaptive equalization processing |
| CN107800655B (en) * | 2016-09-05 | 2020-06-16 | 深圳市中兴微电子技术有限公司 | Method and device for adjusting tap coefficient of adaptive equalizer |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1269639A (en) * | 1999-03-26 | 2000-10-11 | 松下电器产业株式会社 | Signal treatment apparatus |
| US8705977B1 (en) * | 2011-01-22 | 2014-04-22 | Viasat, Inc. | Equalizer tap correction and reseeding |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6985523B2 (en) * | 2001-01-10 | 2006-01-10 | Hughes Electronics Corp. | Method and system for adaptive equalization for receivers in a wide-band satellite communications system |
-
2014
- 2014-07-25 CN CN201410360122.1A patent/CN105281792B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1269639A (en) * | 1999-03-26 | 2000-10-11 | 松下电器产业株式会社 | Signal treatment apparatus |
| US8705977B1 (en) * | 2011-01-22 | 2014-04-22 | Viasat, Inc. | Equalizer tap correction and reseeding |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105281792A (en) | 2016-01-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7187873B2 (en) | Compensation of polarization mode dispersion in single mode fiber for maximum-likelihood sequence estimation | |
| JP2015122632A (en) | Optical communication receiver | |
| US8611762B2 (en) | System and method for frequency-domain chromatic dispersion and polarization mode dispersion compensation with time-domain channel estimation | |
| US9722769B2 (en) | Equalizer | |
| CN107493247B (en) | Self-adaptive equalization method and device and equalizer | |
| TW201406085A (en) | Methods and apparatus for adapting transmitter equalization coefficients based on receiver gain adaptation | |
| CN105281792B (en) | The method of adjustment and device at adaptive equalizer coefficients center | |
| WO2016206521A1 (en) | Self-adaptive equalizer and method of implementing adaptive equalization | |
| US9917712B2 (en) | Frequency domain equalizer for optical communications | |
| CN105262707B (en) | Method and device for self-adaptive equalization of high-speed optical transmission system | |
| WO2017077802A1 (en) | Wavelength dispersion estimation circuit, optical receiver, and wavelength dispersion estimation method | |
| JP6378654B2 (en) | Spatial multiplexed optical transmission system and spatial multiplexed optical transmission method | |
| US9237046B2 (en) | Receiver circuit | |
| US7848401B2 (en) | Method and system for static data-pattern compensated adaptive equalizer control | |
| WO2016101541A1 (en) | Device, method, and computer storage medium for clock recovery and equalization | |
| US10749720B1 (en) | Receiver adaptation using stochastic gradient hill climbing with genetic mutation | |
| CN108667521B (en) | Method and device for adaptive equalization of optical transmission system | |
| US20170286250A1 (en) | Signal Compensation Method and Device | |
| US7848471B2 (en) | Method and system for on-line data-pattern compensated adaptive equalizer control | |
| US11381431B2 (en) | Receiver and transmitter adaptation using stochastic gradient hill climbing with genetic mutation | |
| Chaudhary et al. | Adaptive Filters Design and Analysis using Least Square and Least Pth Norm | |
| US10461968B2 (en) | Adjustment method and device for equalizer coefficient, and computer storage medium | |
| US20110211847A1 (en) | Optical receiver and clock generation method | |
| CN107800655B (en) | Method and device for adjusting tap coefficient of adaptive equalizer | |
| Kaur et al. | Simulation and comparative analysis of SS-LMS & RLS algorithms for electronic dispersion compensation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |