CN103731392A - Regional separation nonlinear phase noise recovery method - Google Patents

Abstract

The invention discloses a regional separation nonlinear phase noise recovery method. A single channel nonlinear schrodinger equation is utilized to deduce that the nonlinear phase noise of different amplitudes of symbols forms into a certain proportion under non-constant-modulus data modulation, the same phase deflection treatment is utilized during phase recovery of the different amplitudes of symbols in the former 16 QAM systems, and accordingly a certain error is produced during nonlinear phase noise recovery of the systems. According to the regional separation nonlinear phase noise recovery method, the complexity of the phase recovery algorithm in the 16 QAM systems is reduced and the accuracy of the nonlinear noise recovery is improved; meanwhile the nonlinear phase recovery method can be applied to a Nyquist technology adopted optical transmission system; the practicability is strong, the structure is simple, the cost is low, and the signal processing requirement is low.

Description

A kind of subregional nonlinear phase noise restoration methods

Technical field

The present invention relates to a kind of subregional nonlinear phase noise restoration methods, belong to the category of optical fiber telecommunications system.

Background technology

With developing rapidly for the communication technology, the technology increased popularity such as HD video and multimedia application, the data volume for interconnecting transfers on network recent years presents the trend being skyrocketed through.Cisco System Co. estimates that before 2012, the flow of Global Internet will be added to monthly 1,000,000,000 GB, doubles and has a surplus than current flow, and the popularity degree of many Online Video websites can also further increase.In " high definition, three-dimensional, user's creation content（UGC）" driving under, the generation of magnanimity information has triggered the arrival of " digital mighty torrent ".From now on, the mankind will produce the digital content for the byte that ended more than 1000 every year.Under the impact of digital mighty torrent, the flow of telecommunications backbone network increases the speed every year with 50%~80%, and network traffics will show ten times of even hundred times of growths.The growth of network traffics needs the raising of the network bandwidth, and why the network bandwidth can constantly raise speed, and is the development for having benefited from supporting the optical communication technique of fiber optic network.Therefore, ultrahigh speed, vast capacity have turned into the main target that present information transmission is pursued.

It is one of the following raising availability of frequency spectrum, the main method for increasing optical fiber telecommunications system transmission capacity that higher order modulation formats and coherent reception technology, which are combined,.Current coherent light communication modulation is used as modulation format using QPSK (QPSK) mostly, in order to improve the transmission capacity of system, the modulation format of higher order can be used, and 16QAM has been increasingly becoming the important modulation format used in 100Gbit/s and above speed coherent optical fiber transmission system.One 16QAM symbol carries 4 bit informations, and a QPSK symbol only carries 2 bit informations, therefore 16QAM has the spectrum efficiency higher than QPSK.In coherent communication system, with the increase of transmission power and transmission range, phase recuperation technique must be used by causing phase noise to turn into the key factor of limitation systematic function by Kerr effect in optical fiber link, thus obtaining correct information in receiving terminal.

There is some 16QAM Phase Retrieve Algorithm at present, method in document 1 is preceding feedback method, in the method, to the phase estimation of each symbol independent of the phase estimation result to other symbols, so can be with parallelization, multiple symbols can be done with phase estimation with multiple digital signal processing units simultaneously, this parallelization can substantially reduce the requirement to DSP processing speeds.Although method in document 2 is feedforward, but it is based on pre- judgement (pre-decision), phase estimation dependence and the phase estimation result of former symbol to current sign, so can not parallelization.The method introduced in document 3 is blind phase search algorithm（BPS）, although BPS algorithms effectively can be compensated and can be flexibly applied in the QAM modulation of higher order to 16QAM signal phase noises, but it significantly has the disadvantage that computation complexity is higher in the presence of one.

The method to document 1 in the prior art is simply introduced below.

Fig. 1, which is shown preferably, does not have noisy 16QAM constellations.Fig. 2, which is shown, to be done biquadratic to preferable 16QAM constellation points and keeps the constant obtained constellation point of amplitude.Fig. 3 shows the handling process of the phase recovery device using document 1.

As shown in figure 1,16QAM symbols have three kinds of amplitudes, three groups can be divided into according to the difference of amplitude.On inner ring is the 1st group of point, with minimum amplitude R₁；On the circle of centre one is the 2nd group of point, with medium amplitude R₂;On outer ring is the 3rd group of point, with maximum amplitude R₃(R₁<R₂<R₃).（" point " refers to constellation point herein, is expression of the symbol in I-Q planes.Based on context, the implication of " point " mentioned in implementation of the present invention can be equal to " symbol "）.1st, the argument of 3 groups of points is

（K=0,1,2,3）, the argument of the 2nd group of point is

After 4 powers are carried out to symbol, the 1st, the argument of 3 groups of points be π, in transverse axis negative direction, and the amplitude of the 2nd group of point has two possible values（As shown in Figure 2）, they are with the angle of width axle negative direction

As shown in figure 3, in the existing method of document 1, for each symbol in symbolic blocks, 4 powers are carried out first（Step 301）, then the symbol is grouped according to amplitude in step 302（It is divided into three groups）.Then in the phase estimation of step 303, merely with the 1st group and the 3rd group of symbol.This phase estimation algorithm, is recovered for every kind of symbol using identical phase noise.But in fact, the nonlinear phase noise of the symbol of different amplitudes is different, so can not accurately estimate phase noise using aforesaid way, particularly when transmission power is more long than larger and transmission range, this inaccuracy is more obvious.

The present inventor has found in the process of research invention：In a symbolic blocks, there is certain relation in the phase noise between each group, the present invention utilizes this relational expression, in single-carrier system, using a kind of subregional nonlinear phase noise restoration methods, the speed of Phase Retrieve Algorithm can be effectively improved and accurately determined.

Document 1:M.Seimetz,“Performance of coherent optical square16QAM systems based on IQ-transmitters and homodyne receivers with digital phase estimation,”OFC2006,NWA4.

Document 2:H.Louchet,K.Kuzmin,A.Richter,“Improved DSP algorithms for coherent16-QAM transmission,”Proc.ECOC'08(2008).

Document 3:D.-S.Ly-Gagnon,“Coherent detection of optical quadrature phase-shift keying signals with carrier phase estimation,”J.Lightw.Technol.,vol.24,no.1,2006.

The content of the invention

The implementation of the present invention is in view of the above mentioned problem of prior art is made, and one or more problems for eliminating or alleviating prior art provide at a kind of intentional selection.

Below in conjunction with the accompanying drawing in the present invention, the present invention is clearly and completely described using the subregional phase noise restoration methods of nonlinear theory.Obviously, described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained on the premise of creative work is not paid belongs to the scope of protection of the invention.

If system is single channel, SPM influences are only considered, then being known that by non-linear Schrodinger equation

$\frac{\&PartialD;U}{\&PartialD;z}=-\frac{j}{L_{\mathrm{NL}}}{e}^{-\mathrm{az}}{\left|U\right|}^{2}U---\left(1\right)$

Wherein,

$L_{\mathrm{NL}}=\frac{1}{\mathrm{\&gamma;}{P}_0}---\left(2\right)$

It can be obtained by solving equation：

U (z, T)=U (0, T) exp (- j Φ_NL(z,T))        （3）

And

${\mathrm{\&Phi;}}_{\mathrm{NL}}(z,T)={\left|U(0,T)\right|}^2\&CenterDot;\frac{z_{\mathrm{eff}}}{L_{\mathrm{NL}}}---\left(4\right)$

By formula（4）It is known that in single channel, the nonlinear noise of system is directly proportional with power.

This is also the theoretical foundation of our present invention, one relational expression can be deduced by existing theory, three regions are mainly divided into according to radius for 16QAM single-channel systems, as described above, we can see that the mould of the symbol of same area is identical, this is reflected on the power of this symbol in above-mentioned formula, it is possible thereby to which the nonlinear noise deflection angle for being inferred to these three regions is：(1²+1²):(1²+3²):(3²+3²)=1:5:9.This ratio is the foundation of the nonlinear noise phase recovery under single-carrier system 16QAM modulation formats, theoretical foundation of the invention.

The principle of the same present invention can also be widely suitable for mQAM modulation formats, such as 32QAM can be divided into five regions altogether, and nonlinear noise ratio is

(1²+1²):(1²+3²):(3²+3²):(1²+5²):(3²+5²)=1:5:9:13:17

Wherein it can be used to carry out nonlinear phase noise estimation in first region and trizonal symbol.

Lower mask body is analyzed under 16QAM modulation formats this theory.

Fig. 4 is the principle flow chart of the invention based on subregional nonlinear phase noise restoration methods in single-carrier system.

Assuming that the data of receiving terminal are

Rx=[N₁,N₂,N₃,N₄......N_m]         （5）

First carrying out biquadratic is

Rx⁴=[N₁ ⁴,N₂ ⁴,N₃ ⁴,N₄ ⁴......N_m ⁴]         （6）

Then data are divided into three regions according to amplitude, it is assumed that this m data, first area is a, Two Areas is b, and the 3rd region is c.

The average deflection angle in so first region is

$\mathrm{angle}\_R_1=\frac{1}{4}\arg \left(\underset{n=1}{\overset{a}{\mathrm{\&Sigma;}}}{N_{\mathrm{na}}}^4\right)---\left(7\right)$

Wherein, N_naFor first group of symbol

The trizonal average deflection angle is

$\mathrm{angle}\_R_3=\frac{1}{4}\arg \left(\underset{n=1}{\overset{c}{\mathrm{\&Sigma;}}}{N_{\mathrm{nc}}}^4\right)---\left(8\right)$

Wherein, N_ncFor the 3rd group of symbol

Because it is understood that there is numerical relation in first region with nonlinear phase noise between the 3rd region, we can carry out more detailed estimation using this numerical value, make whole estimation more accurate.

Assuming that the data of transmitting terminal are random equal-probability distributions, then it is 1/4 that first region and the 3rd region, which respectively account for symbol quantity ratio, and symbol quantity ratio shared by Two Areas is 1/2.

So we can show that the nonlinear noise deflection in first region can be processed as

$\mathrm{angle}\_{\tilde{R}}_1=\frac{1}{2}(\mathrm{angle}\_R_1+\frac{1}{9}\mathrm{angle}\_R_3)---\left(9\right)$

Same trizonal nonlinear noise deflection can be processed as

$\mathrm{angle}\_{\tilde{R}}_3=\frac{1}{2}(9*\mathrm{angle}\_R_1+\mathrm{angle}\_R_3)---\left(10\right)$

Because the deflection angle of Two Areas is theoretically equal to first and the trizonal and be averaged.

So trizonal average deflection angle is

$\mathrm{angle}\_R_2=\frac{\mathrm{angle}\_{\tilde{R}}_1+\mathrm{angle}\_{\tilde{R}}_3}{2}---\left(11\right)$

Brief description of the drawings

In order to illustrate more clearly of technical method proposed by the invention, the accompanying drawing to be used needed for the embodiment of the present invention will be briefly described below, obviously, drawings in the following description are only some embodiments of the present invention, for those of ordinary skill in the art, on the premise of not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.

Fig. 1, which is shown preferably, does not have noisy 16QAM constellations.

Fig. 2, which is shown, preferable 16QAM constellation points are done the holding constant obtained constellation point of amplitude after 4 powers.

Fig. 3 has shown out the flow chart of the phase recovery method of prior art.

Fig. 4 shows a kind of flow chart of the phase recovery method of implementation according to the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is also only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not made belongs to the scope of protection of the invention.

Referring to Fig. 4, Fig. 4 is the system flow schematic diagram that the present invention is implemented.

Step 401：As shown in figure 4, the symbol of receiving is carried out into identical decile, length is m, and for each symbol in symbolic blocks, 4 powers are carried out to the symbol using 4 power devices.

Step 402：The symbol is classified according to the amplitude of the symbol after 4 powers, the less symbol of amplitude (first group of point) is divided into the first kind and the larger symbol of amplitude (the 3rd group of point) is divided into the 3rd class, amplitude is in into middle symbol (second group of point) is divided into Equations of The Second Kind.

If H₁=(R₁+R₂)⁴/ 16, H₂=(R₂+R₃)⁴/16        （12）

Wherein R₁、R₂、R₃For the amplitude of preferable 16QAM constellation points.

The amplitude of the symbol after 4 powers and reservation threshold are compared first：If receiving the amplitude of signal in H₁And H₂Between, second group of point is judged to, Equations of The Second Kind is divided into；If amplitude is more than H₂, the 3rd group of point is judged to, is divided into the 3rd class；If amplitude is less than H₁, first group is judged to, the first kind is divided into.

Step 403：After each symbol in symbolic blocks all calculates 4 powers and has carried out classification, first group and the 3rd group of phase deflection is calculated.

Assuming that this m data, first area is a, Two Areas is b, and the 3rd region is c.Due to carrying out 4 power processing to symbol before, thus first group with the angle divided by 4 that the 3rd group of phase deflection angle is after the symbol biquadratic each organized is added.

The average deflection angle in so first region is

$\mathrm{angle}\_R_1=\frac{1}{4}\arg \left(\underset{n=1}{\overset{a}{\mathrm{\&Sigma;}}}{N_{\mathrm{na}}}^4\right)$

Wherein, N_naFor first group of symbol

The trizonal average deflection angle is

$\mathrm{angle}\_R_3=\frac{1}{4}\arg \left(\underset{n=1}{\overset{c}{\mathrm{\&Sigma;}}}{N_{\mathrm{nc}}}^4\right)---\left(14\right)$

Wherein, N_ncFor the 3rd group of symbol

Assuming that the data of transmitting are random equal-probability distributions, then it is 1/4 that first region and the 3rd region, which respectively account for symbol quantity ratio, and symbol quantity ratio shared by Two Areas is 1/2.

So we can show that the nonlinear noise deflection in first region can be processed as

$\mathrm{angle}\_{\tilde{R}}_1=\frac{1}{2}(\mathrm{angle}\_R_1+\frac{1}{9}\mathrm{angle}\_R_3)---\left(15\right)$

Same trizonal nonlinear noise deflection can be processed as

$\mathrm{angle}\_{\tilde{R}}_3=\frac{1}{2}(9*\mathrm{angle}\_R_1+\mathrm{angle}\_R_3)---\left(16\right)$

Step 404：Second group of phase deflection angle is asked for using the deflection angle required by first group and the 3rd group.

Because the deflection angle of Two Areas is theoretically equal to first and the trizonal and be averaged.

So trizonal average deflection angle is

$\mathrm{angle}\_R_2=\frac{\mathrm{angle}\_{\tilde{R}}_1+\mathrm{angle}\_{\tilde{R}}_3}{2}---\left(17\right)$

Step 405：Phase recovery is carried out using above-mentioned required phase noise.

Claims (10)

1. a kind of subregional nonlinear phase noise restoration methods, it is characterised in that： 

Derive that non-permanent modulus is proportional according to the symbol nonlinear phase noise of different amplitudes under modulation using single channel non-linear Schrodinger equation, and handled when all about 16QAM systems are recovered for the symbol phase of different amplitudes before using the deflection of identical phase, this make it that the nonlinear phase noise of system recovers have certain error, the present invention is compared with conventional method, both the degree of accuracy of nonlinear phase reply can have been improved, algorithm complex is reduced again. 

2. a kind of subregional nonlinear phase noise restoration methods according to claim 1, it is characterised in that： 

It is segmented into three parts：First, 16QAM symbols carry out biquadratic processing and carry out classified part according to amplitude information；2nd, nonlinear phase noise part is calculated using first region and the 3rd region；3rd, using the nonlinear noise of above-mentioned calculating, according to existing relational expression, the nonlinear phase noise part of Two Areas is calculated；4th, using first region and trizonal relational expression, part is modified to the nonlinear phase noise of one's respective area. 

3. a kind of subregional nonlinear phase noise restoration methods according to claim 1-2, are applicable to the light single-carrier system of different modulating form, it is characterised in that： 

The single-carrier system of 16QAM modulation formats is applicable not only to, and suitable for the single-carrier system of the various modulation formats of mQAM. 

4. a kind of subregional nonlinear phase noise restoration methods according to claim 1-3, biquadratic processing is carried out to symbol and is classified according to amplitude information, it is characterised in that： 

Biquadratic processing is carried out to the symbol of reception, and is made decisions according to the amplitude after symbol biquadratic and is divided into different regions. 

5. a kind of subregional nonlinear phase noise restoration methods according to claim 1-3, the nonlinear phase noise estimation of second area, it is characterised in that： 

This mainly uses the estimation noise in existing region, by the ratio for the nonlinear phase noise of different radii symbol asked before, asks for the nonlinear phase noise of second area. 

6. a kind of subregional nonlinear phase noise restoration methods according to claim 1-3, using first region and trizonal relational expression, part is modified to the nonlinear phase noise of one's respective area, it is characterised in that： 

Phase estimation algorithm in the past, recovered for each region using identical phase, and the present invention is recovered for each region using different phases, the nonlinear phase noise for understanding first area and the 3rd region by us has relation, and the probability distribution of their symbols is also aware that, first area and the 3rd region nonlinear phase noise can be modified, revised nonlinear phase noise is more accurate by these numerical relations. 

7. a kind of subregional nonlinear phase noise restoration methods according to claim 1-6, it is main to pass through the existing theoretical probability distribution situation to the estimation of nonlinear phase noise, the amplitude of symbol and every kind of symbol, carry out comprehensive estimation nonlinear phase noise, so as to which the error phase used when recovering to the nonlinear phase in each region is also different, so that the phase information of recovery is more accurate. 

8. a kind of subregional nonlinear phase noise restoration methods according to claim 1-7, can be used together with channel coding with combined signal source coding, improve the precision of channel estimation, realized while may map to specific hardware. 

9. a kind of subregional nonlinear phase noise restoration methods according to claim 1-8, suitable for the single-carrier system under 16QAM, 32QAM, 64QAM etc. different mQAM modulation formats, while can be used together with other phase estimation technologies of single-carrier system. 

10. a kind of subregional nonlinear phase noise restoration methods according to claim 1-9, it is used to not only single-carrier system, and going deep into by theoretical research, also it can be suitably used for wdm system, main method is to use the probability distribution situation by the theoretical estimation of nonlinear phase noise, the amplitude of symbol and every kind of symbol, comprehensive estimation nonlinear phase noise is carried out, this method is goed deep into DSP technologies, and the degree of accuracy of phase recovery can also be improved. 

Images