CN103746950A - Frequency domain dispersion search method based on training sequences - Google Patents

Frequency domain dispersion search method based on training sequences Download PDF

Info

Publication number
CN103746950A
CN103746950A CN201410006652.6A CN201410006652A CN103746950A CN 103746950 A CN103746950 A CN 103746950A CN 201410006652 A CN201410006652 A CN 201410006652A CN 103746950 A CN103746950 A CN 103746950A
Authority
CN
China
Prior art keywords
dispersion
frequency domain
search
modulation symbol
cost function
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.)
Granted
Application number
CN201410006652.6A
Other languages
Chinese (zh)
Other versions
CN103746950B (en
Inventor
於涛
许渤
邱昆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201410006652.6A priority Critical patent/CN103746950B/en
Priority to PCT/CN2014/072121 priority patent/WO2015103805A1/en
Publication of CN103746950A publication Critical patent/CN103746950A/en
Application granted granted Critical
Publication of CN103746950B publication Critical patent/CN103746950B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0224—Channel estimation using sounding signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/0222—Estimation of channel variability, e.g. coherence bandwidth, coherence time, fading frequency

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Optical Communication System (AREA)

Abstract

The invention discloses a frequency domain dispersion search method based on training sequences. The frequency domain dispersion search method comprises the following steps: performing frequency domain dispersion search on modulation symbol sequences obtained through modulating the training sequences; performing frequency domain maximum cost function-based dispersion search on received continuous sampling signals by a receiving end by adopting shift factors OmegaCT, wherein OmegaCT is equal to NFFT/2<K+2>, and K is an integer not smaller than 0; meanwhile, correspondingly performing the following preprocessing on modulation symbol sequences of a sending end: enabling 2<K> modulation symbol sequences for one group, setting all modulation symbols to be 0 every two groups, and keeping the rest modulation symbols constant, and thus, not only is the problem that amplitude of a frequency domain maximum cost function is changed along with the ADC sampling time changes solved but also the dispersion search can be performed by adopting a larger dispersion search step, and therefore, the search time is reduced, the algorithm complexity is lowered, and system resources are saved.

Description

Frequency domain dispersion searching method based on training sequence
Technical field
The invention belongs to frequency domain dispersion search technique field, more specifically say, relate to a kind of frequency domain dispersion searching method based on training sequence.
Background technology
Development along with the coherent fiber communication technology, the relevant detection receiver that includes digital equalising technology can compensate all linear channel losses or low nonlinearity fiber channel, this just makes can be at receiving terminal to chromatic dispersion (CD in transmitting procedure, Chromatic Dispersion) carry out digital equalising and needn't pass through dispersion compensating fiber (DCF, Dispersion Compensating Fiber) and compensate in light territory.In conventional digital equalising method, there are two kinds of modes of time domain equalization and frequency domain equalization.Because frequency domain equalization has low-down computation complexity and good degree of parallelism, optical fiber dispersion finite impulse response (FIR) (FIR with time domain, Finite Impulse Response) equalizer and self adaptation lowest mean square (LMS, Least Mean Square) equalizer is compared, and frequency-domain equalizer becomes the chromatic dispersion balancing technique that industry is mainly paid close attention to.
In the method based on frequency domain, first by receiving terminal, the time-domain signal after over-sampling utilizes fast Fourier transform (FFT) to transform to frequency domain and is multiplied by frequency domain dispersion compensation function again, this penalty function can be regarded as the inverse function of the Dispersion Function of fibre channel transmission, finally the later frequency-region signal of compensation is transformed to symbol recovery by inverse fast fourier transform (IFFT) and becomes continuous data flow.The shortcoming of the method is on the basis of knowing in advance channel dispersion size, to utilize frequency domain dispersion compensation function.Yet, because the dispersion values of transmission link under dynamic optical network environment is variable, need first the dispersion size in channel to be estimated.
A kind of conventional dispersive estimates method is frequency domain dispersion searching algorithm at present, this algorithm adopts a kind of mode of optimum Match search by the parameter of one group of Dispersion filter of observation, and this parameter is according to the cost function that calculates every group of balanced later signal, to make differentiating realizing.
Fig. 1 is the coherent fiber communication control processor structural representation based on frequency domain dispersion search.As shown in Figure 1, the signal that the polarisation-multiplexed signal receiving is divided into two polarization directions through polarization beam apparatus carries out independently coherent demodulation.First the 4 road signals of telecommunication after coherent demodulation need carry out the AD sampling of dual-rate and quantize, then the later signal of sampling is transformed to frequency domain by FFT and be multiplied by again frequency domain dispersion compensation function, frequency domain dispersion compensation function is divided into some groups to meet the needs to different dispersion step-size in searchs according to the scope of the channel dispersion of transmission, according to the later frequency-region signal of compensation, do dispersion search, rule out the dispersion range of transmission.Finally, select that best group frequency-region signal of compensation effect to transform to time domain by IFFT and do subsequent treatment.
Dispersion searching algorithm based on frequency domain maximum cost function is that the method for estimating is searched in a kind of conventional dispersion in channel.The key of this dispersion searching algorithm is that the sampled signal based on receiving is calculated frequency domain maximum cost function to some probable values of channel dispersion, then chooses channel dispersion value corresponding to maximum frequency domain cost function value as the result of dispersion search and estimation.Fig. 2 is the dispersion searching algorithm schematic diagram based on frequency domain maximum cost function.The dispersion searching algorithm based on frequency domain maximum cost function as shown in Figure 2, this algorithm carries out Combined Treatment by the signal on X, Y two-way polarization direction and carries out dispersive estimates, and the signal processing method on single polarization direction is:
First, by the continuous sampling signal r[k receiving] by FFT conversion points N fFTbe divided into N group for r p[k], p=1,2 ..., N, utilizing FFT that each group signal is transformed to frequency domain generating digital spectrum is R p[n], n=-N fFT/ 2 ..., N fFT/ 2-1.
Secondly, according to the channel dispersion scope of transmission, M possible transmission channel dispersion values defined to frequency domain dispersion compensation function H CD , i [ n ] = exp [ - j 2 &beta; 2 L i ( 2 &pi;n&Delta;f ) 2 ] , i = 1,2 , . . . , M 。L wherein i=L min, L min+ Δ L, L min+ 2 Δ L ..., L maxsome transmission range values of selecting while representing dispersion search, step delta L represents the resolution of dispersion search, β 2represent 2nd order chromatic dispersion effect coefficient, Δ f represents frequency domain sample interval delta f=R s/ N fFT, R srepresent sampling rate.
Then, utilize this frequency domain dispersion compensation function can obtain corresponding to the signal group R after the frequency domain compensation of M possible transmission channel dispersion values p,i[n]=H cD, i[n] R p[n], can define frequency domain cost function according to the signal after this compensation C p , i [ &Omega; CT ] = &Sigma; n = 1 N FFT ( R p , i [ n ] &CenterDot; R p , i * [ n + &Omega; CT ] ) , Ω wherein cTrepresent translocation factor, subscript * represents complex conjugate.Frequency domain cost function is relevant to clock recovery, and it is to utilize phase place and the frequency information of the clock frequency obtaining in the auto-correlation function of the complex signal from frequency domain compensation to represent.
Finally, defining the frequency domain maximum cost function that each possible transmission channel dispersion values is corresponding is:
Figure BDA0000454094550000023
wherein | C p,i[Ω cT] | represent C p,i[Ω cT] amplitude.In order to reduce the impact of polarization effect on search, in the dispersion searching algorithm based on frequency domain maximum cost function as shown in Figure 2, be that the maximum cost of frequency domain corresponding on X, Y two-way polarization direction is averaged, J FD [ i ] = 1 2 ( &Sigma; p = 1 N | C x , p , i [ &Omega; CT ] | + &Sigma; p = 1 N | C y , p , i [ &Omega; CT ] | ) . In M frequency domain maximum cost, transmission channel dispersion values corresponding to maximum is the channel dispersion that estimation obtains.
Yet, in the above-mentioned dispersion searching algorithm based on frequency domain maximum cost function, there are two problems: first, under the most frequently used dual-rate sampling, adopt Ω cT=N fFT/ 2, however in this case, the amplitude of frequency domain maximum cost function can change along with the variation in the ADC sampled point moment, causes very large performance difference, even causes dispersion search module normally to work.Secondly, in dispersion searching algorithm, the number of the dispersion values that need to attempt has determined the complexity of dispersion searching algorithm.In traditional dispersion searching algorithm, need to use very little dispersion step-size in search Δ L guarantee can obtain correct dispersive estimates value, in actual processing procedure, need many group frequency domain dispersion compensation functions to realize the search to desired dispersion range, this is to waste very much resource on hardware is realized.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, a kind of frequency domain dispersion searching method based on training sequence is provided, adjusted Ω cTthe modulation symbol of translocation factor and training sequence, both can solve the impact of ADC sampling instant, allows again to use larger dispersion step-size in search, reduces the complexity of dispersion search.
For achieving the above object, the present invention is based on the frequency domain dispersion searching method of training sequence, comprise the following steps:
S1: at transmitting terminal, the modulation symbol sequence that training sequence is obtained through ovennodulation carries out preliminary treatment, and preprocess method is: by modulation symbol sequence with 2 kindividual is one group, and K is more than or equal to 0 integer, at interval of 1 group, modulation symbol is set to 0 entirely, and all the other remain unchanged;
S2: digital coherent light receiver carries out to the received signal coherent demodulation and obtains the 4 road signals of telecommunication, and the Dui Mei road signal of telecommunication carries out respectively dual-rate sampling and quantification;
S3: the continuous sampling signal in step S2 is carried out to the dispersion range that dispersion based on frequency domain maximum cost function search obtains estimating channel, wherein translocation factor Ω cT=N fFT/ 2 k+2.
The present invention is based on the frequency domain dispersion searching method of training sequence, the modulation symbol sequence that adopts training sequence to obtain through ovennodulation carries out frequency domain dispersion search, and receiving terminal adopts different translocation factor Ω when the continuous sampling signal receiving is carried out to the dispersion search based on frequency domain maximum cost function cT=N fFT/ 2 k+2, K is more than or equal to 0 integer, simultaneously corresponding the modulation symbol sequence of transmitting terminal is carried out to following preliminary treatment: by modulation symbol sequence with 2 kindividual is one group, at interval of 1 group, modulation symbol is set to 0 entirely, all the other remain unchanged, the amplitude that had so not only solved frequency domain maximum cost function changes constantly with ADC sampled point and the problem that changes but also can allow to adopt larger dispersion step-size in search to carry out dispersion search, thereby minimizing searching times, reduce algorithm complex, save system resource.
Accompanying drawing explanation
The coherent fiber communication control processor structural representation of Fig. 1 based on frequency domain dispersion search;
Fig. 2 is the dispersion searching algorithm schematic diagram based on frequency domain maximum cost function;
Fig. 3 is the preliminary treatment result schematic diagram of modulation symbol sequence;
Fig. 4 is four kinds of dispersion step-size in search comparison diagrams that translocation factor is corresponding;
Fig. 5 is frequency domain maximum cost function changes in amplitude comparison diagram under different ADC sampling instants;
Fig. 6 is the simulation result figure to 4 channel dispersion in 200 kilometer range;
Fig. 7 is 5000 transmission channel dispersions and the distribution map of estimating channel dispersion;
Fig. 8 is the simulation result figure to 4 channel dispersion in 500 kilometer range;
Fig. 9 is the simulation result figure of 4 channel dispersion in 500 kilometer range after adjusting.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described, so that those skilled in the art understands the present invention better.Requiring particular attention is that, in the following description, when perhaps the detailed description of known function and design can desalinate main contents of the present invention, these are described in here and will be left in the basket.
Embodiment
The present invention is based on the frequency domain dispersion searching method of training sequence, is mainly that its concrete steps comprise by the adjustment of training sequence modulation symbol and translocation factor is realized to goal of the invention:
S101: at transmitting terminal, the modulation symbol sequence that training sequence is obtained through ovennodulation carries out preliminary treatment, and preprocess method is: by modulation symbol sequence with 2 kindividual is one group, and K is more than or equal to 0 integer, at interval of 1 group, modulation symbol is set to 0 entirely, and all the other remain unchanged.
Fig. 3 is the preliminary treatment result schematic diagram of modulation symbol sequence.As shown in Figure 3, for same basic modulation symbol, when K equals respectively 0,1,2, one group comprises respectively 1,2,4 modulation symbol, at interval of 1 group, modulation symbol is set to 0 entirely.The problem that the amplitude that the variation that follow-up translocation factor is added in this preliminary treatment can solve frequency domain maximum cost function changes along with the variation constantly of ADC sampled point, the later data that also can suppress to sample are cancelled out each other in phase place.
S102: digital coherent light receiver carries out to the received signal coherent demodulation and obtains the 4 road signals of telecommunication, and the Dui Mei road signal of telecommunication carries out respectively dual-rate sampling and quantification.
S103: the continuous sampling signal in step S2 is carried out to the dispersion range that dispersion based on frequency domain maximum cost function search obtains estimating channel, wherein translocation factor Ω cT=N fFT/ 2 k+2.Visible, when K equals respectively 0,1,2, corresponding translocation factor Ω cTbe respectively N fFT/ 4, N fFT/ 8, N fFT/ 16.Adopt translocation factor of the present invention, with respect to being N in prior art fFT/ 2 translocation factor, can adopt larger dispersion step-size in search Δ L to realize dispersion search.
Estimate after the dispersion range of channel obtaining, in actual data transfer process, can directly use this estimation channel dispersion to carry out dispersion compensation.
Fig. 4 is four kinds of dispersion step-size in search comparison diagrams that translocation factor is corresponding.As shown in Figure 4, four kinds of translocation factor N fFT/ 2, N fFT/ 4, N fFT/ 8, N fFT/ 16.This emulation realizes in 112Gb/s PDM-QPSK system, and wherein transmission symbol speed is 28GBaud/s, and the channel dispersion of setting is 100 kilometers, the peak point of each frequency domain maximum cost function in corresponding diagram 4.Can find out, as translocation factor Ω cT=N fFT/ 2 o'clock, the actual dispersion step-size in search that can adopt is very little need to calculate the scope that a large amount of frequency domain maximum cost functions is determined channel dispersion.Adopt other three kinds of translocation factors can adopt larger dispersion step-size in search, for example Ω cT=N fFT, can realize the dispersion step-size in search of search 120 kilometers of left and right at/16 o'clock.Certainly, dispersion step-size in search is less, and the result of its search will be more accurate.But to algorithm complex and hardware resource limitations, larger in the situation that, adopting the present invention can allow to adopt larger dispersion step-size in search to realize the search of dispersion more accurately, and be unlikely to make to estimate channel dispersion and transmission channel dispersion deviation excessive.
Fig. 5 is frequency domain maximum cost function changes in amplitude comparison diagram under different ADC sampling instants.In the contrast simulation of Fig. 5, as translocation factor Ω cT=N fFT/ 2 o'clock, the modulation symbol sequence that training sequence is obtained through ovennodulation does not carry out preliminary treatment, visible under different ADC sampling instants the changes in amplitude of frequency domain maximum cost function differ greatly, even also not necessarily can obtain maximum amplitude in the situation that full remuneration does not exist residual dispersion.And as translocation factor Ω cT=N fFT/ 16 o'clock, the modulation symbol that training sequence is obtained through ovennodulation carried out preliminary treatment, and visible different ADC sampling instant is on the almost not impact of the amplitude of frequency domain maximum cost function.
Below with translocation factor Ω cT=N fFT/ 16 carry out dispersion of the present invention for example searches for emulation.For Ω cT=N fFT/ 16, now only need 4 frequency domain dispersion compensation functions can realize by the dispersion search based on frequency domain maximum cost function channel dispersion search of dispersion values arbitrarily in 200 kilometer range.L is set min=25 kilometers, Δ L=50 kilometer, now L 1=25, L 2=75, L 3=125, L 4=175, compensate respectively 25 kilometers, 75 kilometers, 125 kilometers and 175 kilometers.Finally by four groups, compensate later frequency-region signal and by the maximum cost function amplitude of asking for, rule out the scope of estimating channel dispersion respectively, the amplitude of which frequency domain maximum cost function is maximum, its L idefinite dispersion range is the transmission dispersion range of estimation.
In 112Gb/s PDM-QPSK modulating system, 2nd order chromatic dispersion coefficient is-20ps^2/km, and Optical Signal To Noise Ratio is 12dB, and the stable state frequency deviation of laser is made as 3G and adds transient state frequency deviation.4096 modulation symbols that utilize training sequence modulation to obtain in transmission carry out emulation, receiving terminal is to twice of each symbol sampler, totally 8192 sampled points are divided into 8 groups, every group of 1024 points are FFT and are adopted respectively 4 transmission distance values to calculate frequency domain cost function, and 4 frequency domain maximum cost function amplitudes that finally calculate are also made judgement.
Fig. 6 is the simulation result figure to 4 channel dispersion in 200 kilometer range.As shown in Figure 6, actual transmission range value corresponding to channel dispersion is respectively 10,50,125,200 kilometers.4 values in every curve are respectively the frequency domain maximum cost function amplitude of 25,75,125,175 kilometers of corresponding channel dispersion of compensation.When actual transmissions distance value is 10 kilometers, in 4 frequency domain maximum cost function amplitudes, 25 kilometers of corresponding frequency domain maximum cost function amplitudes are maximum, therefore estimate that the transmission range value obtaining is in the scope of 25 kilometers.Because the dispersion of 50 kilometers of actual transmissions distance values is positioned at the midpoint of 25,75 kilometers of compensation, therefore 25 kilometers and 75 kilometers of corresponding frequency domain maximum cost function amplitude difference little, the transmission dispersion ruling out is all accurately in 25 kilometers or 75 kilometer range.The dispersion that actual transmissions distance value is 125 kilometers is just in time positioned at 125 kms of compensation, at this time can obtain maximum frequency domain maximum cost function.When actual transmissions distance value is 200 kilometers, 175 kilometers of corresponding frequency domain maximum cost function amplitudes are maximum, therefore estimate that the transmission range value obtaining is in 175 kilometer range.
In order to make simulation result have more generality, produce 5000 different transmission channel dispersions at random in 200 kilometers of dispersion range in emulation, the same with above of simulated conditions.Fig. 7 is 5000 transmission channel dispersions and the distribution map of estimating channel dispersion.As shown in Figure 7, in figure, transmission channel dispersion is that with the reason of difference slightly that is distributed with of estimating channel dispersion it is all correct to any one side that the dispersion of transmission is adjudicated in the time of 50,100,150 kilometers after compensation, dispersion such as 50 kilometers of transmission, it is the same with the distance of 25 kilometers, 75 kilometers of compensation, and it is all accurately that the result of search belongs to 25 kilometers or 75 kilometers.
In addition, also the dispersion search in 500 kilometer range has been carried out to emulation.Fig. 8 is the simulation result figure to 4 channel dispersion in 500 kilometer range.Similarly, transmission range value corresponding to channel dispersion actual in this emulation is respectively 10,250,312.5,500 kilometers, the dispersion step-size in search Δ L=125 kilometer of frequency domain dispersion compensation function is set, compensate respectively 62.5 kilometers, 187.5 kilometers, 312.5 kilometers and 437.5 kilometers, translocation factor Ω cT=N fFT/ 16.By the present Ω of simulation hair cT=N fFT, although can make search accurately, only have the maximum of channel dispersion time-frequency domain maximum cost function and the Bizet of minimum value when the close compensation of transmission channel dispersion to be more conducive to judgement at/16 o'clock.Therefore further change translocation factor Ω cTsize, even K=3 makes Ω cT=N fFT/ 32, correspondingly, the training sequence modulation symbol of transmission should be take 8 modulation symbols as one group, is set to 0 at interval of one group, and other modulation symbol remains unchanged.Fig. 9 is the simulation result figure of 4 channel dispersion in 500 kilometer range after adjusting.As shown in Figure 9, compare with Fig. 8, can find out that the difference of 4 frequency domain maximum cost function amplitudes is more obvious, be more conducive to make search accurately.Visible, along with the increase of hunting zone, can correspondingly increase K value.At present conventional digital coherent light communications distance value generally arrives several hundred kilometers tens, and therefore conventional K value is 0,1,2,3.
Although above the illustrative embodiment of the present invention is described; so that those skilled in the art understand the present invention; but should be clear; the invention is not restricted to the scope of embodiment; to those skilled in the art; as long as various variations appended claim limit and definite the spirit and scope of the present invention in, these variations are apparent, all utilize innovation and creation that the present invention conceives all at the row of protection.

Claims (2)

1. the frequency domain dispersion searching method based on training sequence, is characterized in that, comprises the following steps:
S1: at transmitting terminal, the sequence modulation symbol sebolic addressing that training sequence is obtained through ovennodulation carries out preliminary treatment, and preprocess method is: by modulation symbol sequence with 2 kindividual is one group, and K is more than or equal to 0 integer, at interval of 1 group, modulation symbol is set to 0 entirely, and all the other remain unchanged;
S2: digital coherent light receiver carries out to the received signal coherent demodulation and obtains the 4 road signals of telecommunication, and the Dui Mei road signal of telecommunication carries out respectively dual-rate sampling and quantification;
S3: the continuous sampling signal in step S2 is carried out to the scope that dispersion based on frequency domain maximum cost function search obtains estimating channel dispersion, wherein translocation factor Ω cT=N fFT2 k+2.
2. frequency domain search method according to claim 1, is characterized in that, the span of described K is 0≤K≤3.
CN201410006652.6A 2014-01-07 2014-01-07 Frequency domain dispersion search method based on training sequences Expired - Fee Related CN103746950B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201410006652.6A CN103746950B (en) 2014-01-07 2014-01-07 Frequency domain dispersion search method based on training sequences
PCT/CN2014/072121 WO2015103805A1 (en) 2014-01-07 2014-02-16 Frequency-domain dispersion search method based on training sequence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410006652.6A CN103746950B (en) 2014-01-07 2014-01-07 Frequency domain dispersion search method based on training sequences

Publications (2)

Publication Number Publication Date
CN103746950A true CN103746950A (en) 2014-04-23
CN103746950B CN103746950B (en) 2017-04-12

Family

ID=50503939

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410006652.6A Expired - Fee Related CN103746950B (en) 2014-01-07 2014-01-07 Frequency domain dispersion search method based on training sequences

Country Status (2)

Country Link
CN (1) CN103746950B (en)
WO (1) WO2015103805A1 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010094342A1 (en) * 2009-02-20 2010-08-26 Nokia Siemens Networks Gmbh & Co. Kg Method and arrangement for adaptive dispersion compensation.
CN102006261A (en) * 2010-12-09 2011-04-06 北京邮电大学 Channel estimating method for increasing performance of coherent light communication system
CN102742239A (en) * 2011-12-01 2012-10-17 华为技术有限公司 Method and apparatus for frame synchronization in optical fiber system
CN103401829A (en) * 2013-06-26 2013-11-20 吉林大学 IQ imbalance compensation method for coherent-light orthogonal frequency division multiplexing (OFDM) communication system

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2487226B (en) * 2011-01-14 2017-03-08 Nvidia Tech Uk Ltd Equalisation of a signal received over a wireless channel
CN103430003B (en) * 2012-02-01 2016-03-30 华为技术有限公司 For estimating the method for the dispersion of the light signal received
CN103312645B (en) * 2013-06-05 2016-04-13 华中科技大学 A kind of damage self-adapting compensation method of digital coherent optical communication system and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010094342A1 (en) * 2009-02-20 2010-08-26 Nokia Siemens Networks Gmbh & Co. Kg Method and arrangement for adaptive dispersion compensation.
CN102006261A (en) * 2010-12-09 2011-04-06 北京邮电大学 Channel estimating method for increasing performance of coherent light communication system
CN102742239A (en) * 2011-12-01 2012-10-17 华为技术有限公司 Method and apparatus for frame synchronization in optical fiber system
CN103401829A (en) * 2013-06-26 2013-11-20 吉林大学 IQ imbalance compensation method for coherent-light orthogonal frequency division multiplexing (OFDM) communication system

Also Published As

Publication number Publication date
WO2015103805A1 (en) 2015-07-16
CN103746950B (en) 2017-04-12

Similar Documents

Publication Publication Date Title
CN101965698A (en) Apparatus and method for equalizing chromatic dispersion and digital coherent optical receiver
EP2820781B1 (en) Chromatic dispersion processing apparatus and method
US5987063A (en) Method for compensating channel errors in a digital data communication system
Secondini et al. Enhanced split-step Fourier method for digital backpropagation
EP2997676B1 (en) Method, device and communication system for reducing optical transmission impairments
EP2709328B1 (en) Coefficient determining apparatus, equalizer, receiver and transmitter
CN107395282A (en) Phase noise compensation method of time domain unscented Kalman filtering large-linewidth CO-OFDM system
CN103338171B (en) A kind of receiving terminal equalization methods based on channel estimation in frequency domain and system
CN109347562A (en) CO-OFDM system phase noise optimization compensation method
EP3048746B1 (en) Method and device for estimation of chromatic dispersion in optical coherent communication
US20100209115A1 (en) Frequency-Domain Equalization of the Fiber Optic Channel
CN101534266A (en) Channel estimation method for Orthogonal Frequency Division Multiplexing system and device thereof
EP2465219B1 (en) Filter device
WO2010145075A1 (en) Method for carrier frequency recovery and optical intradyne coherent receiver
EP3096469B1 (en) Chromatic dispersion measurement method and device and digital coherent receiver
CN107078982A (en) The dispersive estimates of digital coherent optical receiver
Lu et al. Vertical blind phase search for low-complexity carrier phase recovery of offset-QAM Nyquist WDM transmission
Zhou et al. A fast and robust blind chromatic dispersion estimation based on fractional fourier transformation
CN101964765B (en) Signal compensation method and device
Kamalian et al. Spectral efficiency estimation in periodic nonlinear Fourier transform based communication systems
CN102255667A (en) Method and device for dispersion compensation
EP3143708B1 (en) Non-linear propagation impairment equalisation in optical transmission
CN103746950A (en) Frequency domain dispersion search method based on training sequences
Qiu et al. Fourier Neural Operator Based Fibre Channel Modelling for Optical Transmission
Abdullah et al. Frequency domain technique for reducing chromatic dispersion

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20170412

Termination date: 20200107

CF01 Termination of patent right due to non-payment of annual fee