WO2013124986A1

WO2013124986A1 - Polarization estimator, polarization splitter, optical receiver, polarization estimation method, and polarization splitting method

Info

Publication number: WO2013124986A1
Application number: PCT/JP2012/054284
Authority: WO
Inventors: 吉田　剛
Original assignee: 三菱電機株式会社
Priority date: 2012-02-22
Filing date: 2012-02-22
Publication date: 2013-08-29
Also published as: JPWO2013124986A1; JP5657168B2

Abstract

A polarization estimator (301) imparts different polarization rotations to an inputted signal with two polarizations (two polarization signals), carries out differential detection for each set of two polarization signals to which the same polarization rotation is imparted, acquires only the real parts of the outputs, divides the results of multiplying the real parts with each other by the sum of squares of the real parts, and on the basis of the results of division, calculates an estimated value for the polarization rotation angle of the two polarization signals. As a result, polarization splitting can be carried out significantly faster than in the prior art, and even if the polarizations fluctuate rapidly, transmission quality can be maintained.

Description

Polarization estimator, polarization separator, optical receiver, polarization estimation method, and polarization separation method

The present invention relates to a polarization estimator, a polarization separator, an optical receiver, a polarization estimation method, and a polarization separation method, and more particularly to a polarization estimator, a polarization separator, and an optical device using a digital coherent method. The present invention relates to a receiver, a polarization estimation method, and a polarization separation method.

For high-capacity optical transmission, overcoming the optical signal-to-noise power limit and high-density wavelength multiplexing are challenges. As a technology to overcome the optical signal power vs. noise power limit, binary phase-shift keying (BPSK) and quaternary PSK (Quaternary) are compared to conventional on-off keying (OOK). Phase-Shift Keying (QPSK) is used. For high-density wavelength multiplexing, a method of doubling the number of transmission bits per symbol by polarization multiplexing that assigns independent signals to two orthogonal polarization components, QPSK, or 16-value orthogonal There is known a method of increasing the signal multiplicity and increasing the number of transmission bits per symbol, such as amplitude modulation (QAM). QPSK and 16QAM are transmitted by assigning signals to the same phase axis (In-Phase axis: I axis) and quadrature phase axis (Quadrature-Phase axis: Q axis) in the optical transmitter.

Also, a digital coherent method that receives these optical modulation signals by combining digital signal processing with the synchronous detection method has attracted attention. In this method, chromatic dispersion and polarization mode dispersion (Polarization-Mode Dispersion) generated in a transmission line are achieved by linear photoelectric conversion by synchronous detection and fixed, semi-fixed and adaptive linear equalization by digital signal processing. It is possible to realize excellent equalization characteristics and excellent noise tolerance against linear waveform distortion caused by PMD).

The polarization multiplexed signal is separated by a linear equalizer in the receiver. The filter parameters of the linear equalizer are normally updated sequentially by digital signal processing based on an adaptive algorithm including a feedback path as described in Non-Patent Documents 1 to 4 and Patent Documents 1 to 3.

JP2011-0033195A US Pat. No. 7,623,797 Special table 2010-539789

However, according to the conventional techniques described in Non-Patent Documents 1 to 4 and Patent Documents 1 to 3, the feedback path is required for updating the filter parameters. There was a problem that separation did not function effectively and communication was lost.

For example, in the example described in Non-Patent Document 4, the ability to follow polarization fluctuation is 50 kHz at the maximum, and when the polarization fluctuates in the order of MHz, the polarization cannot be appropriately separated and communication is impossible. There was a problem of falling into.

The present invention has been made to solve such a problem, and can prevent deterioration in transmission quality even when the polarization fluctuates at high speed. The polarization estimator, the polarization separator, and the optical It is an object to obtain a receiver, a polarization estimation method, and a polarization separation method.

The present invention provides a polarization rotation unit that receives two polarization signals, separates the polarization signals into four systems, and applies different polarization rotations for each system to the two polarization signals included in each system And an inter-polarization differential detection unit that performs differential detection between the two polarization signals by inputting a two-polarization signal that has been polarized by the polarization rotation unit for each of the four systems. , For each of the four systems, a real part acquisition unit that acquires only a real part from the output of the inter-polarization differential detection unit, and two sets of two sets formed as two sets of the four systems In each set, a complex multiplication unit that performs multiplication between the two real parts included in the set, and a square sum operation that calculates a square sum of the two real parts included in the set in each set And the calculation result of the complex multiplication unit in each set is the calculation result of the square sum calculation unit. A division unit for calculating, an averaging unit for storing and averaging the calculation results of the division unit in each set, and the polarization rotation unit based on the calculation results of the averaging units of the two sets The polarization estimator includes a polarization rotation angle calculation unit that calculates an estimated value of the polarization rotation angle of the two input polarization signals.

The present invention provides a polarization rotation unit that receives two polarization signals, separates the polarization signals into four systems, and applies different polarization rotations for each system to the two polarization signals included in each system And an inter-polarization differential detection unit that performs differential detection between the two polarization signals by inputting a two-polarization signal that has been polarized by the polarization rotation unit for each of the four systems. , For each of the four systems, a real part acquisition unit that acquires only a real part from the output of the inter-polarization differential detection unit, and two sets of two sets formed as two sets of the four systems In each set, a complex multiplication unit that performs multiplication between the two real parts included in the set, and a square sum operation that calculates a square sum of the two real parts included in the set in each set And the calculation result of the complex multiplication unit in each set is the calculation result of the square sum calculation unit. A division unit for calculating, an averaging unit for storing and averaging the calculation results of the division unit in each set, and the polarization rotation unit based on the calculation results of the averaging units of the two sets Since the polarization estimator includes a polarization rotation angle calculation unit that calculates an estimation value of the polarization rotation angle of the two input polarization signals, the polarization rotation angle estimation in the transmission path is performed in a feedforward manner. Can be performed only by processing and blind processing, polarization separation can be performed much faster than before, and transmission quality can be maintained even when the polarization fluctuates at high speed. it can.

It is a figure which shows the polarization rotation in an optical transmission line. It is a block diagram which shows the structure of the polarization estimator which concerns on Embodiment 1 of this invention. It is a figure which shows the input polarization dependence of the polarization estimation result by the polarization estimator which concerns on Embodiment 1 of this invention. It is a block diagram which shows the structure of the polarization separator which concerns on Embodiment 2 of this invention, and its peripheral circuit. It is the block diagram which showed the structure for the offline analysis for confirming the polarization splitting capability of the polarization splitter which concerns on Embodiment 2 of this invention. It is an offline analysis result which shows the polarization separation capability of the polarization separator which concerns on Embodiment 2 of this invention. It is a block diagram which shows the structure of the optical receiver which concerns on Embodiment 3 of this invention.

Embodiments of a polarization estimator, a polarization separator, and an optical receiver according to the present invention will be described below in detail with reference to the drawings. In addition, embodiment described below is one form at the time of actualizing this invention, Comprising: It is not for limiting this invention in the range.

Embodiment 1 FIG.
FIG. 2 is a block diagram showing the configuration of the polarization estimator 301 according to Embodiment 1 of the present invention. The polarization estimator 301 according to the first embodiment is mounted and used in a polarization separator or an optical receiver described later. The polarization estimator 301 of the first embodiment receives two polarization signals (Erx, Ery), performs polarization estimation based on them, and outputs an estimated value θest of the polarization rotation angle. is there. In FIG. 2, 41 and 43 are both X polarization received optical signals (Erx), and 42 and 44 are both Y polarization received optical signals (Ery). These polarization signals (Erx, Ery) are all digital signals.

As shown in FIG. 2, the polarization estimator 301 according to the first embodiment divides the input two-polarization signals (Erx, Ery) into four systems and has their own polarization rotations (0, − π / 4, + π / 8, -π / 8), and a polarization differential detector (complex conjugate calculation) that performs differential detection between two polarizations that have the same polarization rotation. Sections 61 to 64 and complex multiplication sections 71 to 74), real part acquisition sections 81 to 84 for acquiring only the real part of the differential detection results between the polarizations, and two sets of the four systems as two sets.

Complex multipliers

101 and 102 that form two sets and perform multiplication between the real parts, and

coefficient input units

91 and 92 that input predetermined coefficients to the complex multipliers 101 and 102 (here, coefficient = 2) )), And a sum of squares calculation unit (square calculation units 111 to 114 and Adders 121 and 122),

division units

131 and 132 that divide the complex multiplication result by the square sum operation result, averaging

units

141 and 142 that store and average the division result, and two sets of the averages And a polarization rotation angle calculation unit 151 that calculates a polarization rotation angle based on the conversion result.

Hereinafter, the operation of the polarization estimator 301 according to the first embodiment will be described. Hereinafter, an example of generating a 128 Gbit / s polarization multiplexed QPSK optical signal will be described. The 128 Gbit / s polarization multiplexed QPSK optical signal can communicate 4 bits per symbol, and the symbol repetition frequency fs is 32 GHz. The symbol repetition period is Ts (= 1 / fs). The first embodiment can be applied to other transmission rates and various modulation schemes, and is not limited to this example.

The input two-polarized signals (Erx, Ery) are electric signals that have undergone substantially linear optical / electrical conversion by synchronous detection (intradyne detection) of polarization multiplexed optical signals.

The polarization rotation units 51 to 54 apply polarization rotation based on a 2 × 2 complex matrix (Jones matrix) to the input two polarization signals (Erx, Ery), respectively, and 2 after applying polarization rotation. Outputs a polarization signal.

For example, the polarization rotation angle in the polarization rotation unit 51 is 0, the polarization rotation angle in the polarization rotation unit 52 is −π / 4, the polarization rotation angle in the polarization rotation unit 53 is + π / 8, and the polarization rotation unit The polarization rotation angle at 54 may be −π / 8. However, these combinations are merely examples, and the polarization rotation angles of the polarization rotation units 51 to 54 may be appropriately set to arbitrary angles, and are not limited to the above angles.

The polarization rotation unit 51 applies polarization rotation to each of the polarization signals 41 and 42 (Erx, Ery) at a polarization rotation angle of 0, and outputs two polarization signals (ErIxp, ErIyp).
The polarization rotation unit 52 applies polarization rotation to each of the polarization signals 41 and 42 (Erx, Ery) at a polarization rotation angle of −π / 4, and outputs two polarization signals (ErIxn, ErIyn).
The polarization rotation unit 53 applies polarization rotation to each of the polarization signals 43 and 44 (Erx, Ery) at a polarization rotation angle + π / 8, and outputs a two-polarization signal (ErRxp, ErRyp).
The polarization rotation unit 54 applies polarization rotation to each of the polarization signals 43 and 44 (Erx, Ery) at a polarization rotation angle of −π / 8, and outputs two polarization signals (ErRxn, ErRyn).

The complex conjugate calculation unit 61 receives ErIyp as an input and outputs the complex conjugate ErIyp * to the complex multiplication unit 71.
The complex conjugate calculation unit 62 receives ErIyn as an input and outputs the complex conjugate ErIyn * to the complex multiplication unit 72.
The complex conjugate calculation unit 63 receives ErRyp as an input and outputs the complex conjugate ErRyp * to the complex multiplication unit 73.
The complex conjugate calculation unit 64 receives ErRyn as an input and outputs the complex conjugate ErRyn * to the complex multiplication unit 74.

The complex multiplication unit 71 performs complex multiplication of ErIxp and ErIyp *, and outputs the resulting complex signal to the real part acquisition unit 81.
The complex multiplier 72 performs a complex multiplication of ErIxn and ErIyn *, and outputs the resulting complex signal to the real part acquisition unit 82.
The complex multiplier 73 performs a complex multiplication of ErRxp and ErRyp * and outputs the resulting complex signal to the real part acquisition unit 83.
The complex multiplier 74 performs complex multiplication of ErRxn and ErRyn * and outputs the resulting complex signal to the real part acquisition unit 84.

The real part acquisition unit 81 acquires the real part of the complex signal output from the complex multiplication unit 71 and outputs the real part signal A _I to the multiplication unit 101 and the square calculation unit 111.
The real part acquisition unit 82 acquires the real part of the complex signal output from the complex multiplier 72 and outputs the real part signal _BI to the multiplier 101 and the square calculator 112.
The real part acquisition unit 83 acquires the real part of the complex signal output from the complex multiplication unit 73 and outputs the real part signal A _R to the multiplication unit 102 and the square calculation unit 113.
The real part acquisition unit 84 acquires the real part of the complex signal output from the complex multiplication unit 74 and outputs the real part signal _BR to the multiplication unit 102 and the square calculation unit 114.

The multiplication unit 101 includes a real part signal A _I input from the real part acquisition unit 81, a real part signal B _I input from the real part acquisition unit 82, and a coefficient (coefficient = 2) input from the coefficient input unit 91. ) And the product (2A _I B _I ) is output to the division unit 131.
The multiplication unit 102 receives the real part signal A _R input from the real part acquisition unit 83, the real part signal B _R input from the real part acquisition unit 84, and the coefficient input from the coefficient input unit 92 (coefficient = 2). ) And the product (2A _R B _R ) is output to the division unit 132.

The square calculation unit 111 squares the real part signal A _I input from the real part acquisition unit 81, and outputs the square result (A _I ² ) to the addition unit 121.
The square calculation unit 112 squares the real part signal B _I input from the real part acquisition unit 82, and outputs the square result (B _I ² ) to the addition unit 121.
The square calculation unit 113 squares the real part signal A _R input from the real part acquisition unit 83 and outputs the square result (A _R ² ) to the addition unit 122.
Square calculation unit 114 squares the real part signal B _R input from the real acquisition unit 84, and outputs the squared result (B _R ²⁾ to the adder 122.

The adding unit 121 adds the square result A _I ² input from the square calculation unit 111 and the square result B _I ² input from the square calculation unit 112, and divides the addition result (A _I ² + B _I ² ). To the unit 131.
The adding unit 122 adds the square result A _R ² input from the square calculation unit 113 and the square result B _R ² input from the square calculation unit 114, and divides the addition result (A _R ² + B _R ² ). Output to the unit 132.

The division unit 131 divides the multiplication result 2A _I B _I input from the multiplication unit 101 by the addition result A _I ² + B _I ² input from the addition unit 121, and the division result C _I to the averaging unit 141. Output.
Divider 132, a multiplication result 2A _R B _R input from the multiplication unit 102, divided by the sum A _{_R} ^² + B _R ² input from the addition section 122, the averaging unit 142 and the division result C _R Output.

The averaging unit 141 averages the division result C _I input from the division unit 131 and outputs the average result C _{I, avg} to the polarization rotation angle calculation unit 151. For example, although it is assumed that a moving average for 20 samples is taken, the present invention is not limited to this, and any other averaging process may be performed. Further, without averaging, it may be output as it is C _I.
The averaging unit 142 averages the division result C _R input from the division unit 132 and outputs the averaged result C _{R, avg} to the polarization rotation angle calculation unit 151. For example, although it is assumed that a moving average for 20 samples is taken, the present invention is not limited to this, and any other averaging process may be performed. Further, without averaging, it may be output as it is C _R.
Basically, it is possible to suppress noise components by performing averaging processing, but the longer the averaging time, the lower the ability to follow the fluctuation of the transmission line, so optimal averaging according to the transmission line It is most desirable to select a treatment. Further, simply, it is not necessary to average.

The polarization rotation angle calculation unit 151 is based on C _{I, avg} input from the averaging unit 141 and C _{R, avg} input from the averaging unit 142 _, and polarization of the two-polarization signal (Erx, Ery). An estimated value θest of the rotation angle is obtained by the following equation (1) and output to the outside. Arg {·} is a function for obtaining the argument of a complex number.

θest = 1/4 × Arg {(CR _{, avg} ) + j (CI _{, avg} )} (1)

As described above, the polarization rotation angle of the input polarization signal (Erx, Ery) by the polarization estimator 301 according to the first embodiment can be obtained only by feed-forward processing and (pilot tone etc. Can be estimated by blind processing.

Hereinafter, it will be explained that the polarization multiplexed signal can be separated by estimating the polarization rotation angle.

In the optical transmission line, the polarization state changes variously. Considering unmodulated light, the polarization state is generally represented by elliptical polarization, and as a special case, linear polarization or circular polarization can be taken. These polarization state changes are caused by a delay difference in the wavelength order between orthogonal polarizations.

In polarization multiplexed signals, independent signals are superimposed on two orthogonal polarizations. At this time, in the transmitter, for example, two systems of linearly polarized unmodulated signals are independently modulated, and the polarization relationship is orthogonalized and multiplexed.

The polarization multiplexed signal can be handled as two linearly polarized signals that are independent up to the receiving end in the linear analysis.

In the receiver, an optical signal is converted into an electric signal. At this time, since the response of the electrical device does not follow the polarization that changes in the wavelength order of the light, for example, when receiving a circularly polarized optical signal, the X polarization and Y polarization of the polarization diversity receiver In both cases, it is simply received with half the power. Even in the case of the more generalized elliptically polarized wave, similarly, only the X-polarized wave and Y-polarized wave of the polarization diversity receiver appear to be received with a certain proportion of power. As for the polarization diversity receiver, for example, Optical Internetworking Forum, “100G Ultra Long Haul DWDM Framework Document”, (http://www.oiforum.com/ public / documents / OIF-FD-100G-DWDM-01.0 .pdf), refer to the receiver described in June 2009.

These are generalized when receiving linearly polarized signals. This is because when the polarization angle of the linearly polarized signal is changed and received, the power ratio between the X polarization and the Y polarization of the polarization diversity receiver changes. Therefore, the polarization state fluctuation in the transmission line may be formulated by polarization rotation of linearly polarized waves.

FIG. 1 shows polarization state fluctuations in an optical transmission line described by polarization rotation using a Jones matrix. In FIG. 1, 11 is an X polarization transmission optical signal (Etx), 12 is a Y polarization transmission optical signal (Ety), 31 is an X polarization reception optical signal (Erx), and 32 is a Y polarization reception optical signal. (Ery) and 21 are optical transmission lines. It is only necessary to simplify the transmission

optical signals

11 and 12 and the reception

optical signals

31 and 32 so that two orthogonal linearly polarized waves are data-modulated. You can think of the waves rotating.

From the above, by estimating the polarization rotation angle in the optical transmission line, the signal power ratio between the X polarization and Y polarization of the polarization diversity receiver can be found, and the original linear polarization state is restored. It turns out that it will be possible. This does not depend on whether the transmission side is polarization multiplexed.

FIG. 3 is a diagram showing the input polarization dependence of the polarization estimation result by the polarization estimator 301 according to the first embodiment. In FIG. 3, the horizontal axis represents the input polarization rotation angle, and the vertical axis represents the estimated value of the polarization rotation angle. This is the same characteristic as when the phase estimation of the QPSK signal is performed by the Viterbi & Viterbi algorithm, which is a method suitable for performing the carrier phase estimation process, which is necessary for the receiver of the same synchronous detection (intradyne detection). This indicates that the polarization state can be estimated at the same level as the phase estimation of the QPSK signal. As for the Viterbi & Viterbi algorithm, for example, Andrew J. Viterbi and Audrey M. Viterbi, “Nonlinear Estimation of PSK-Modulated Carrier Phase with Application to Burst Digital Transmission”, IEEE Transactions on Information Theory,. Please refer to 4, pp.543-551, 1983.

Also, according to the polarization estimator 301 according to the first embodiment, the polarization state estimation capability does not change even if the number of phases of the phase modulation signal changes. That is, even if 8-value PSK or 16-value PSK is adopted, the polarization estimation capability is not changed at all. However, it is not suitable for application to a modulation method in which the amplitude level is multilevel.

As described above, according to the first embodiment of the present invention, the polarization estimator 301 receives two polarization signals, separates the polarization signals into four systems, and is included in each system. Polarization rotating units 51 to 54 that apply different polarization rotations to the polarization signal for each system, and two polarization signals that are polarized by the polarization rotation unit for each of the four systems. Are input for each of the four systems, and differential polarization detectors (complex conjugate arithmetic units 61 to 64 and complex multipliers 71 to 74) for differential detection between two polarization signals. Included in the set are the real part acquisition units 81 to 84 that acquire only the real part from the output of the inter-wave differential detection unit, and two sets formed by setting two systems out of the four systems as one set In the

complex multipliers

101 and 102 that perform multiplication of the real parts of two systems, A sum of squares computing unit (square computing units 111 to 114 and adding units 121 and 122) for calculating the sum of squares of the two real parts included in the set, and the computation result of the complex multiplication unit in each set The

division units

131 and 132 that divide by the calculation result of the square sum calculation unit, the averaging

units

141 and 142 that store and average the calculation result of the division unit in each set, and the two sets of the above A polarization rotation angle calculation unit 151 that calculates an estimated value of the polarization rotation angle of the two polarization signals input to the polarization rotation unit based on the calculation result of the averaging unit; The polarization rotation angle estimation in the transmission path can be performed only by feedforward processing and blind processing, and even when the polarization fluctuates at high speed, the polarization estimation capability does not change and transmission quality is improved. Can be maintained.

Embodiment 2. FIG.
FIG. 4 is a block diagram showing a configuration example of the polarization separator and its peripheral circuit according to the second embodiment of the present invention. In FIG. 4A,

reference numerals

204 and 206 denote polarization separators according to the second embodiment, and the other is a peripheral circuit thereof. Accordingly, the peripheral circuit includes a preprocessing unit 201, a chromatic dispersion compensation unit 202, a timing extraction unit 203, an adaptive equalization unit 205, a carrier frequency offset compensation unit 207, a carrier phase offset compensation unit 208, and an identification unit. 209.

As shown in FIG. 4A, the polarization separator 204 is provided between the timing extraction unit 203 and the adaptive equalization unit 205. The polarization separator 206 is provided between the adaptive equalization unit 205 and the carrier frequency offset compensation unit 207.

The polarization separator 204 and the polarization separator 206 have the same configuration. FIG. 4B is a diagram showing the internal configuration of the

polarization separators

204 and 206. As shown in FIG. 4B, the

polarization separators

204 and 206 are composed of the polarization estimator 301 of FIG. 2 described in the first embodiment and the polarization rotation unit 302. Details will be described later.

Hereinafter, the polarization separator and its peripheral circuit according to the second embodiment will be described.

The pre-processing unit 201 receives a 4-lane digital signal from the outside. These digital signals are composed of an I-ch signal and a Q-ch signal, respectively, for the X polarization signal and the Y polarization signal. Therefore, in the following, the 4-lane digital signals are represented as XI, XQ, YI, and YQ. The preprocessing unit 201 corrects amplitude variations, delay errors, and the like of the 4-lane digital signals (XI, XQ, YI, YQ), and outputs the corrected 4-lane signal to the chromatic dispersion compensation unit 202.

The chromatic dispersion compensator 202 compensates the chromatic dispersion received on the optical transmission line by digital signal processing for the 4-lane signal input from the preprocessor 201 and outputs the compensated 4-lane signal to the timing extractor 203. .

The timing extraction unit 203 extracts symbol timing information based on the 4-lane signal input from the chromatic dispersion compensation unit 202, and reproduces the identification point. The discrimination lane regenerated 4-lane signal is output to the polarization separator 204.

The polarization separator 204 includes the polarization estimator 301 described in the first embodiment and the polarization rotation unit 302 connected thereto.

The polarization estimator 301 estimates the polarization rotation angle based on the 4-lane signal input from the timing extraction unit 203, and outputs the polarization rotation angle estimated value θest to the polarization rotation unit 302. Since the operation of the polarization estimator 301 is the same as that described in Embodiment 1, the description thereof is omitted here. However, here, XI and YI are considered to be Erx41 and Ery42 shown in the first embodiment, and similarly, XQ and YQ are considered to be Erx43 and Ery44 shown in the first embodiment.

The polarization rotation unit 302 gives a reverse rotation of the estimated value θest of the polarization rotation angle input from the polarization estimator 301 to the 4-lane signal (two polarization signals) input from the timing extraction unit 203. Then, a 4-lane signal (two-polarized signal) obtained by reversely rotating the polarization is output to the adaptive equalization unit 205.

The adaptive equalization unit 205 adaptively equalizes linear waveform distortion due to residual chromatic dispersion, polarization mode dispersion, bandwidth limitation, and the like with respect to the 4-lane signal input from the polarization separator 204, and after equalization Are output to the polarization separator 206. Although the transmission polarization is generally restored by the polarization separator 204, the polarization separator 204 can only give a polarization rotation, so an adaptive equalization unit 205 that performs waveform equalization is necessary in many cases. It is. However, it can be made unnecessary when a signal that has passed through a transmission path in good condition is received.

The polarization separator 206 is basically the same in configuration and operation as the polarization separation unit 204, and includes a polarization estimator 301 and a polarization rotation unit 302 as shown in FIG. Yes. The polarization separator 206 is used to remove the remaining polarization interference component that cannot be separated by the polarization separator 204. In the polarization separator 206, first, the polarization estimator 301 estimates the polarization rotation angle based on the 4-lane signal input from the adaptive equalization unit 205, and uses the polarization rotation angle estimated value θest as the polarization. Output to the rotating unit 302. Next, the polarization rotation unit 302 reverses the estimated value θest of the polarization rotation angle input from the polarization estimator 301 with respect to the 4-lane signal (two polarization signals) input from the timing extraction unit 203. A four-lane signal (two-polarized signal) that is rotated and reversely polarized is output to the carrier frequency offset compensator 207.

The carrier frequency offset compensation unit 207 compensates for the frequency difference between the carrier wave of the received signal and the local oscillation wave for the 4-lane signal input from the polarization separation unit 206, and converts the 4-lane signal after frequency difference compensation to the carrier phase. Output to the offset compensator 208.

The carrier phase offset compensation unit 208 compensates the phase difference between the carrier wave of the received signal and the local oscillation wave for the 4-lane signal input from the carrier frequency offset compensation unit 207, and identifies the 4-lane signal after the phase offset compensation Output to the unit 209.

The identification unit 209 performs a hard decision on the 4-lane signal input from the carrier phase offset compensation unit 208, adds reliability information in some cases, and outputs the 4-lane signal to the outside.

FIG. 5 shows a block diagram for off-line analysis for confirming the polarization separation capability of the

polarization separators

204 and 206 according to the present invention. The same components as those in FIG. 4 are denoted by the same reference numerals, and description thereof is omitted here. In FIG. 5, 210 is a polarization rotation emulator unit that simulates polarization fluctuations, 211 is a sampling rate conversion unit that eliminates the mismatch between the sampling rate and the symbol repetition frequency, and sets the sampling rate to an approximately integer multiple of the symbol repetition frequency, Reference numeral 212 denotes a Q value calculation unit. The polarization rotation emulator unit 210 intentionally rotated the polarization in the off-line analysis, and confirmed the reception characteristics.

FIG. 6 is a result of confirming the polarization separation capability of the polarization separation unit according to the present invention by the configuration of FIG. As an example of a normal polarization separation method having a feedback path, a standard CMA was used as a comparison target. Noise is added to a 43 Gb / s polarization multiplexed quaternary phase shift keying signal, and sine is intentionally received at the receiving end signal processing under the condition of optical signal power to noise power ratio of 10.9 dB (noise band 0.1 nm). Waveform polarization rotation: θ (t) = πsin (πft) was given to confirm the Q value characteristic. FIG. 6 shows the following three cases.

1) When not using the polarization separator of the present invention (standard CMA)
2) When the polarization separator of the present invention is used only in the preceding stage of the adaptive equalization unit 205 (that is, when only the polarization separator 204 of FIG. 5 is used)
3) When the polarization separator of the present invention is used in both the front stage and the rear stage of the adaptive equalization unit 205 (that is, when the

polarization separators

204 and 206 in FIG. 5 are used).

The adaptive equalization unit 205 is configured by a butterfly finite impulse response filter with a 9-tap 1/2 symbol interval, and is optimized for a series of 1024 symbol received signals by standard CMA. The horizontal axis of FIG. 6 represents the polarization rotation speed obtained by normalizing the maximum polarization change speed f with the symbol repetition frequency Bs = 10.7 GHz. The vertical axis in FIG. 6 indicates the Q value characteristic. When it is specified by the 0.5 dBQ value deterioration point, it can be seen that the

polarization followers

204 and 206 of the present invention increase the polarization tracking speed by two digits under this condition. The contribution to the increase in speed is dominated by the polarization separation unit in front of the adaptive equalization unit.

As described above, according to the second embodiment, the

polarization separators

204 and 206 have the polarization estimator 301 described in the first embodiment and the polarization rotation angle estimated by the polarization estimator 301. And the polarization rotation unit 302 that reversely rotates the polarization based on the polarization rotation angle estimation in the transmission line can be performed only by the feedforward process and the blind process. It is possible to maintain the transmission quality even when the polarization separation of the wave-multiplexed signal is significantly faster than in the past and the polarization fluctuates at high speed.

4 and 5 show examples in which the

polarization separators

204 and 206 are provided in the front stage and the rear stage of the adaptive equalization unit 205, respectively. However, only one of the

polarization separators

204 and 206 may be provided. Even in such a case, sufficient effects can be obtained as shown in FIG.

Embodiment 3 FIG.
FIG. 7 is a block diagram showing a configuration of an optical receiver according to Embodiment 3 of the present invention. The optical receiver according to the third embodiment includes a local oscillation light source 401, an optical interference unit 402, a photoelectric conversion unit 403, an analog / digital conversion unit 404, and a digital signal processing unit 405.

The local oscillation light source 401, the optical interference unit 402, and the photoelectric conversion unit 403 constitute an optical receiving unit that receives an optical signal (hereinafter referred to as a received optical signal) from the outside and performs reception processing. .

The local oscillation light source 401 generates CW (Continuous Wave) light having a center wavelength approximately coincident with the center wavelength of the optical signal after transmission, and outputs the CW light to the optical interference unit 402.

The optical interference unit 402 receives the received optical signal from the outside and also receives the CW light from the local oscillation light source 401, interfers and mixes the received optical signal and the CW light, and converts the optical signal after the interference into an electrical signal. The data is output to the conversion unit 403.

The photoelectric converter 403 performs synchronous detection (intradyne detection) by converting the optical signal after interference input from the optical interference unit 402 into an electrical signal, and outputs the electrical signal to the analog / digital conversion unit 404. .

The analog / digital conversion unit 404 converts the 4-lane analog electrical signal input from the photoelectric conversion unit 403 into a discrete-time digital signal, and outputs the discrete-time digital signal to the digital signal processing unit 404. In analog-digital conversion, both discrete time conversion of continuous-time signals and quantization of amplitude levels are performed. The ratio of the sampling rate by the discrete time to the symbol rate is set to 1 Sample / symbol or higher, and is generally set to 2 Sample / symbol in many cases.

The digital signal processing unit 405 includes the polarization separators 204 and / or 206 shown in the second embodiment as shown in FIG. Based on the 4-lane signal input from the analog / digital conversion unit 404, the digital signal processing unit 405 performs reception electric signal processing including the polarization separation processing described in the second embodiment, and outputs the signal processing result. Output.

Also in the third embodiment, the configuration of the

polarization separators

204 and 206 is as shown in FIG. 4 of the second embodiment, that is, the polarization estimator 301 shown in the first embodiment is provided. is doing. Since the configurations and operations of the

polarization separators

204 and 206 are as described in the second embodiment, a detailed description thereof is omitted here.

Note that the digital signal processing unit 405 may include both of the

polarization separators

204 and 206 shown in FIG. 4, or may include only one of them.

Furthermore, the digital signal processing 405 may include all the configurations (201 to 209) of FIG. 4 shown in the second embodiment. In that case, the digital signal processing unit 405 receives electric power received from the preprocessing unit 201 to the identification unit 209 described in the second embodiment based on the 4-lane signal input from the analog / digital conversion unit 404. Perform signal processing and output signal processing results. Since each process from the preprocessing unit 201 to the identification unit 209 is as described in the second embodiment, the description thereof is omitted here.

As described above, according to the third embodiment, the optical receiver oscillates the CW light at the center wavelength substantially coincident with the received optical signal, and the CW light generated by the local oscillation light source 401. And an optical signal output from the optical receiver. The optical receiver includes an optical interference unit 402 that interferes with the received optical signal, and an optical / electrical converter 403 that converts an output from the optical interference unit 402 into an electrical signal. Is converted to a digital signal, and a digital signal processing unit 405 that performs polarization separation is provided, so that the polarization rotation angle estimation in the transmission path can be performed only by feedforward processing and blind processing. In addition, polarization separation of orthogonal polarization multiplexed signals can be made much faster than before, and transmission quality can be maintained even when the polarization fluctuates at high speed. Kill.

As described above, the polarization estimator, the polarization separator, and the optical receiver according to the present invention are useful for an optical transmission system using a digital coherent method, and in particular, light whose polarization changes rapidly. Suitable for transmission system.

11 X polarization transmission optical signal, 12 Y polarization transmission optical signal, 21 optical transmission line, 31 X polarization reception optical signal, 32 Y polarization reception optical signal, 41, 43 X polarization reception signal, 42, 44 Y Polarization reception signal, 51, 52, 53, 54 Inter-polarization differential detection unit, 61, 62, 63, 64 Complex conjugate operation unit, 71, 72, 73, 74 Complex multiplication unit, 81, 82, 83, 84 Part acquisition unit, 101, 102 complex multiplication unit, 111, 112, 113, 114 square operation unit, 121, 122 addition unit, 131, 132 division unit, 141, 142 averaging unit, 151 polarization rotation angle calculation unit, 201 Pre-processing unit, 202 chromatic dispersion compensation unit, 203 timing extraction unit, 204 polarization separation unit, 205 adaptive equalization unit, 206 polarization separation unit, 207 carrier frequency offset compensation unit, 20 Carrier phase offset compensation unit, 209 identification unit, 301 polarization estimator, 302 polarization rotation unit, 401 local oscillation light source, 402 optical interference unit, 403 photoelectric conversion unit, 404 analog / digital conversion unit, 405 digital signal processing unit .

Claims

A signal having two polarizations (= 2 polarization signals) is inputted, the two polarization signals are separated into four systems, and different polarization rotations are performed for each system with respect to the two polarization signals included in each system. A polarization rotator to give,
For each of the four systems, a two-polarization signal that is polarized by the polarization rotation unit is input, and one of the two polarization signals and the other of the polarization signals A differential detection section between polarized waves that performs differential detection between
For each of the four systems, a real part acquisition unit that acquires only a real part from the output of the differential detection section between polarizations,
A complex multiplier that performs multiplication of the two real parts included in the set in each of two sets formed as a set for every two systems of the four systems,
In each set, a sum of squares calculation unit that calculates the sum of squares of the real parts of the two systems included in the set;
In each set, a division unit that divides the calculation result of the complex multiplication unit by the calculation result of the square sum calculation unit;
In each set, an averaging unit that accumulates and averages the calculation results of the division unit;
A polarization rotation angle calculation unit that calculates an estimated value of the polarization rotation angle of the two polarization signals input to the polarization rotation unit based on the calculation results of the two sets of the averaging units; A polarization estimator characterized by comprising.
A polarization rotation unit that receives two polarization signals, separates the two polarization signals into four systems, and applies different polarization rotations for each system to the two polarization signals included in each system;
For each of the four systems, a two-polarization signal that is polarized by the polarization rotation unit is input, and one of the two polarization signals and the other of the polarization signals A differential detection section between polarized waves that performs differential detection between
For each of the four systems, a real part acquisition unit that acquires only a real part from the output of the differential detection section between polarizations,
A complex multiplier that performs multiplication of the two real parts included in the set in each of two sets formed as a set for every two systems of the four systems,
In each set, a sum of squares calculation unit that calculates the sum of squares of the real parts of the two systems included in the set;
In each set, a division unit that divides the calculation result of the complex multiplication unit by the calculation result of the square sum calculation unit;
In each set, an averaging unit that accumulates and averages the calculation results of the division unit;
A polarization rotation angle calculation unit that calculates an estimated value of the polarization rotation angle of the two polarization signals input to the polarization rotation unit, based on the calculation results of the two sets of the averaging units;
Two polarization signals identical to the two polarization signals input to the polarization rotation unit are input, and input based on the estimated value of the polarization rotation angle output from the polarization rotation angle calculation unit And a polarization rotation unit that performs polarization separation of the two polarization signals by applying reverse rotation to the two polarization signals.
A local oscillation light source that oscillates light at a center wavelength substantially coincident with the received optical signal, an optical interference unit that causes interference between the received optical signal and the light generated by the local oscillation light source, and an output from the optical interference unit An optical receiver having a photoelectric converter for converting into an electrical signal;
An analog / digital converter that converts an electrical signal output from the optical receiver into a digital signal;
A digital signal processing unit that performs polarization separation on the digital signal output from the analog-digital conversion unit,
The digital signal processor is
Four digital signals generated by separating the two polarized signals into four systems are input from the analog / digital converter,
A polarization rotation unit that applies different polarization rotations for each system to the two polarization signals included in each system;
For each of the four systems, a two-polarization signal that is polarized by the polarization rotation unit is input, and one of the two polarization signals and the other of the polarization signals A differential detection section between polarized waves that performs differential detection between
For each of the four systems, a real part acquisition unit that acquires only a real part from the output of the differential detection section between polarizations,
A complex multiplier that performs multiplication of the two real parts included in the set in each of two sets formed as a set for every two systems of the four systems,
In each set, a sum of squares calculation unit that calculates the sum of squares of the real parts of the two systems included in the set;
In each set, a division unit that divides the calculation result of the complex multiplication unit by the calculation result of the square sum calculation unit;
In each set, an averaging unit that accumulates and averages the calculation results of the division unit;
A polarization rotation angle calculation unit that calculates an estimated value of the polarization rotation angle of the two polarization signals input to the polarization rotation unit, based on the calculation results of the two sets of the averaging units;
Two polarization signals identical to the two polarization signals input to the polarization rotation unit are input, and input based on the estimated value of the polarization rotation angle output from the polarization rotation angle calculation unit An optical receiver comprising: a polarization rotation unit that performs polarization separation of the two polarization signals by applying reverse rotation to the two polarization signals.
Two polarization signals are input, the polarization signals are separated into four systems, and a polarization rotation step for applying different polarization rotations for each system to the two polarization signals included in each system,
For each of the four systems, a two-polarization signal to which polarization rotation is given by the polarization rotation step is input, and one of the two polarization signals and the other polarization signal are A differential detection step between polarizations for differential detection between
For each of the four systems, a real part acquisition step for acquiring only a real part from the output of the differential detection step between polarizations;
A complex multiplication step of multiplying the two real parts of the two systems included in the set in each of two sets formed as a set for every two systems of the four systems,
In each set, a square sum operation step of calculating the sum of squares of the real part of the two systems included in the set;
In each set, a division step of dividing the operation result of the complex multiplication step by the operation result of the square sum operation step;
In each set, an averaging step for storing and averaging the calculation results of the division step;
A polarization rotation angle calculation step for calculating an estimated value of the polarization rotation angle of the two polarization signals input in the polarization rotation step based on the calculation results of the two sets of the averaging steps; A polarization estimation method comprising:
Two polarization signals are input, the polarization signals are separated into four systems, and a polarization rotation step for applying different polarization rotations for each system to the two polarization signals included in each system,
For each of the four systems, a two-polarization signal to which polarization rotation is given by the polarization rotation step is input, and one of the two polarization signals and the other polarization signal are A differential detection step between polarizations for differential detection between
For each of the four systems, a real part acquisition step for acquiring only a real part from the output of the differential detection step between polarizations;
A complex multiplication step of multiplying the two real parts of the two systems included in the set in each of two sets formed as a set for every two systems of the four systems,
In each set, a square sum operation step of calculating the sum of squares of the real part of the two systems included in the set;
In each set, a division step of dividing the operation result of the complex multiplication step by the operation result of the square sum operation step;
In each set, an averaging step for storing and averaging the calculation results of the division step;
A polarization rotation angle calculating step for calculating an estimated value of the polarization rotation angle of the two polarization signals input in the polarization rotation step based on the calculation results of the two sets of the averaging steps; Two polarization signals identical to the two polarization signals input in the polarization rotation step are input and input based on the estimated value of the polarization rotation angle output in the polarization rotation angle calculation step. And a polarization rotation step for performing polarization separation of the two polarization signals by applying reverse rotation to the two polarization signals.