JP6673881B2 - Optical transmission characteristic compensation system and optical transmission characteristic compensation method - Google Patents

Description

  The present invention relates to an optical transmission characteristic compensation system and an optical transmission characteristic compensation method for compensating transmission characteristics of an optical transceiver in optical communication.

  In order to cope with an increase in communication traffic, high-speed and large-capacity optical transceivers are required. 2. Description of the Related Art In recent years, optical transceivers that have been increasingly used use digital coherent technology that combines digital signal processing (DSP) and coherent detection.

  In an optical transceiver of 100 Gb / s per channel, the baud rate (baud rate) and modulation method are, for example, 32 Gbaud PDM-QPSK (Polarization Division Multiplexing-Quadrature Phase Shift Keying). The optical transmitter generates a PDM-QPSK optical signal by modulating orthogonal linearly polarized light (X polarization and Y polarization) with a QPSK baseband signal. The optical receiver converts the optical signal into a baseband signal by coherently detecting the received optical signal and local light, and demodulates QPSK by digital signal processing (DSP) to reproduce transmission data.

  In order to increase the transmission capacity per channel, in a 400 Gb / s optical transceiver, the baud rate and modulation method are, for example, 64 Gbaud PDM-16QAM (PDM-16 Quadrature amplitude modulation), or 43Gbaud PDM-64QAM. As described above, in future optical transceivers, the baud rate increases and the modulation scheme becomes multi-valued in order to increase the transmission capacity per channel.

  As the baud rate increases and the modulation scheme becomes multi-valued, an optical transceiver is required to have good transmission characteristics over a wide band. The transmission characteristic of the transmission signal in the optical transceiver is expressed by a transfer function. In general, the optical transceiver includes a plurality of lanes (X-polarized in-phase component XI, X-polarized orthogonal component XQ, Y-polarized in-phase component YI, Since the transfer function difference between the lanes has the orthogonal component YQ of the Y polarization and causes a deterioration in the overall transmission characteristics of the system, it is also required to sufficiently suppress the difference between the transfer functions between the lanes. When the frequency characteristic of the transfer function of the optical transceiver is insufficient or there is a difference between lanes, it is necessary to compensate the transmission characteristic or the difference between the lanes by, for example, a DSP. On the other hand, a method of compensating for chromatic dispersion of an optical transmission line or a difference between lanes on a receiving side on a receiving side (for example, see Non-Patent Documents 1 and 2), and a method for compensating a difference between lanes on a transmitting side on a transmitting side (For example, see Patent Literature 1 and Non-Patent Literature 3).

  When the transfer function of the optical transceiver is compensated by the DSP, it is necessary to grasp in advance the transfer function of the optical circuit or the analog electric circuit constituting the optical transceiver, and to set a compensation value as needed based on them. is there. Each of the optical transmitter and the optical receiver has a frequency characteristic of a transfer function in a range that requires compensation according to the baud rate. When setting a compensation value for compensating the above transfer function in a conventional optical transceiver, for example, a specification value of a transfer function indicated by an optical circuit or analog electric circuit vendor or a previously measured transfer function evaluation result of a representative individual By setting compensation values for the transmitter compensator and the receiver compensator based on the above, sufficient overall transmission characteristics could be obtained.

  However, in a high-speed transmission system such as 400 Gb / s, due to an increase in the baud rate and the increase in the number of values, individual variations in the transfer function of an optical circuit or an analog electric circuit cause the specification value indicated by the vendor or the evaluation result of the representative individual. With the compensation value setting based on this, a sufficient total transmission characteristic may not be obtained. On the other hand, by transmitting and receiving a known pattern (known signal), it becomes possible to grasp a transfer function of an optical circuit or an analog electric circuit constituting an optical transceiver for each individual with high accuracy.

Japanese Patent No. 6077769

RR Muller, J. Renaudier, and Gabriel Charlet, "Blind Receiver Skew Compensation and Estimation for Long-Haul Non-Dispersion Managed Systems Using Adaptive Equalizer", JOURNAL OF LIGHTWAVE TECHNOLOGY, VOL.33, NO.7, pp.1315-1318 , APRIL 1, 2015. J. C.M.Diniz, E. P da Silva, M. Piels, and D. Zibar, "Joint IQ Skew and Chromatic Dispersion Estimation for Coherent Optical Communication Receivers", Advanced Photonics Congress 2016. Ginni Khanna, Bernhard Spinnler, Stefano Calabro, Erik De Man, and Norbert Hanik, "A Robust Adaptive Pre-Distortion Method for Optical Communication Transmitters", IEEE PHOTONICS TECHNOLOGY LETTERS, Vol.28, No.7, pp.752-755, APRIL 1, 2016.

  However, in order to improve the transmission characteristics by the above-described conventional technology, a finite impulse response (FIR) filter corresponding to an inverse transfer function of a transfer function of an optical circuit or an analog electric circuit constituting an optical transceiver is required. Parameter (the number of taps) needs to be increased. As a result, there has been a problem that the circuit scale of the DSP increases, and the manufacturing cost and power consumption of the device increase.

  The present invention has been made in order to solve the above-described problems, and an optical transmission characteristic compensation system and an optical transmission characteristic compensation method capable of performing highly accurate transceiver distortion compensation while suppressing an increase in circuit scale. It is intended to provide.

  One aspect of the present invention is a method for transmitting a first known signal from a transmission unit of an optical transceiver to a reception unit, the first data acquired by the reception unit, and a temporary transfer function of the optical receiver of the reception unit. Or, from a temporary inverse transfer function, a transmitter transfer function estimator for estimating the transfer function or inverse transfer function of the optical transmitter of the transmitter, and a second known signal transmitted from the transmitter to the receiver. A receiver transfer function estimator for estimating the transfer function or inverse transfer function of the optical receiver from the second data obtained by the receiver at the time and the estimated transfer function or inverse transfer function of the optical transmitter. A transmitter compensator that compensates for the transmission characteristics of the optical transmitter using the estimated transfer function or inverse transfer function of the optical transmitter, and the estimated transfer function or inverse transfer function of the optical receiver. A receiver compensator for compensating the transmission characteristics of the optical receiver. At least one of the transmitter compensator and the receiver compensator uses the tap coefficient sequence at which the position where the tap coefficient of the digital filter has the maximum value is shifted from the center position of the tap coefficient sequence. This is an optical transmission characteristic compensation system for compensation.

  Further, one aspect of the present invention is the above-described optical transmission characteristic compensation system, wherein the transmitter compensator is configured such that a position where the tap coefficient has a maximum value is a signal from a center position of the tap coefficient sequence. Compensation is performed using a tap coefficient sequence that is shifted to the tap side that is convolved earlier, and the receiver compensator is configured such that the position at which the tap coefficient has the maximum value is located at the center of the tap coefficient sequence. Compensation is performed using a tap coefficient sequence that is shifted from the position to the tap side where the signal is convolved later.

  Further, one embodiment of the present invention is the above-described optical transmission characteristic compensation system, wherein a shift amount between a center position of the tap coefficient row and a position shifted from the center position of the tap coefficient row is a wiring amount. This is a shift amount calculated based on the length.

  Further, according to one aspect of the present invention, the first data acquired by the receiving unit when the first known signal is transmitted from the transmitting unit of the optical transceiver to the receiving unit, and the temporary data of the optical receiver of the receiving unit. From a transfer function or a provisional inverse transfer function, a first step of estimating a transfer function or an inverse transfer function of the optical transmitter of the transmitting unit, and transmitting a second known signal from the transmitting unit to the receiving unit. A second step of estimating the transfer function or inverse transfer function of the optical receiver from the second data obtained by the receiving unit at times and the estimated transfer function or inverse transfer function of the optical transmitter; When compensating the transmission characteristics of the optical transmitter and the optical receiver using the transfer function or inverse transfer function of the optical transmitter and the transfer function or inverse transfer function of the optical receiver, the tap coefficient of the digital filter Is the maximum value in the tap coefficient sequence. A third step of compensating using the tap coefficient string to be position shifted from the position is a light transmission characteristic compensation method with.

  According to the present invention, highly accurate transceiver distortion compensation can be performed while suppressing an increase in circuit scale.

It is a figure showing the composition of the optical transceiver provided with the optical transmission characteristic compensation system concerning one embodiment of the present invention. 5 is a flowchart illustrating a process of estimating a transfer function or an inverse transfer function of an optical transceiver by the optical transmission characteristic compensation system according to one embodiment of the present invention. 5 is a flowchart illustrating a process of estimating a temporary transfer function of the optical receiver by the optical transmission characteristic compensation system according to the embodiment of the present invention. FIG. 3 is a diagram illustrating a configuration of a first receiver transfer function estimator of the optical transmission characteristic compensation system according to one embodiment of the present invention. FIG. 2 is a diagram illustrating a configuration of a receiver compensator of the optical transmission characteristic compensation system according to the embodiment of the present invention. FIG. 3 is a diagram illustrating a configuration of a transmitter transfer function estimating unit of the optical transmission characteristic compensation system according to one embodiment of the present invention. 5 is a flowchart illustrating a process of estimating a transfer function or an inverse transfer function of an optical transmitter by the optical transmission characteristic compensation system according to one embodiment of the present invention. FIG. 5 is a diagram illustrating a configuration of a second receiver transfer function estimator of the optical transmission characteristic compensation system according to one embodiment of the present invention. 6 is a flowchart illustrating a process of estimating a true transfer function or an inverse transfer function of the optical receiver by the optical transmission characteristic compensation system according to one embodiment of the present invention. FIG. 2 is a diagram illustrating a configuration of an FIR filter configuring a transmitter compensator and a receiver compensator of the optical transmission characteristic compensation system according to one embodiment of the present invention. It is a figure showing the tap coefficient of the usual FIR filter. FIG. 3 is a diagram illustrating tap coefficients of an FIR filter included in a transmitter compensator of the optical transmission characteristic compensation system according to the embodiment of the present invention. FIG. 4 is a diagram illustrating a received Q value improvement amount by the optical transmission characteristic compensation system according to one embodiment of the present invention.

  Hereinafter, an optical transmission characteristic compensation system and an optical transmission characteristic compensation method according to an embodiment of the present invention will be described with reference to the drawings. The same or corresponding components are denoted by the same reference numerals, and description thereof may not be repeated. Note that the term “transfer function” used below may be any function, expression, circuit, line, or the like that represents transmission characteristics of devices, components, propagation paths, and the like. Further, the transfer function is not limited to a linear function, and may be a function representing a non-linear characteristic. Further, “transmission” and “transmission” are basically regarded as consent within the scope of the present invention.

<Embodiment>
FIG. 1 is a diagram illustrating a configuration of an optical transceiver including an optical transmission characteristic compensation system according to an embodiment of the present invention. The transmission unit 1 transmits an optical signal to the reception unit 3 via the transmission path 2. The transmission path 2 includes, for example, an optical fiber and an optical amplifier.

  The transmission unit 1 includes a transmission signal processing unit 4, a known signal insertion unit 5, a transmitter compensation unit 6, and an optical transmitter 7. Part or all of the transmission signal processing unit 4, the known signal insertion unit 5, and the transmitter compensation unit 6 may be a hardware such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array). Can be configured with hardware. Further, some or all of them can be configured by software that functions by a processor such as a CPU (Central Processing Unit) executing a program stored in a storage unit.

  The known signal insertion unit 5 modulates the XI lane (first lane), XQ lane (second lane), YI lane (third lane), and YQ lane (fourth lane) generated by the transmission signal processing unit 4. Each known signal sequence is inserted into the sequence. The sequence of the known signal is shared between the transmitting unit 1 and the receiving unit 3. The known signal can be composed of predetermined bits or symbols, and for example, is composed of a signal sequence of about 2000 symbols. The sequence length of the known signal is required to be at least longer than the calculated FIR filter length.

  The transmission signal processing unit 4 generates frame data based on the transmission data sequence. The frame data is a signal sequence (modulation target signal sequence) for performing a modulation process in the optical transmitter 7. The transmission signal processing unit 4 transmits the frame data into which the known signal sequence has been inserted to the transmitter compensation unit 6.

  The transmitter compensator 6 obtains a transfer function estimation result of the optical transmitter 7 from a transmitter transfer function estimator 8 of the receiver 3 described later. The transmitter compensator 6 compensates the transfer functions of the XI lane, XQ lane, YI lane, and YQ lane of the optical transmitter 7 and the difference between the lanes based on the estimation result. The transmitter compensator 6 can be constituted by, for example, a digital filter such as an FIR filter. Further, the transmitter compensator 6 may include a functional unit having a function of individually compensating for a delay time difference between the four lanes.

  The optical transmitter 7 generates an optical signal of a signal sequence to be modulated by modulating the orthogonal change orthogonal with the compensated frame data. The optical transmitter 7 includes a driver amplifier 7a, a laser module 7b (signal LD), a 90 ° combiner 7c, and a polarization combiner 7d. The driver amplifier 7a amplifies the compensated electric signal of the frame data so as to have an appropriate amplitude and transmits the amplified electric signal to the 90 ° combiner 7c. The 90 ° combiner 7c is a Mach-Zehnder type vector modulator, and separates linearly polarized CW (Continuous Wave) light transmitted from the laser module 7b into orthogonal linearly polarized light, and separates the linearly polarized light into respective linearly polarized light. On the other hand, an optical signal of a signal sequence to be modulated is generated by modulating with frame data. The optical signal based on the horizontal polarization and the optical signal based on the vertical polarization are combined by the polarization combiner 7d and supplied to the receiving unit 3 via the transmission line 2.

  The receiving unit 3 includes an optical receiver 9, a data buffer 10, a receiver compensating unit 11, a received signal processing unit 12, first and second receiver transfer function estimating units 13 and 14, and a transmitter transfer function estimating unit 8. Is provided. The first and second receiver transfer function estimators 13 and 14 and the transmitter transfer function estimator 8 constitute an optical transmission characteristic estimation system for estimating the optical transmission characteristics of the optical transceiver. The optical transmission characteristic estimating system, the transmitter compensator 6 and the receiver compensator 11 constitute an optical transmission characteristic compensating system for compensating the optical transmission characteristics of the optical transceiver. In FIG. 1, the transmitter compensator 6 and the receiver compensator 11 are represented by separate blocks. However, the transmitter compensator 6 may be a part of the transmission signal processor 4, The unit 11 may be a part of the reception signal processing unit 12.

  The optical receiver 9 includes a polarization separator 9a, a laser module 9b (local LD), a polarization diversity 90 ° hybrid 9c, a photodiode (PD; Photo Diode) (not shown), a TIA 9d (Transimpedance Amplifier; transimpedance). Amplifier) and an A / D converter 9e.

  The laser module 9b sends the linearly polarized CW light to the polarization diversity 90 ° hybrid 9c. The polarization diversity 90 ° hybrid 9c causes the received optical signal to interfere with the CW light. A photodiode converts it photoelectrically. The TIA 9d converts the current signal into a voltage signal. A / D converter 9e A / D converts the voltage signal. As a result, the received optical signal is converted into a baseband digital signal.

  A / D converter 9e of optical receiver 9, data buffer 10, receiver compensator 11, received signal processor 12, first and second receiver transfer function estimators 13, 14, and transmitter transfer function estimation Part or all of the unit 8 can be configured by hardware such as an ASIC or an FPGA. Further, some or all of them can be configured by software that functions when a processor such as a CPU executes a program stored in a storage unit. The first and second receiver transfer function estimators 13 and 14 and the transmitter transfer function estimator 8 correspond to an external device independent of the optical transceiver, for example, a PC (Personal Computer) or a personal computer. It can be configured by a device. In addition, the received signal processing unit 12 can have the same function as the first and second receiver transfer function estimating units 13 and 14 and the transmitter transfer function estimating unit 8, and can be shared with them. is there.

  The data buffer 10 can be generally configured by a memory circuit (RAM (Random Access Memory; readable / writable memory)), and temporarily stores data obtained by A / D converting a signal received by the optical receiver 9. . The data stored in the data buffer 10 is sequentially sent to a receiver compensation unit 11 and a reception signal processing unit 12 at the subsequent stage. The data can be obtained by the first and second receiver transfer function estimating units 13 and 14 and the transmitter transfer function estimating unit 8. Note that the first and second receiver transfer function estimators 13 and 14 and the transmitter transfer function estimator 8 directly obtain the A / D converted data in real time without using the data buffer 10. Is also good. Hereinafter, all the examples described using the digital data of the data buffer 10 also include a method of directly acquiring the received data in real time.

  The receiver compensating unit 11 obtains the estimation result of the transfer function of the optical receiver 9 from the second receiver transfer function estimating unit 14, and based on the estimation result, the XI lane, the XQ lane, and the YI lane of the optical receiver 9. The transfer function of the lane and the YQ lane and the difference between the lanes are compensated. The receiver compensator 11 can be configured by a digital filter such as an FIR filter. Further, the receiver compensating unit 11 may have a functional unit having a function of individually compensating for a delay time difference between the four lanes.

  A digital signal is input from the receiver compensation unit 11 to the reception signal processing unit 12. In the transmission line 2, waveform distortion occurs in the optical signal due to, for example, chromatic dispersion, polarization mode dispersion, polarization fluctuation, or nonlinear optical effect. The reception signal processing unit 12 compensates for waveform distortion generated in the transmission path 2. Further, the reception signal processing unit 12 compensates for a difference between the frequency of light of the laser module 7b of the optical transmitter 7 and the frequency of local light of the laser module 9b of the optical receiver 9. Further, the reception signal processing unit 12 compensates for phase noise according to the line width of light of the laser module 7b of the optical transmitter 7 and the line width of local light of the laser module 9b of the optical receiver 9.

  The first receiver transfer function estimating unit 13 converts the digital data acquired by the receiving unit 3 when an ASE (Amplified Spontaneous Emission) signal corresponding to white noise is input to the input end of the optical receiver 9 from the optical receiver. Nine tentative transfer functions or inverse transfer functions are estimated. The ASE signal can be generated from an optical amplifier. When outputting only ASE, an optical amplifier is used in a state where nothing is input. Although this optical amplifier may be separately prepared, the optical amplifier of the transmission line 2 may be used. Since the spectrum (frequency characteristic) of the ASE signal is uniform, the frequency characteristic can be obtained by passing the signal through the spectrum. Therefore, when the first receiver transfer function estimating unit 13 acquires the data stored in the data buffer 10 in a state where the ASE signal is input, the frequency characteristics can be estimated. These can be estimated for each lane. A configuration example of the first receiver transfer function estimating unit 13 will be described later.

  The estimation of the frequency characteristic is obtained as a transfer function by performing a Fourier transform on the digital data. Further, as a method of obtaining an inverse transfer function, there is a method of obtaining a solution of an adaptive filter in addition to calculating an inverse number. As a method of obtaining a solution of an adaptive filter, a method of generally obtaining a Wiener solution, a method of obtaining an LMS (Least Mean Square) algorithm or an RLS (Recursive Least Square) algorithm, or the like. There is. Here, since the transfer function does not change relatively with time, “adaptation” does not mean a temporal correspondence. Hereinafter, "adaptation" means adaptation to a feedback circuit for obtaining a convergent solution. A detailed configuration example of the first receiver transfer function estimator 13 will be described later. In the above description, the ASE signal is used. However, the present invention is not limited to the ASE signal, and any test signal having a known spectrum can be used.

  The transmitter transfer function estimating unit 8 transmits the first digital data (first data) acquired by the receiving unit 3 when the first known signal is transmitted from the transmitting unit 1 to the receiving unit 3, and the light of the receiving unit 3. The transfer function or the inverse transfer function of the optical transmitter 7 is estimated from the temporary transfer function or the inverse transfer function of the receiver 9. As an estimation method, for example, the transfer function of the optical transmitter 7 is estimated using an adaptive filter. The adaptive filter is, for example, a filter based on the LMS algorithm or a filter based on the RMS algorithm.

  The second receiver transfer function estimator 14 estimates the second digital data (second data) obtained by the receiver 3 when the second known signal is transmitted from the transmitter 1 to the receiver 3. From the transfer function or the inverse transfer function of the optical transmitter 7, the true transfer function or the inverse transfer function of the optical receiver 9 is estimated. As an estimation method, an inverse transfer function of the optical receiver 9 is estimated using, for example, an adaptive filter. The adaptive filter is, for example, a filter based on the LMS algorithm or a filter based on the RLS algorithm. Also in this case, estimation can be performed for each lane.

  Subsequently, a method for estimating the optical transmission characteristics of the optical transceiver by the optical transmission characteristics estimation system according to the present embodiment will be described with reference to the drawings. FIG. 2 is a flowchart showing a process for estimating the transfer function or inverse transfer function of the optical transceiver by the optical transmission characteristic compensation system according to one embodiment of the present invention.

First, the first receiver transfer function estimator 13 estimates a provisional transfer function or inverse transfer function of the optical receiver 9 (step S1).
Next, the transmitter transfer function estimator 8 estimates the transfer function or inverse transfer function of the optical transmitter 7 (step S2).
Next, the second receiver transfer function estimator 14 estimates the true transfer function or inverse transfer function of the optical receiver 9 (step S3).

  Next, a detailed operation of each step will be described. FIG. 3 is a flowchart illustrating a process of estimating a temporary transfer function of the optical receiver by the optical transmission characteristic compensation system according to the embodiment of the present invention.

First, an ASE signal is inserted into the input of the optical receiver 9 (Step S101).
Since it is known that the spectrum of the ASE signal is uniform, it is possible to obtain the frequency characteristic by passing the signal. Next, while the ASE signal is being input, the data buffer 10 acquires the received data (Step S102).
Next, the first receiver transfer function estimator 13 acquires digital data from the data buffer 10 and performs FFT processing to acquire a temporary transfer function (step S103).
Next, a temporary inverse transfer function is calculated from the obtained temporary transfer function (step S104).
Next, the calculated tentative inverse transfer function is set in the receiver compensator 11 (step S105).

  FIG. 4 is a diagram illustrating a configuration of a first receiver transfer function estimator of the optical transmission characteristic compensation system according to one embodiment of the present invention. The first receiver transfer function estimating unit 13 performs FFT (Fast Fourier Transform) processing on the X-polarized received signal and the Y-polarized received signal, and performs 1 / transfer function processing on their outputs. A circuit for calculating an inverse transfer function. Although the received signal of X polarization is XI + jXQ and the received signal of Y polarization is YI + jYQ, it is assumed that there is no delay difference between XI and XQ and between YI and YQ. If there is a delay difference, XI, XQ, YI, and YQ can be individually subjected to Fourier transform and 1 / transfer function processing. It is not necessary to limit to FFT processing as long as Fourier transform can be performed, and another method may be used. The notation “FFT” hereinafter refers to the function of the Fourier transform.

  Since the digital data acquired by the data buffer 10 is data in the time domain, the data is converted into data in the frequency domain by FFT processing on the X polarization lane and the Y polarization lane, respectively.

X _R (n) is digital data acquired by the data buffer 10, and X _R (k) is data subjected to FFT processing. FFT means high-speed processing of DFT (Discrete Furrier Transfer; discrete Fourier transform). In the general FFT processing for a continuous signal, the processing is performed for each finite number of data N. However, it is needless to say that the processing is performed by overlapping data between adjacent processings (overlap Add, overlap Save, etc.). Method). The same applies to the subsequent FFT processing. The absolute value of X _R (k) indicates amplitude information, which is obtained as a temporary transfer function. By calculating the reciprocal thereof, a temporary inverse transfer function can be obtained. This inverse transfer function can be set in the receiver compensator 11. The inverse transfer function is also used when estimating the transfer function of the optical transmitter 7. In this case, it is not always necessary to set the inverse transfer function of the temporary transfer function in the receiver compensator 11.

  FIG. 5 is a diagram illustrating a configuration of a receiver compensator of the optical transmission characteristic compensation system according to one embodiment of the present invention. The receiver compensator 11 performs IQ vector processing (time domain processing). That is, the X polarization and the Y polarization are expressed as complex vector signals as XI + jXQ and YI + jYQ, respectively, and the transmission characteristics are compensated by the FIR filter. The inverse transfer function calculated in step S104 is converted into a time response signal by IFFT processing (not shown), and is set as a tap coefficient of an FIR filter.

  Note that the receiver compensator 11 is not limited to the above-described configuration, and may have any configuration as long as the transfer function can be compensated. The compensation in the frequency domain by the receiver compensator 11 is represented by the following equation. However, it goes without saying that compensation can be performed using an FFT filter in the time domain by IFFT.

Here, X _out and Y _out are data after compensation of X _in = XI + jXQ and Y _in = YI + jYQ, respectively. XI _out , XQ _out , YI _out , and YQ _out are data after compensation of XI _in , XQ _in , YI _in , and YQ _in , respectively. H1 to H16 indicate the inverse transfer functions in that case.

  Although not shown, IQ vector processing and IQ individual processing can be combined. For example, as described below, after a filtering process is once performed by a complex filter as an IQ vector, a real part and an imaginary part are divided, and filtering is performed on each of them by a real filter.

・ X * complex filter ⇒ XI * real filter, XQ * real filter (tap coefficient can be set individually)
・ Y * complex filter ⇒ YI * real filter, YQ * real filter (tap coefficient can be set individually)

  Here, X and Y represent complex vector representations, XI, XQ, YI, and YQ represent real numbers, respectively, "*" represents a process, and "?" Represents a process flow.

  For complex filtering of a complex signal such as X or Y, the circuit scale can be reduced by performing batch processing (FFT → transfer function multiplication → IFFT) in the frequency domain as compared to processing individually. In addition, the real number filtering is more efficient in terms of circuit scale when it is processed by processing in the time domain (FIR filter (convolution operation)). As described above, the frequency characteristics and the delay difference can be compensated.

  FIG. 6 is a diagram illustrating a configuration of a transmitter transfer function estimating unit of the optical transmission characteristic compensation system according to one embodiment of the present invention. The transmitter transfer function estimating unit 8 includes a known signal synchronizing unit 8a, various transmission characteristic compensating units 8b, a receiver compensating unit 8c, and an adaptive filter having an FIR filter 8d and a square error minimizing unit 8e. The various transmission characteristic compensators 8b include various compensation circuits for compensating for transmission distortion such as chromatic dispersion compensation, frequency offset compensation, polarization dispersion / polarization rotation compensation, clock phase compensation, and phase noise compensation. The known signal synchronizer 8a has a function of extracting a known signal from digital data, and estimates compensation data to be set for various transmission characteristic compensations at a later stage from various extracted blocks based on the state of the extracted known signal. That is, the estimation of the transfer function or the inverse transfer function of the optical transmitter 7 includes a process of estimating the transmission characteristics of the transmission path 2. It should be noted that the receiver compensating section 8c can be arranged before the various transmission characteristic compensating sections 8b.

  The chromatic dispersion compensator can be arranged at a stage prior to the known signal synchronizer 8a. The order of the compensators of the various transmission characteristic compensators can be interchanged. The meaning of (1 TAP 2 × 2 MIMO (Multi Input Multi Output)) of the polarization dispersion / polarization rotation compensation means that the tap number of the filter is set to 1 and the band characteristics of the optical transceiver are not compensated by this block. This indicates that only the polarization rotation is performed (the band is compensated in a general multi-tap 2 × 2 MIMO filter).

  Further, the transmitter transfer function estimator 8 processes each of the X polarization and the Y polarization as a complex vector signal similarly to the first receiver transfer function estimator 13 of FIG. It is also possible to independently process each lane of XQ, YI, and YQ. In this case, it is possible to extract and compensate for the delay difference between lanes. Processing the X polarization as a complex vector signal assumes that the delay difference (Skew; skew) between XI and XQ is zero. If the delay difference cannot be ignored, it is necessary to extract and compensate for the transfer function for each lane. The same applies to the Y polarization.

  FIG. 7 is a flowchart showing a process for estimating the transfer function or inverse transfer function of the optical transmitter by the optical transmission characteristic compensation system according to one embodiment of the present invention.

First, a known signal is input to the input of the transmission signal processing unit 4, and an optical modulation signal is transmitted from the optical transmitter 7 (step S201).
At this time, the transmitter compensator 6 bypasses. Note that the transmitter compensator 6 can have the same configuration as the receiver compensator 11 shown in FIG.
Next, reception data is acquired by the data buffer 10 on the reception side (step S202).
Next, the transmitter transfer function estimator 8 acquires digital data from the data buffer 10 (step S203). The known signal synchronizer 8a extracts a known signal from the acquired digital data. Compensation of various transmission characteristics and optical receiver compensation are performed on the extracted known signal. The optical receiver compensation is performed using the temporary inverse transfer function of the optical receiver 9 estimated in step S1. FIG. 6 shows a configuration in which the receiver compensator 11 compensates with the provisional inverse transfer function at the subsequent stage of the data buffer 10, but this compensation is not particularly necessary for the processing of the transmitter transfer function estimator 8 described above. .

  The effect of the transfer function of the optical transmitter 7 remains in the known signal that has been subjected to various transmission characteristic compensation and optical receiver compensation. Therefore, the FIR filter 8d in which the inverse characteristic is set is applied to the signal as an adaptive filter, and the inverse characteristic is corrected again so that the square of the difference between the output and the known signal is minimized. By this processing, the tap coefficient of the FIR filter 8d constituting the adaptive filter can be obtained as the time response of the inverse transfer function. The method of obtaining the inverse transfer function is generally known as a Wiener solution or an LMS algorithm described below.

  Here, s (n) is a known signal, y (n) is an output of the adaptive filter, e (n) is a difference between s (n) and y (n), and h (n) is a time response of the adaptive filter. .

  In the above example, since the inverse transfer function of the optical transmitter 7 can be directly obtained by the adaptive equalization circuit, step S203 and step S204 can be processed integrally. On the other hand, once the transfer function of the optical transmitter 7 is obtained, the inverse transfer function is calculated (step S204).

  Next, the estimated inverse transfer function of the optical transmitter 7 is set in the transmitter compensator 6 (step S205). The setting method is the same as the method shown in step S105. At this time, as described above, it is also possible to independently process each lane of XI, XQ, YI, and YQ. In this case, it is possible to extract and compensate for the delay difference between lanes.

  FIG. 8 is a diagram illustrating a configuration of a second receiver transfer function estimator of the optical transmission characteristic compensation system according to one embodiment of the present invention. The second receiver transfer function estimating unit 14 simulates a known signal synchronization unit 14a, chromatic dispersion compensation, frequency offset compensation, polarization dispersion / polarization rotation addition, clock phase addition, phase noise addition, and other transmission distortion. Circuit 14b, an FIR filter 14c for adaptive equalization, and a square error minimizing circuit 14d. The known signal synchronizer 14a has a function of extracting a known signal from digital data, and estimates additional data to be set in a circuit that simulates distortion at a subsequent stage from various states of the extracted known signal by various estimation blocks. That is, the estimation of the transfer function or the inverse transfer function of the optical receiver 9 includes a process of estimating the transmission characteristics of the transmission path 2. Note that the order of the circuits 14b for simulating distortion during transmission such as chromatic dispersion compensation, frequency offset compensation, polarization dispersion / polarization rotation addition, clock phase addition, and phase noise addition can be changed.

  Although the second receiver transfer function estimator 14 processes each of the X polarization and the Y polarization as a complex vector signal, as in the case of the first receiver transfer function estimator 13 in FIG. , XI, XQ, YI, and YQ, respectively. In this case, it is possible to extract and compensate for the delay difference between lanes. Processing the X polarization as a complex vector signal assumes that the delay difference between XI and XQ is zero. If the delay difference cannot be ignored, it is necessary to extract and compensate for the transfer function for each lane. The same applies to the Y polarization.

  FIG. 9 is a flowchart showing a process of estimating the true transfer function or the inverse transfer function of the optical receiver by the optical transmission characteristic compensation system according to one embodiment of the present invention.

  First, a known signal is input to the input of the transmission signal processing unit 4, and an optical modulation signal is transmitted from the optical transmitter 7 of the transmission unit 1 to the reception unit 3 (Step S301). At this time, the inverse transfer function of the optical transmitter 7 estimated in step S2 of FIG. 2 is set in the transmitter compensator 6, and the transmission characteristics of the optical transmitter 7 are compensated. Note that the transmitter compensator 6 can have the same configuration as the receiver compensator 11 shown in FIG.

Next, on the receiving side, the received data is acquired by the data buffer 10 (step S302).
The second receiver transfer function estimating unit 14 acquires digital data from the data buffer 10 (Step S303). The known signal synchronizer 14a extracts a known signal from the acquired digital data. The extracted known signal is supplied to an FIR filter 14c as an adaptive filter. On the other hand, chromatic dispersion, frequency offset, polarization dispersion / polarization rotation, clock phase, and phase noise estimated as transmission line distortion are added to the known signal, and compared with the output of the adaptive filter. Chromatic dispersion, frequency offset, polarization dispersion / polarization rotation, clock phase, and the amount of phase noise added are estimated by various estimation blocks from the state of the known signal.

  Here, at the output of the adaptive filter, the transfer function of the optical transmitter 7 is regarded as being compensated by the transmitter compensator 6. If the transfer function of the optical receiver 9 is compensated by the adaptive filter, the output of the adaptive filter is affected only by the transmission line distortion. This signal is compared with a known signal to which transmission line distortion is added, and the difference (square error) is minimized, so that the tap coefficient of the FIR filter 14c, which is an adaptive filter, is calculated by the inverse transfer function of the optical receiver 9. Can be obtained as the time response of The method of obtaining the inverse transfer function is generally known as a Wiener solution or an LMS algorithm described below.

  Here, d (n) is a known signal, y (n) is an output of the adaptive filter, e (n) is a difference between d (n) and y (n), and h (n) is a time response of the adaptive filter. .

  In the above example, since the true inverse transfer function of the optical transmitter 7 can be directly obtained by the adaptive equalization circuit, step S303 and step S304 can be integrally processed. On the other hand, when a true transfer function of the optical receiver 9 is obtained, a true inverse transfer function is calculated from the transfer function (step S304).

  Next, the estimated true inverse transfer function of the optical receiver 9 is set in the receiver compensator 11 (step S305). The setting method is the same as the method shown in step S105. At this time, as described above, it is also possible to independently process each lane of XI, XQ, YI, and YQ. In this case, it is possible to extract and compensate for the delay difference between lanes.

  With the configuration described above, the transfer function or inverse transfer function of the optical transmitter 7 and the transfer function or inverse transfer function of the optical receiver 9 can be estimated. That is, the transmission characteristics of the optical transmitter 7 and the optical transmitter 9 can be estimated. By setting those transfer functions or inverse transfer functions in the transmitter compensator 6 and the receiver compensator 11, the transfer function in the optical transmitter 7 and the transfer function in the optical receiver 9 can be individually compensated.

  In order to improve the transmission characteristics by the configuration of the optical transmission characteristic compensation system described above, conventionally, as described above, an FIR corresponding to an inverse transfer function of a transfer function of an optical circuit or an analog electric circuit constituting an optical transceiver is used. It was necessary to increase the filter parameters (the number of taps).

  The optical transmission characteristic compensating system according to the present embodiment generally has a parameter sequence in which the maximum value of the compensation parameter (tap coefficient) is located at the center of the tap coefficient sequence, while the position of the maximum value is intentionally shifted. have. By displacing the tap center, it is possible to compensate for signal distortion due to the influence of reflection of electric wiring and the like, and to achieve higher accuracy compensation despite a small circuit scale.

  FIG. 10 is a diagram showing a configuration of an FIR filter constituting a transmitter compensator and a receiver compensator of the optical transmission characteristic compensation system according to one embodiment of the present invention. As shown in the figure, in the present embodiment, the tap center of the FIR filter for compensating distortion is shifted. Here, the tap center is the maximum value of the tap coefficient of the FIR filter.

Specifically, the center of the tap is shifted to the left on the transmitter compensator 6 side, and the tap center is shifted to the right on the receiver compensator 11 side. In FIG. 10, X (t) is a signal before compensation, h (t−τ) is a tap coefficient, τ is the amount of time to be shifted from the tap center, Z− ¹ is a block corresponding to the delay amount τ, and Y (t) is This is the signal after compensation.

Hereinafter, the distortion compensating FIR filter on the transmitter compensating unit 6 side will be described in more detail as an example. In the case of the transmitting side, the tap on the right side of the tap center (right tap) compensates for distortion due to reflection between devices and the like, so that it is necessary to lengthen the tap in order to perform highly accurate compensation. Incidentally, as shown in FIG. 10, right side tap is n value is greater than the taps in h _n (taps convolved more later to the signal), whereas the left tap its opposite h _n Are the taps with smaller values of n (the taps that are convolved earlier for the signal). When FIR filters having the same tap length are used, the compensation effect can be maximized by shifting the center of each tap on the transmitting and receiving sides to an optimum position.

  FIG. 11 is a diagram showing tap coefficients of a normal FIR filter. The vertical axis of the graph shown in FIG. 11 represents a tap coefficient, and the horizontal axis represents a tap position. As shown, in a normal FIR filter, the peak value (tap center) of the tap coefficient is at the center of the tap coefficient sequence (hereinafter, also referred to as “tap center”).

  FIG. 12 is a diagram illustrating tap coefficients of an FIR filter included in a transmitter compensator of the optical transmission characteristic compensation system according to one embodiment of the present invention. The vertical axis of the graph shown in FIG. 12 represents a tap coefficient, and the horizontal axis represents a tap position. As shown in the figure, the FIR filter constituting the transmitter compensator 6 of the optical transmission characteristic compensating system according to the present embodiment is configured to shift the peak value of the tap coefficient (tap center) to the left from the center of the tap. Note that the FIR filter included in the receiver compensator 11 of the optical transmission characteristic compensation system according to the present embodiment is configured to shift the peak value of the tap coefficient (tap center) to the right from the center of the tap.

  In addition, as a method of determining the amount by which the center of the tap is shifted from the center of the tap (shift amount), for example, a method of sweeping the shift amount and determining the shift amount at a position where the transmission characteristic is the best can be considered. Alternatively, a method of determining a range in which compensation by reflection is performed by theoretical calculation based on the wiring length, and determining a shift amount based on the range may be considered.

  As a specific method of shifting the tap center, for example, a method of calculating a filter in a state where a time delay is previously given to the reference signal sequence s (n) as s (n−τ) or the like in Equation 3 described above. And so on. Here, τ is the amount of time to shift from the center of the tap.

  FIG. 13 is a diagram showing the amount of improvement in the reception Q value by the optical transmission characteristic compensation system according to one embodiment of the present invention. In the graph shown in FIG. 13, the vertical axis represents the received Q value improvement amount (unit: dB (decibel)), and the horizontal axis represents the position of the tap center. The graph shown in FIG. 13 shows the amount of improvement of the reception Q value improved when the FIR filter according to the present embodiment (with the tap center shifted) is used for a normal (with the tap center not shifted) FIR filter. Is a result of an experiment in which is shown.

  In the above experiment, since a 65-tap FIR filter was used, the position of the center of the tap was about 33 on the horizontal axis. Since this position corresponds to the position of the center of the tap of the normal FIR filter (the center of the tap is not shifted), as shown in FIG. 13, the received Q value improvement amount is 0 dB (r1 in FIG. 13).

  As illustrated in FIG. 13, when the center of the tap is shifted from the center of the tap and is moved closer to the position 60 than the position 50, the reception Q value improvement amount exceeds 0.8 dB. As described above, when the FIR filter according to the present embodiment (with the tap center shifted) is used for an ordinary (with the tap center shifted) FIR filter, an experimental result that the received Q value is improved is shown. Obtained.

  As described above, in the optical transmission characteristic compensation system according to the present embodiment, the maximum value of the compensation parameter (tap coefficient) is usually at the center of the tap coefficient sequence, but the position of the maximum value is intentionally set. Has a parameter series that shifts. By displacing the tap center, it is possible to compensate for signal distortion due to the influence of reflection of electric wiring and the like, and to achieve higher accuracy compensation despite a small circuit scale.

  As a result, the optical transmission characteristic compensation system and optical transmission characteristic compensation method according to the present embodiment can perform highly accurate transceiver distortion compensation while suppressing an increase in circuit scale. Further, with this, the optical transmission characteristic compensation system and the optical transmission characteristic compensation method according to the present embodiment can reduce the number of taps of the FIR filter to be used and simplify the configuration of the FIR filter. , And high-precision optical transceiver distortion compensation with low power consumption can be realized.

  As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to the embodiments and includes a design and the like within a range not departing from the gist of the present invention.

Some functions of the optical transmission characteristic compensation system in the above-described embodiment may be realized by a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read and executed by a computer system.
Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, and a CD-ROM, and a storage device such as a hard disk built in a computer system.
Further, a "computer-readable recording medium" refers to a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line, and dynamically holds the program for a short time. Such a program may include a program that holds a program for a certain period of time, such as a volatile memory in a computer system serving as a server or a client in that case. Further, the above-mentioned program may be for realizing a part of the above-mentioned functions, and may be for realizing the above-mentioned functions in combination with a program already recorded in the computer system.

REFERENCE SIGNS LIST 1 transmitter, 2 transmission path, 3 receiver, 6 transmitter compensator, 7 optical transmitter, 8 transmitter transfer function estimator, 9 optical receiver, 11 receiver compensator, 13 first receiver transfer function Estimator, 14 second receiver transfer function estimator

Claims (4)

The first data obtained by the receiving unit when the first known signal is transmitted from the transmitting unit of the optical transceiver to the receiving unit by bypassing the transmitter compensating unit, and the input terminal of the optical receiver of the receiving unit. From the provisional transfer function or the provisional inverse transfer function estimated from the digital data obtained by the reception unit when a signal with a known spectrum is input, the transfer function or inverse transfer function of the optical transmitter of the transmission unit is calculated. A transmitter transfer function estimator for estimating,
From the transmitting unit to the receiving unit, the second data acquired by the receiving unit when transmitting the second known signal compensated by the inverse transfer function of the transmitting unit by the transmitter compensating unit, and the estimated From the transfer function or the inverse transfer function of the optical transmitter, a receiver transfer function estimator for estimating the transfer function or the inverse transfer function of the optical receiver,
When the inverse transfer function of the optical transmitter is set by the transmitter transfer function estimating unit, transmitter compensation for compensating the transmission characteristics of the optical transmitter using the transfer function or inverse transfer function of the optical transmitter. Department and
A receiver compensating unit that compensates for the transmission characteristics of the optical receiver using the estimated transfer function or inverse transfer function of the optical receiver,
With
At least one of the transmitter compensator and the receiver compensator is compensated by using the tap coefficient sequence in which the position where the tap coefficient of the digital filter has the maximum value is shifted from the center position of the tap coefficient sequence. And
When the transmitter compensator compensates, the shifted position is a position shifted in a first direction from a center position of the tap coefficient sequence,
When the receiver compensator compensates, the shifted position is a position shifted from a center position of the tap coefficient sequence in a second direction opposite to the first direction. Characteristic compensation system. The transmitter compensator, a position where the tap coefficient has the maximum value, a tap coefficient sequence that is shifted from the center position of the tap coefficient sequence to the tap side where the signal is convolved earlier. Compensation using
The receiver compensating unit uses a tap coefficient sequence in which a position where the tap coefficient has the maximum value is shifted from a center position of the tap coefficient sequence to a tap side where the signal is convolved later. The optical transmission characteristic compensation system according to claim 1. The shift amount between the center position of the tap coefficient sequence and a position shifted from the center position of the tap coefficient sequence is a shift amount calculated based on a wiring length. Optical transmission characteristics compensation system. The first data obtained by the receiving unit when the first known signal is transmitted from the transmitting unit of the optical transceiver to the receiving unit by bypassing the transmitter compensating unit, and the input terminal of the optical receiver of the receiving unit. From the provisional transfer function or the provisional inverse transfer function estimated from the digital data obtained by the reception unit when a signal with a known spectrum is input, the transfer function or inverse transfer function of the optical transmitter of the transmission unit is calculated. A first step of estimating;
From the transmitting unit to the receiving unit, the second data acquired by the receiving unit when transmitting the second known signal compensated by the inverse transfer function of the transmitting unit by the transmitter compensating unit, and the estimated A second step of estimating a transfer function or an inverse transfer function of the optical receiver from a transfer function or an inverse transfer function of the optical transmitter;
When the inverse transfer function of the optical transmitter is set by the first step, the light is transmitted using the transfer function or inverse transfer function of the optical transmitter and the transfer function or inverse transfer function of the optical receiver. When compensating the transmission characteristics of the transmitter and the optical receiver, the position where the tap coefficient of the digital filter is the maximum value is compensated using the tap coefficient sequence that is a position shifted from the center position of the tap coefficient sequence ,
When compensating for the transmission characteristics of the optical transmitter, the shifted position is a position shifted in a first direction from a center position of the tap coefficient sequence,
When compensating for the transmission characteristics of the optical receiver, the shifted position is a position shifted from a center position of the tap coefficient sequence in a second direction opposite to the first direction. The third step;
An optical transmission characteristic compensation method comprising:

Images