WO2012108421A1 - デジタル信号処理装置 - Google Patents
デジタル信号処理装置 Download PDFInfo
- Publication number
- WO2012108421A1 WO2012108421A1 PCT/JP2012/052724 JP2012052724W WO2012108421A1 WO 2012108421 A1 WO2012108421 A1 WO 2012108421A1 JP 2012052724 W JP2012052724 W JP 2012052724W WO 2012108421 A1 WO2012108421 A1 WO 2012108421A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- signal
- filter
- input
- delay
- phase
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/06—Receivers
- H04B1/10—Means associated with receiver for limiting or suppressing noise or interference
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/60—Receivers
- H04B10/61—Coherent receivers
- H04B10/616—Details of the electronic signal processing in coherent optical receivers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03178—Arrangements involving sequence estimation techniques
- H04L25/03248—Arrangements for operating in conjunction with other apparatus
- H04L25/03254—Operation with other circuitry for removing intersymbol interference
- H04L25/03261—Operation with other circuitry for removing intersymbol interference with impulse-response shortening filters
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H21/00—Adaptive networks
- H03H21/0012—Digital adaptive filters
- H03H2021/0085—Applications
- H03H2021/0092—Equalization, i.e. inverse modeling
Definitions
- the present invention relates to a digital signal processing device that is included in an optical signal receiver for optical fiber communication and that processes a digital signal converted from an optical signal by a photoelectric conversion device and an analog-digital converter.
- Non-Patent Documents 1 to 3 In the digital coherent transmission technology that coherently detects a phase-modulated polarization multiplexed optical signal and restores the signal by digital signal processing, the transmission capacity per wavelength channel and the frequency utilization efficiency are dramatically increased (for example, Non-Patent Documents 1 to 3).
- the digital signal processing circuit generally includes an adaptive equalization filter.
- Typical control algorithms for the adaptive equalization filter include CMA (Constant Modulus Algorithm) and DD-LMS (Decision-Directed Least Mean Squared Algorithm). These control algorithms realize the adaptive operation of the equalization filter by calculating in real time a transfer function that compensates for waveform distortion caused by the transmission path using known information about the signal waveform (for example, non-patented). Reference 4).
- MLSE maximum likelihood sequence estimation
- the adaptive equalization filter described above has only a limited effect with respect to the compensation of the band-limited distortion when performing a strong band narrowing below the Nyquist limit.
- Compensation by adaptive equalization of band-limited distortion is an operation for shaping the spectrum of a signal in the frequency domain.
- the Nyquist band that is, when the attenuation by the frequency filter extends to a frequency component of 1/2 or less of the symbol rate, the compensation by the linear equalization filter of the band limited distortion amplifies the high frequency signal component attenuated by the band limitation. It becomes operation. Since the signal-to-noise ratio is reduced in the attenuated band, amplification of the band results in a decrease in the overall signal-to-noise ratio.
- Optical transmission systems include wavelength multiplexing / demultiplexing filters and optical filters for ADD-DROP, and the frequency band of signal light is cut during transmission. It is necessary to transmit signal light having a wide band. Therefore, it is practically difficult to achieve the Nyquist limit frequency utilization efficiency. If signal light is transmitted in the Nyquist limit band, it is highly possible that the receiving side is cut below the Nyquist band and a large transmission penalty occurs. Furthermore, in an adaptive equalization filter, it is impossible in principle to compensate for distortion caused by a nonlinear optical effect.
- Maximum Likelihood Sequence Estimation which is another method for compensating for signal distortion, generates a plurality of reference signals by applying a transmission line model to a plurality of signal sequence candidates.
- This is a method for estimating the most likely transmission time sequence by evaluating the difference between the two and is advantageous over the adaptive equalization filter in terms of the above-described points. Since the operation of shaping the received signal cannot be performed, problems such as noise amplification do not occur when a band-limited signal is received. For this reason, the degradation of the reception quality due to the band limitation is more gradual than when the equalization filter is used. In addition, nonlinear distortion can be compensated by providing the transmission path model with nonlinearity.
- MLSE has a drawback that the calculation amount is large.
- M is a multi-level number of symbols
- M 2 for on / off keying (OOK)
- M 8 for polarization multiplexed quaternary phase modulation (DP-QPSK).
- ODK on / off keying
- DP-QPSK polarization multiplexed quaternary phase modulation
- the DP-QPSK signal is not directly input to the MLSE decoder, but is first subjected to polarization separation, or polarization separation and quadrature phase separation, and then input to a plurality of MLSE decoders. Can be reduced.
- Polarization separation can be realized using a suitably adaptively controlled butterfly filter.
- the butterfly filter is a 2-input 2-output signal processing circuit having a configuration in which linear filters hxx, hxy, hyx, and hyy are arranged. Input two lane signals of X and Y, perform a convolutional linear sum under appropriately selected two sets of four transfer functions, and output a two-lane signal. That is, when the inputs are X in , Y in , outputs X out , and Y out , the input / output relationship is as shown in the following equation (1).
- n is the sample number of the signal and T is the sample interval time.
- t is the time.
- hij is a time domain transfer function of four linear filters, and directly corresponds to a tap coefficient when the butterfly filter is implemented by an FIR filter. * Represents a convolution operation.
- the output signal is in a state where the polarization is separated. Also, as mentioned in the background section, various linear distortions can be compensated.
- the adaptive equalization filter in the previous stage leaves intersymbol interference due to band-limited distortion, while compensating for other signal distortion and polarization crosstalk, It is necessary to output.
- Conventionally known adaptive equalization algorithms for optical receivers cannot intentionally leave part of such distortion.
- the present invention has been made in consideration of such circumstances, the purpose of which is to adopt an adaptive equalization filter that performs decision feedback, and by inserting a fixed filter corresponding to the band limitation into the feedback loop,
- An object of the present invention is to provide a digital signal processing apparatus capable of optimizing the parameters of the adaptive filter so that the intersymbol interference corresponding to the amount of the inserted fixed filter remains.
- the present invention is a digital signal processing device that is included in an optical signal receiver for optical fiber communication and that processes a digital signal converted from an optical signal by a photoelectric conversion device and an analog-digital converter.
- a linear adaptive filter that inputs the digital signal, outputs a dynamically controllable linear transfer function by acting on the input digital signal, and inputs an output signal of the linear adaptive filter,
- a plurality of reference signals are generated by applying a transfer function of the transmission path model to the signal sequence candidates of the input signal, and the difference between the input signal of the linear adaptive filter and the reference signal is evaluated to most likely transmit time sequence
- a maximum likelihood sequence decoder that decodes a received signal by maximum likelihood sequence estimation, and decoded data from the maximum likelihood sequence decoder.
- a signal regenerator that generates a signal corresponding to the data generator, and an output signal from the signal regenerator, and a distortion equivalent to the transmission path model used in the maximum likelihood sequence decoder is input to the signal regenerator.
- a feedback distortion adding filter that is output in addition to the output signal, an output signal of the feedback distortion adding filter is input as a target signal, the digital signal input to the linear adaptive filter is input, and the target signal and input
- an adaptive equalization filter control block that updates a tap coefficient of the linear adaptive filter by an LMS (Least Mean Square) algorithm using a difference from the digital signal as an error signal.
- an amount of delay equal to the sum of a delay of the maximum likelihood sequence decoder and a delay of the feedback distortion addition filter is given to the digital signal input to the linear adaptive filter, and the adaptive equalization is performed. It is preferable to further include an input-side delay circuit that outputs to the filter control block.
- the output signal of the linear adaptive filter is input, a carrier frequency / phase offset is estimated, a phase compensation signal, and a carrier phase estimation circuit that outputs an inverse compensation signal that is the inverse of the phase compensation signal;
- a carrier frequency / phase offset is estimated, a phase compensation signal, and a carrier phase estimation circuit that outputs an inverse compensation signal that is the inverse of the phase compensation signal;
- Arranged between the linear adaptive filter and the maximum likelihood sequence decoder input the output signal of the linear adaptive filter as a main signal, input the phase compensation signal output from the carrier phase estimation circuit, A phase compensation circuit that applies the phase compensation signal to the output signal of the linear adaptive filter and outputs the phase compensation signal to the maximum likelihood sequence decoder; a feedback distortion addition filter; and an adaptive equalization filter control block.
- the output signal of the feedback distortion addition filter is input as a main signal, and the inverse complement output from the carrier phase estimation circuit is disposed between Inputs a signal, preferably further comprises a phase inverse compensation circuit for outputting to the adaptive equalization filter control block on an acting the inverse compensation signal to the output signal of the feedback distortion adding filter.
- a delay of an amount equal to the delay of phase estimation is given to the output signal of the linear adaptive filter, and the main signal delay circuit supplied to the phase compensation circuit and the preceding stage of the adaptive equalization filter control block
- the digital signal input to the linear adaptive filter is input, and an amount equal to the sum of the delay of the maximum likelihood sequence decoder, the delay of the feedback distortion addition filter, and the delay of the main signal delay circuit.
- An input-side delay circuit that provides a delay and supplies the adaptive equalization filter control block, and is arranged between the carrier phase estimation circuit and the phase inverse compensation circuit, and is output from the carrier phase estimation circuit.
- a phase inverse compensation signal delay circuit for providing a compensation signal with an amount of delay equal to the sum of the delay of the maximum likelihood sequence decoder and the delay of the feedback distortion addition filter; It is preferably provided.
- a bandwidth limitation estimation is performed by estimating a bandwidth limitation width and a roll-off shape received by the optical signal in a transmission process in an optical transmission link, and generating and outputting a parameter characterizing the bandwidth limitation.
- the maximum likelihood sequence decoder uses the parameter for generating the reference signal, and the feedback distortion adding filter applies a band-limited distortion corresponding to the parameter to the output signal of the signal regenerator. It is preferable to add.
- the present invention is included in an optical signal receiver for optical fiber communication, and is converted into two lanes converted from a polarization multiplexed optical signal by a photoelectric conversion device and an analog-digital converter.
- a digital signal processing apparatus for processing a digital electric signal of a complex type, straight and crossed through four linear adaptive filters having a transfer function that can be controlled from the outside between two inputs and two outputs
- a butterfly filter having a connectable configuration, inputting a digital electric signal of the two lanes and outputting a digital signal of two lanes on the X side and the Y side, and inputting an X side output signal of the butterfly filter,
- a plurality of control signals on the X side are generated by causing the transfer function of the transmission path model to act on the signal sequence candidates, and the input of the butterfly filter
- An X-side maximum likelihood sequence decoding block for decoding an X-side received signal by maximum likelihood sequence estimation for estimating a most likely transmission signal time sequence by evaluating a difference between a side output signal and the
- the Y side feedback distortion adding filter to be added and the output signal of the X side feedback distortion adding filter are inputted as the X side target signal, and the X side input signal and the Y side input signal to the butterfly filter are inputted.
- LMS Local Mobile System
- An X-side adaptive equalization filter control block for updating tap coefficients of two linear adaptive filters connected to the X-side output of the butterfly filter by means of the Mean Squared algorithm, and an output signal of the Y-side feedback distortion adding filter as Y Input side Y signal, the input side Y signal and the input side X signal are input, the difference between the Y side target signal and the input side Y signal, and the Y side target signal and the input side X signal
- the tap coefficients of the two linear adaptive filters connected to the Y-side output of the butterfly filter are updated by the LMS algorithm using the difference between the two as an error signal.
- a digital signal processor and a Y-side adaptive equalization filter control block for updating tap coefficients of two linear adaptive filters connected to the X-side output of the butterfly filter by means of the Mean Squared algorithm, and an output signal of the Y-side feedback distortion adding filter as Y Input side Y signal, the input side Y signal and the input side X signal are input, the difference between the Y side target signal
- the X-side input signal to the butterfly filter is given a delay of an amount equal to the sum of the delay of the X-side maximum likelihood sequence decoding block and the delay of the X-side feedback distortion addition filter, X input side delay circuit supplied as the input side X signal of the X side adaptive equalization filter control block and the input side X signal of the Y side adaptive equalization filter control block, and the Y side input to the butterfly filter
- the signal is given a delay equal to the sum of the delay of the Y-side maximum likelihood sequence decoding block and the delay of the Y-side feedback distortion addition filter, and the input side Y of the Y-side adaptive equalization filter control block
- an X-side carrier phase estimation block that inputs the X-side output signal of the butterfly filter, estimates a carrier phase offset, and outputs an X-side phase compensation signal and an X-side phase inverse compensation signal, and the butterfly filter
- the X side output signal is input as a main signal
- the X side phase compensation signal from the X side carrier phase estimation block is input
- the X side phase compensation signal is applied to the X side output signal of the butterfly filter.
- an X-side phase compensation circuit to be supplied to the X-side maximum likelihood sequence decoding block and the output signal from the X-side feedback distortion addition filter are input as main signals, and from the X-side carrier phase estimation block
- the X-side phase inverse compensation signal is input, and the X-side phase inverse compensation signal is applied to the output signal from the X-side feedback distortion addition filter.
- An X-side phase inverse compensation circuit that outputs the X-side target signal of the X-side adaptive equalization filter control block and the Y-side output signal of the butterfly filter are input, a carrier phase offset is estimated, and a Y-side phase compensation is performed.
- a Y-side carrier phase estimation block that outputs a signal and a Y-side phase inverse compensation signal, and the Y-side output signal of the butterfly filter as a main signal, and the Y-side phase compensation signal from the Y-side carrier phase estimation block
- a Y-side phase compensation circuit for applying the Y-side phase compensation signal to the Y-side output signal of the butterfly filter and supplying the Y-side maximum likelihood sequence decoding block, and the Y-side feedback distortion
- the output signal from the additional filter is input as a main signal
- the Y-side phase inverse compensation signal from the Y-side carrier phase estimation block is input
- a Y-side phase inverse compensation circuit for applying the Y-side phase inverse compensation signal to the output signal from the Y-side feedback distortion adding filter and outputting the Y-side target signal of the Y-side adaptive equalization filter control block; Is preferably further provided.
- an X-side main signal delay circuit that gives a delay equal to the delay of the X-side carrier phase estimation block to the X-side output signal of the butterfly filter and outputs the delayed signal to the X-side phase compensation circuit;
- a Y-side main signal delay circuit that gives a delay equal to a delay of the Y-side carrier phase estimation block to the Y-side output signal of the butterfly filter and outputs the delayed signal to the Y-side phase compensation circuit;
- a delay of an amount equal to the sum of the delay of the X-side maximum likelihood sequence decoding block and the delay of the X-side feedback distortion addition filter is given to the X-side phase inverse compensation signal of the side carrier phase estimation block,
- An X-side phase inverse compensation signal delay circuit for outputting to the side-phase inverse compensation circuit, and the Y-side maximum likelihood sequence decoding block for the Y-side phase inverse compensation signal of the Y-side carrier phase estimation block
- a Y-side phase inverse compensation signal delay circuit that gives an amount of
- the digital electric signal output from the analog-digital converter is input, crosstalk between the X polarization channel and the Y polarization channel is removed, and the polarization-separated X side signal and Y side signal are separated.
- a one-tap butterfly filter for generating the X-side signal and the frequency offset of the X-side signal and the Y-side signal output from the one-tap butterfly filter, and the X-side signal and the Y-side signal that have been frequency offset compensated for the butterfly filter It is preferable to further comprise a frequency offset compensation circuit that outputs to the output.
- an adaptive equalization filter that performs decision feedback and inserting a fixed filter corresponding to the band limitation in the feedback loop, an intersymbol interference corresponding to the amount of the inserted fixed filter remains.
- the parameters of the adaptive filter can be optimized.
- FIG. 11 is a block diagram of an optical transmission link in which the digital signal processing device shown in FIG. 10 is incorporated in a receiving device. It is a block diagram which shows the structure of the optical receiver incorporating the digital signal processing apparatus by 1st Embodiment of this invention.
- FIG. 3 is a block diagram of an optical transmission link in which the external environment in the first embodiment, that is, a digital signal processing device is incorporated in a receiving device. It is a figure which shows the OSNR dependence of the bit error rate when a zone
- limiting is a symbol rate ratio 0.9 time (9 GHz).
- FIG. 6 is a diagram illustrating OSNR when the symbol rate normalized optical signal ⁇ 3 dB bandwidth is plotted on the horizontal axis and the bit error rate is 10 ⁇ 3 on the vertical axis. It is a block diagram which shows the structure of the optical receiver incorporating the digital signal processing apparatus by 2nd Embodiment of this invention. It is a block diagram of the optical transmission link which incorporated the external environment in this 2nd Embodiment, ie, the digital signal processing apparatus, in the receiver. It is a block diagram which shows the structure of a butterfly filter.
- An adaptive equalization filter according to an embodiment of the present invention is characterized in that a fixed filter corresponding to band limitation is inserted into a feedback loop of an adaptive equalization filter that performs decision feedback. As a result, the intersymbol interference corresponding to the inserted fixed filter can be left.
- FIG. 1 is a block diagram showing a basic configuration (No. 1) of a digital signal processing apparatus according to an embodiment of the present invention.
- the digital signal processing device is included in an optical signal receiver for optical fiber communication, and processes a digital signal converted from an optical signal by a photoelectric conversion device and an analog-digital converter.
- the digital signal processing apparatus includes a linear adaptive filter (FIR) 1, a maximum likelihood sequence decoder (MLSE) 2, a signal regenerator (Data ⁇ Signal) 3, a feedback distortion addition filter (distortion addition) 4, and an adaptive equalization filter control
- FIR linear adaptive filter
- MBE maximum likelihood sequence decoder
- Data ⁇ Signal Data ⁇ Signal
- Distortion addition feedback distortion addition filter
- a block (adaptive control) 5 is provided.
- the linear adaptive filter 1 outputs a dynamically controllable linear transfer function by acting on an input signal (digital signal).
- the maximum likelihood sequence decoder 2 receives the output signal of the linear adaptive filter 1 and generates a plurality of reference signals by applying a transfer function of the transmission path model to a plurality of signal sequence candidates.
- the received signal is decoded by maximum likelihood sequence estimation that estimates the most likely transmission time sequence by evaluating the difference.
- the signal regenerator 3 receives the decoded data from the maximum likelihood sequence decoder 2 and generates a signal corresponding to the decoded data.
- the feedback distortion addition filter 4 receives the output signal from the signal regenerator 3, adds distortion equivalent to the transmission path model used in the maximum likelihood sequence decoder 2, and outputs the added signal.
- the adaptive equalization filter control block 5 inputs the target signal that is the output of the feedback distortion addition filter 4 and the input signal to the linear adaptive filter 1, and uses the difference between the target signal and the input signal as an error signal as an error signal.
- the tap coefficient of the linear adaptive filter (FIR) 1 is updated by the Least Mean Square) algorithm.
- FIG. 2 is a block diagram showing a basic configuration (No. 2) of the digital signal processing apparatus according to the embodiment of the present invention. It should be noted that portions corresponding to those in FIG.
- the input-side delay circuit 6 is arranged in the preceding stage of the adaptive equalization filter control block 5 and inputs an input signal to the linear adaptive filter 1, and calculates the delay of the maximum likelihood sequence decoder 2 and the delay of the feedback distortion addition filter 4. An amount of delay equal to the sum is applied and supplied to the adaptive equalization filter control block 5.
- FIG. 3 is a block diagram showing a basic configuration (part 3) of the digital signal processing apparatus according to the embodiment of the present invention. It should be noted that portions corresponding to those in FIG.
- the carrier phase estimation circuit 7 receives the output signal of the linear adaptive filter 1, estimates the carrier frequency / phase offset, and outputs a compensation signal e ⁇ j ⁇ and an inverse compensation signal e j ⁇ that is the reciprocal of the compensation signal.
- the phase compensation circuit 8 is arranged between the linear adaptive filter 1 and the maximum likelihood sequence decoder 2, receives the output signal of the linear adaptive filter 1 as a main signal, and outputs the phase compensation signal output from the carrier phase estimation circuit 7. As a compensation signal, the compensation signal is applied to the main signal and supplied to the maximum likelihood sequence decoder 2.
- the phase inverse compensation circuit 9 is disposed between the feedback distortion addition filter 4 and the adaptive equalization filter control block 5, receives the output signal of the feedback distortion addition filter 4 as a main signal, and is output from the carrier phase estimation circuit 7.
- the inverse compensation signal is input as a compensation signal, and the compensation signal is applied to the main signal and then supplied to the adaptive equalization filter control block 5.
- FIG. 4 is a block diagram showing a basic configuration (No. 4) of the digital signal processing apparatus according to the embodiment of the present invention.
- the parts corresponding to those in FIG. The main signal delay circuit 6-1 is arranged between the output of the linear adaptive filter 1 and the input of the phase compensation circuit 8, and provides a delay of an amount equal to the delay of the phase estimation.
- the input-side delay circuit 6-2 is arranged in the preceding stage of the adaptive equalization filter control block 5 and inputs an input signal to the linear adaptive filter 1, and the delay of the maximum likelihood sequence decoder 2 and the feedback distortion addition filter 4 An amount of delay equal to the sum of the delay and the delay of the main signal delay circuit 6-1 is given and supplied to the adaptive equalization filter control block 5.
- the phase inverse compensation signal delay circuit 6-3 is inserted between the carrier phase estimation circuit 7 and the phase inverse compensation circuit 9, and the maximum likelihood sequence decoding is performed on the inverse compensation signal output from the carrier phase estimation circuit 7.
- An amount of delay equal to the sum of the delay of the device 2 and the delay of the feedback distortion adding filter 4 is provided.
- FIG. 5 is a block diagram showing the basic configuration (No. 5) of the digital signal processing apparatus according to the embodiment of the present invention. It should be noted that portions corresponding to those in FIG.
- the digital signal processing apparatus shown in FIG. 5 further includes a band limit estimation circuit 10 in addition to the configuration shown in FIG.
- the band limit estimation circuit 10 inputs an input signal (that is, a digital signal converted from an optical signal by an analog / digital converter) to the digital signal processing device, and the optical signal is transmitted to the optical transmitter, A band limit width and a roll-off shape received from a transmission line, an optical receiver, etc. are estimated based on the digital signal, a parameter characterizing the band limit is generated, and the generated parameter is used as the maximum likelihood sequence decoder 2 and feedback distortion.
- an input signal that is, a digital signal converted from an optical signal by an analog / digital converter
- a band limit width and a roll-off shape received from a transmission line, an optical receiver, etc. are estimated based on the digital signal, a parameter characterizing
- the maximum likelihood sequence decoder 2 receives the parameters output from the band limit estimation circuit 10 and is input when generating a plurality of reference signals by applying a transfer function of a transmission path model to a plurality of signal sequence candidates. Parameters are used.
- the feedback distortion adding filter 4 receives the parameter output from the band limit estimation circuit 10 and uses the input parameter to apply a band limit distortion corresponding to the band limit to the output signal from the signal regenerator 3. give. According to the above configuration, an optimum band limitation model corresponding to the input digital signal can be given to the maximum likelihood sequence decoder 2 and the feedback distortion addition filter 4.
- FIG. 6 is a block diagram showing the basic configuration (No. 6) of the digital signal processing apparatus according to the embodiment of the present invention.
- the digital signal processing device is included in the optical signal receiver for optical fiber communication, and processes the two-lane complex digital electric signal converted from the polarization multiplexed optical signal by the photoelectric conversion device and the analog-digital converter. To do.
- the digital signal processing apparatus includes a butterfly filter 20, an X-side maximum likelihood sequence decoder (X-side MLSE decoding block) 21-1, a Y-side maximum likelihood sequence decoder (Y-side MLSE decoding block) 21-2, and an X-side signal reproduction.
- Block 22-1 Y side signal regeneration block 22-2, X side feedback distortion addition filter (distortion addition) 23-1, Y side feedback distortion addition filter 23-2, X side adaptive equalization filter control block (adaptive control) 24-1 and a Y-side adaptive equalization filter control block 24-2.
- the butterfly filter 20 has a configuration in which two inputs and two outputs are connected in a straight and cross manner via four linear adaptive filters having transfer functions that can be controlled from the outside.
- a digital electric signal is input, and digital signals of two lanes on the X side and the Y side are output.
- the X-side maximum likelihood sequence decoder 21-1 receives the X-side output signal of the butterfly filter 20, generates a plurality of reference signals by applying a transfer function of a transmission path model to a plurality of signal sequence candidates, and generates an input signal.
- the X-side received signal is decoded by maximum likelihood sequence estimation that estimates the most likely transmission signal time sequence by evaluating the difference between the control signal and the reference signal.
- the Y-side maximum likelihood sequence decoder 21-2 receives the Y-side output signal of the butterfly filter 20, generates a plurality of reference signals by applying a transfer function of a transmission path model to a plurality of signal sequence candidates, and generates an input signal.
- the Y-side received signal is decoded by maximum likelihood sequence estimation that estimates the most likely transmission signal time sequence by evaluating the difference between the control signal and the reference signal.
- the X-side signal reproduction block 22-1 receives the output data of the X-polarization in-phase component and the X-polarization quadrature component of the X-side maximum likelihood sequence decoder 21-1, and reproduces and outputs the X-side signal.
- the Y-side signal reproduction block 22-2 receives the output data of the Y-polarization in-phase component and Y-polarization quadrature component of the Y-side maximum likelihood sequence decoder 21-2, and reproduces and outputs the Y-side signal.
- the X-side feedback distortion addition filter 23-1 receives the output signal of the X-side signal regeneration block 22-1 and adds distortion equivalent to the transmission path model used in the X-side maximum likelihood sequence decoder 21-1. And output.
- the Y-side feedback distortion addition filter 23-2 receives the output signal of the Y-side signal regeneration block 22-2 and adds distortion equivalent to the transmission path model used in the Y-side maximum likelihood sequence decoder 21-2. And output.
- the X-side adaptive equalization filter control block 24-1 inputs the output of the X-side feedback distortion adding filter 23-1 as a target signal, and converts the X-side input signal and Y-side input signal to the butterfly filter 20 into the input-side X Signal and input side Y signal, and the difference between the target signal and input side X signal and the difference between the target signal and input side Y signal as error signals to the X side output of the butterfly filter 20 by the Last Mean Squared algorithm.
- the tap coefficients of the two connected FIR filters are updated.
- the Y-side adaptive equalization filter control block 24-2 inputs the output of the Y-side feedback distortion adding filter 23-2 as a target signal, and converts the Y-side input signal and the X-side input signal to the butterfly filter 20 into the input side Y Signal and input side X signal, and the difference between the target signal and the input side Y signal and the difference between the target signal and the input side X signal as error signals to the Y side output of the butterfly filter 20 by the Last Mean Squared algorithm
- the tap coefficients of the two connected FIR filters are updated.
- FIG. 7 is a block diagram showing the basic configuration (No. 7) of the digital signal processing apparatus according to the embodiment of the present invention. Note that portions corresponding to those in FIG. 6 are denoted by the same reference numerals and description thereof is omitted.
- the digital signal processing apparatus further includes an X input side delay circuit 25-1 and a Y input side delay circuit 25-2 in addition to the configuration shown in FIG.
- the X input side delay circuit 25-1 receives a signal supplied to the X side input of the butterfly filter 20, and adds the delay of the X side maximum likelihood sequence decoder 21-1 and the X side feedback distortion to the signal.
- An amount of delay equal to the sum of the delay of the filter 23-1 is given and output to the input side X signal input of the X side and Y side adaptive equalization filter control blocks 24-1 and 24-2.
- the Y input side delay circuit 25-2 inputs a signal supplied to the Y side input of the butterfly filter 20, and adds the delay of the Y side maximum likelihood sequence decoder 21-2 and the Y side feedback distortion to the signal.
- An amount of delay equal to the sum of the delay of the filter 23-2 is given and output to the input Y signal input of the Y side and X side adaptive equalization filter control blocks 24-2 and 24-1.
- FIG. 8 is a block diagram showing the basic configuration (No. 8) of the digital signal processing apparatus according to the embodiment of the present invention. Note that portions corresponding to those in FIG. 6 are denoted by the same reference numerals and description thereof is omitted.
- the digital signal processing apparatus has an X-side carrier phase estimation block 26-1, a Y-side carrier phase estimation block 26-2, an X-side phase compensation circuit 27-1, and a Y-side phase compensation circuit 27 for the configuration shown in FIG. -2, an X-side phase reverse compensation circuit 28-1, and a Y-side phase reverse compensation circuit 28-2.
- the X-side carrier phase estimation block 26-1 receives the X-side output signal of the butterfly filter 20, estimates the carrier phase offset, and outputs a phase compensation signal and a phase inverse compensation signal.
- the X-side phase compensation circuit 27-1 is inserted between the X-side output of the butterfly filter 20 and the X-side maximum likelihood sequence decoder 21-1, and inputs the X-side output of the butterfly filter 20 as a main signal.
- the phase compensation signal of the X-side carrier phase estimation block 26-1 is input as a compensation signal, and the compensation signal is applied to the main signal and then output to the X-side maximum likelihood sequence decoder 21-1.
- the X-side phase inverse compensation circuit 28-1 is inserted between the X-side feedback distortion adding filter 23-1 and the X-side adaptive equalization filter control block 24-1, and is supplied from the X-side feedback distortion adding filter 23-1.
- the signal is input as the main signal
- the phase inverse compensation signal from the X-side carrier phase estimation block 26-1 is input as the compensation signal
- the compensation signal is applied to the main signal
- the X-side adaptive equalization filter control block 24 -1 as a target signal input.
- the Y-side carrier phase estimation block 26-2 receives the Y-side output signal of the butterfly filter 20, estimates the carrier phase offset, and outputs a phase compensation signal and a phase inverse compensation signal.
- the Y-side phase compensation circuit 27-2 is inserted between the Y-side output of the butterfly filter 20 and the Y-side maximum likelihood sequence decoder 21-2, and inputs the Y-side output of the butterfly filter 20 as a main signal.
- the phase compensation signal of the Y-side carrier phase estimation block 26-2 is input as a compensation signal, applied to the main signal, and then output to the Y-side maximum likelihood sequence decoder 21-2.
- the Y-side phase inverse compensation circuit 28-2 is inserted between the Y-side feedback distortion adding filter 23-2 and the Y-side adaptive equalization filter control block 24-2, and is supplied from the Y-side feedback distortion adding filter 23-2.
- the signal is input as a main signal
- the phase inverse compensation signal from the Y-side carrier phase estimation block 26-2 is input as a compensation signal
- the Y-side adaptive equalization filter control block 24 -2 output as a target signal input.
- FIG. 9 is a block diagram showing the basic configuration (No. 9) of the digital signal processing apparatus according to the embodiment of the present invention. Note that portions corresponding to those in FIG. 6 are denoted by the same reference numerals and description thereof is omitted.
- the digital signal processing apparatus is different from the configuration shown in FIG. 6 in that an X-side main signal delay circuit 30-1, a Y-side main signal delay circuit 30-2, an X-side phase reverse compensation signal delay circuit 31-1, and a Y-side phase.
- a reverse compensation signal delay circuit 31-2, an X input side delay circuit 32-1, and a Y input side delay circuit 32-2 are further provided.
- the X-side main signal delay circuit 30-1 is inserted between the X-side output of the butterfly filter 20 and the X-side phase compensation circuit 27-1, and has an amount of delay equal to the delay of the X-side carrier phase estimation block 26-1.
- the Y-side main signal delay circuit 30-2 is inserted between the Y-side output of the butterfly filter 20 and the Y-side phase compensation circuit 27-2, and has an amount of delay equal to the delay of the Y-side carrier phase estimation block 26-2. give.
- the X-side phase inverse compensation signal delay circuit 31-1 is inserted between the phase inverse compensation output of the X-side carrier phase estimation block 26-1 and the X-side phase inverse compensation circuit 28-1, and the X-side maximum likelihood sequence decoding is performed.
- An amount of delay equal to the sum of the delay of the device 21-1 and the delay of the X-side feedback distortion adding filter 23-1 is given.
- the Y-side phase inverse compensation signal delay circuit 31-2 is inserted between the phase inverse compensation output of the Y-side carrier phase estimation block 26-2 and the Y-side phase inverse compensation circuit 28-2, and the Y-side maximum likelihood sequence decoding An amount of delay equal to the sum of the delay of the device 21-2 and the delay of the Y-side feedback distortion adding filter 23-2 is given.
- the X input side delay circuit 32-1 inputs a signal supplied to the X side input of the butterfly filter 20, and with respect to the signal, the delay of the X side main signal delay circuit 30-1 and the X side maximum likelihood sequence decoding An amount of delay equal to the sum of the delay of the counter 21-1 and the delay of the X-side feedback distortion adding filter 23-1 is applied to the X-side and Y-side adaptive equalization filter control blocks 24-1 and 24-2. Output to side signal input.
- the Y input side delay circuit 32-2 receives a signal supplied to the Y side input of the butterfly filter 20, and the delay of the Y side main signal delay circuit 30-2 and the Y side maximum likelihood sequence decoding are performed on the signal.
- Y2 of the Y-side and X-side adaptive equalization filter control blocks 24-2 and 24-1 by giving an amount of delay equal to the sum of the delay of the counter 21-2 and the delay of the Y-side feedback distortion adding filter 23-2. Output to side signal input.
- FIG. 10 is a block diagram showing the basic configuration (No. 10) of the digital signal processing apparatus according to the embodiment of the present invention.
- the digital signal processing apparatus further includes a one-tap butterfly filter 33 and a frequency offset compensation circuit 34 in the configuration shown in FIG.
- An existing 1-tap butterfly filter can be used as the 1-tap butterfly filter 33.
- an existing frequency offset compensation circuit can be used for the frequency offset compensation circuit 34.
- the frequency offset compensation circuit 34 includes a circuit using a Viterbi-Viterbi algorithm or a provisional determination circuit and a PLL (Phase Locked Loop) circuit.
- the digital signal processing apparatus shown in FIG. 8 also performs frequency offset compensation, but differs from the digital signal processing apparatus shown in FIG. 10 in the position where frequency offset compensation is performed.
- FIG. 11 is a block diagram of an optical transmission link in which the digital signal processing device shown in FIG. 10 is incorporated in a receiving device.
- This optical transmission link includes a transmitter 110, an optical fiber transmission line 110, a coherent detector 120, a local oscillation laser 130, an analog-digital converter 140, and a reception signal processing circuit 150.
- the transmitter 100 includes an optical modulator 101 and a transmission laser 102.
- the coherent detector 120, the local oscillation laser 130, the analog-digital converter 140, and the reception signal processing circuit 150 constitute a reception device.
- the received signal processing circuit 150 corresponds to the digital signal processing device shown in FIG. 10, and the digital signal output from the analog-digital converter 140 is supplied to the 1-tap butterfly filter 33 shown in FIG.
- the coherent detector 120 and the local oscillation laser 130 correspond to a photoelectric conversion device that is not shown in the above description.
- the frequency offset is removed from the input signals to the butterfly filter 20, the X-side adaptive equalization filter control block 24-1, and the Y-side adaptive equalization filter control block 24-2.
- the digital signal processing apparatus shown in FIG. 10 includes a frequency offset compensation circuit 34 as a circuit for removing the frequency offset from the input signal.
- currently known methods eg, Viterbi-Viterbi algorithm
- the digital signal processing apparatus shown in FIG. 10 includes a one-tap butterfly filter 33 for polarization channel separation before the frequency offset compensation circuit 34.
- the optical modulator 101 modulates the output of the transmission laser 102 with data to be transmitted, maps the data to a polarization multiplexed QPSK optical signal, and sends the polarization multiplexed QPSK optical signal to the optical fiber transmission line 110.
- the coherent detector 120 uses the output signal of the local oscillation laser 130 to coherently detect the polarization multiplexed QPSK optical signal propagated through the optical fiber transmission line 110 and convert it into an analog electric signal.
- the analog-digital converter 140 converts the analog electric signal output from the coherent detector 120 into a digital signal, and outputs the converted digital signal to the reception signal processing circuit 150 (digital signal processing device shown in FIG. 10).
- This digital signal is in a state where the X-polarized channel signal and the Y-polarized channel signal are mixed, and has a frequency offset derived from a frequency shift between the transmission laser 102 and the local oscillation laser 130.
- the maximum amount of frequency offset is about 2 GHz.
- the digital signal is input to the 1-tap butterfly filter 33.
- the 1-tap butterfly filter 33 is adaptively controlled by the CMA, removes the crosstalk between the X polarization channel and the Y polarization channel, separates the X channel polarization and the Y channel polarization, and separates the polarization separated X
- the side signal and the Y side signal are output to the frequency offset compensation circuit 34.
- the frequency offset compensation circuit 34 receives the X-side signal and Y-side signal of the 1-tap butterfly filter 33 as input, removes the frequency offset, and converts the X-side signal and Y-side signal compensated for the frequency offset into the butterfly filter 20 and the X-side signal.
- Examples of the advantages of the digital signal processing apparatus include the following. First, frequency offset compensation can be performed stably. Second, since the polarization state input to the butterfly filter 20 is constant, the initial convergence operation of the butterfly filter 20 is stabilized. Third, it is robust against input polarization fluctuations.
- the butterfly filter 20 is a multi-tap butterfly filter configured using an FIR filter. Since the 1-tap butterfly filter 33 cannot perform distortion compensation unlike the multi-tap butterfly filter, the butterfly filter 20 is required for distortion compensation even in the configuration in which the 1-tap butterfly filter 33 is provided.
- FIG. 12 is a block diagram showing a configuration of an optical receiver incorporating the digital signal processing device according to the first embodiment of the present invention.
- FIG. 13 is a block diagram of an external environment in the first embodiment, that is, an optical transmission link in which a digital signal processing device is incorporated in a receiving device.
- the optical transmitter 60 generates a polarization multiplexed quaternary phase modulation signal and inputs it to the transmission path 62.
- the optical signal is added with chromatic dispersion, polarization mode dispersion, band limiting distortion, rotation of the polarization state, and noise derived from spontaneously emitted light from an optical amplifier (not shown) in the transmission path 62 (band limiting 61, 63), and input to the optical receiver (optical front end 64, digital signal processing device 65).
- the input light to the optical receiver is mixed with the local oscillation light from the local oscillator (LO) 41 in the 90 ° optical hybrid (optical front end) 40 (corresponding to the optical front end 64). Homodyne detection is performed with four sets of balanced photodiodes.
- the optical signal is converted into a 4-lane baseband analog electrical signal. Four lanes correspond to the X polarization in-phase component and the quadrature phase component, the Y polarization in-phase component and the quadrature phase component, respectively, but at this stage, due to the influence of the rotation of the polarization state in the transmission path 62, The signals are linearly mixed.
- the analog electrical signal is digitized by an analog-digital converter (hereinafter referred to as ADC), and further converted into a 4-lane digital signal of 2 samples per symbol by a resample circuit.
- the 4-lane digital signal is treated as a complex number with the in-phase component as the real part and the quadrature component as the imaginary part (imag), and is treated as a 2-lane complex digital signal corresponding to the X polarization and the Y polarization. Is called.
- the conversion circuits 42-1 and 42-2 convert the 4-lane digital signal (actual signal) output from the optical front end 40 into a 2-lane complex digital signal corresponding to the X polarization and the Y polarization. To do.
- the X-polarized and Y-polarized complex digital signals are input to the adaptively controlled butterfly filter 43. If the adaptive equalization algorithm is operating properly, the crosstalk between XY polarizations and linear waveform distortion such as chromatic dispersion and polarization mode dispersion are compensated here.
- the butterfly filter 43 includes four finite impulse response filters (FIR filters) that are straight or cross-connected. The output of the FIR filter is down-sampled to 2: 1, and the output signal of the butterfly filter 43 is a signal of one sample per symbol.
- the output of the butterfly filter 43 is output from the frequency and phase in the X-side phase compensation circuit 50-1 and the Y-side phase compensation circuit 50-2 based on the phase compensation signals from the carrier phase estimation circuits 44-1 and 44-2. Offset is compensated.
- the two-lane complex digital signals output from the X-side phase compensation circuit 50-1 and the Y-side phase compensation circuit 50-2 are converted into in-phase components (real part (real part) by the conversion circuits 49-1 and 49-2, respectively. )) And quadrature component (imaginary part (imag)), and converted into a 4-lane digital signal (real signal).
- These maximum likelihood sequence estimation decoders 45-1 to 45-4 incorporate a transmission path model for generating a reference signal.
- the transmission path model corresponds to band limitation, and is adjusted so that the absolute value of the transfer function actually matches that of the transmission path through which the signal passes.
- the transmission line models of the maximum likelihood sequence estimation decoders 45-1 to 45-4 are adjusted so as to reproduce the band limitation of the actual transmission line.
- the decoding outputs of the maximum likelihood sequence estimation decoders 45-1 to 45-4 are X- and Y-side signal reproduction blocks 46-1 for every two lanes corresponding to the X polarization and every two lanes corresponding to the Y polarization. , 46-2, where it is again converted to a QPSK signal of one sample per symbol.
- the reproduced signal is input to feedback distortion adding filters 47-1 and 47-2 configured by FIR filters.
- the feedback distortion adding filters 47-1 and 47-2 have transfer characteristics equivalent to the transmission path model for generating a reference signal built in the maximum likelihood sequence estimation decoders 45-1 to 45-4, and code the input signal. Output with interference.
- the characteristics of the feedback distortion adding filters 47-1 and 47-2 determine the distortion compensation sharing between the butterfly filter 43 and the maximum likelihood sequence estimation decoders 45-1 to 45-4.
- the outputs of the feedback distortion addition filters 47-1 and 47-2 are supplied to the butterfly filter adaptive control block (LMS) operating with the LMS algorithm via the X-side phase inverse compensation circuit 51-1 and the Y-side phase inverse compensation circuit 51-2. ) 48-1 and 48-2 are input as target signals.
- LMS butterfly filter adaptive control block
- an input signal to the butterfly filter 43 is also input as an input signal to the butterfly filter adaptive control blocks 48-1 and 48-2 via the delay circuits 52-1 and 52-2.
- the butterfly filter adaptive control blocks 48-1 and 48-2 update the tap coefficient of the butterfly filter 43 using the difference between the input signal and the target signal as an error signal.
- the X-side phase compensation circuit 50-1 is inserted between the X-side output of the butterfly filter 43 and the conversion circuit 49-1, and receives the X-side output of the butterfly filter 43 as a main signal.
- ⁇ 1 phase compensation signal is input as a compensation signal, the compensation signal is applied to the main signal, and then output to the conversion circuit 49-1.
- the Y-side phase compensation circuit 50-2 is inserted between the Y-side output of the butterfly filter 43 and the conversion circuit 49-2, and receives the Y-side output of the butterfly filter 43 as a main signal.
- the carrier phase estimation block 44 -2 phase compensation signal is input as a compensation signal, the compensation signal is applied to the main signal, and then output to the conversion circuit 49-2.
- the X-side phase inverse compensation circuit 51-1 is inserted between the X-side feedback distortion adding filter 47-1 and the butterfly filter adaptive control block 48-1, and mainly receives the signal from the X-side feedback distortion adding filter 47-1.
- Input as a signal the phase inverse compensation signal from the X-side carrier phase estimation circuit 44-1 is input as a compensation signal, the compensation signal is applied to the main signal, and then the target signal input of the butterfly filter adaptive control block 48-1 Output as.
- the Y-side phase inverse compensation circuit 51-2 is inserted between the Y-side feedback distortion addition filter 47-2 and the butterfly filter adaptive control block 48-2, and mainly receives the signal from the Y-side feedback distortion addition filter 47-2.
- Input as a signal the phase inverse compensation signal from the Y-side carrier phase estimation circuit 44-2 is input as a compensation signal, the compensation signal is applied to the main signal, and then the target signal input of the butterfly filter adaptive control block 48-2 Output as.
- the butterfly filter 43 corresponds to the linear adaptive filter (FIR) 1 shown in FIGS. 1 to 5 and the butterfly filter 20 shown in FIGS.
- the carrier phase estimation circuits 44-1 and 44-2 include the carrier phase estimation circuit 7 in FIGS. 3 and 4, the X-side carrier phase estimation block 26-1 in FIGS. 8 and 9, and the Y-side carrier phase estimation block 26-2. It corresponds to.
- Maximum likelihood sequence estimation decoders 45-1 to 45-4 are the maximum likelihood sequence decoder (MLSE) 2 in FIGS. 1 to 5 and the X side maximum likelihood sequence decoder (X side MLSE decoding block in FIGS. 6 to 10). ) 21-1, corresponding to the Y-side maximum likelihood sequence decoder (Y-side MLSE decoding block) 21-2.
- the signal regeneration blocks 46-1 and 46-2 are the signal regenerator (Data ⁇ Signal) 3 of FIGS. 1 to 5, the X-side signal regeneration block 22-1 and the Y-side signal regeneration block 22- of FIGS. It corresponds to 2.
- the feedback distortion addition filters 47-1 and 47-2 are the feedback distortion addition filter (distortion addition) 4 of FIGS. 1 to 5, the X-side feedback distortion addition filter (distortion addition) 23-1, and FIGS. This corresponds to the side feedback distortion addition filter 23-2.
- the butterfly filter adaptive control blocks (LMS) 48-1 and 48-2 are the adaptive equalization filter control block (adaptive control) 5 in FIGS. 1 to 5 and the X-side adaptive equalization filter control block (in FIG. 6 to FIG. 10).
- the X-side phase compensation circuit 50-1 and the Y-side phase compensation circuit 50-2 are the phase compensation circuit 8 in FIGS. 3 and 4, the X-side phase compensation circuit 27-1, and the Y-side phase compensation circuit in FIGS. It corresponds to 27-2.
- the X-side phase reverse compensation circuit 51-1 and the Y-side phase reverse compensation circuit 51-2 include the phase reverse compensation circuit 9 shown in FIGS. 3 and 4, the X-side phase reverse compensation circuit 28-1 shown in FIGS. This corresponds to the side phase reverse compensation circuit 28-2.
- the delay circuits 52-1 and 52-2 are the input side delay circuit 6 in FIG. 2, the input side delay circuit 6-2 in FIG. 4, the X input side delay circuit 25-1 in FIG. 25-2 corresponds to the X input side delay circuit 32-1 and the Y input side delay circuit 32-2 in FIG.
- the LMS algorithm operates using the decision feedback signal to which band-limited distortion is added as a target signal, and the butterfly filter 43 leaves only a predetermined band-limited distortion, thereby chromatic dispersion and polarization mode dispersion. And so on to compensate for other linear impairments.
- the remaining band-limited distortion is compensated and decoded by the maximum likelihood sequence estimation decoders 45-1 to 45-4 in the subsequent stage.
- ⁇ ⁇ Pseudo random binary string was used as transmission data.
- An NRZ (Non-Return-to-Zero) signal waveform of every two samples is generated from the transmission data. Prepare four copies of the signal waveform, add delays to each other, and set the X-polarized in-phase (XI) channel, quadrature (XQ) channel, Y-polarized in-phase (YI) channel, and quadrature (YQ), respectively. ) Assigned to the channel.
- the symbol rate was 10 GHz. Since the modulation method is polarization multiplexing QPSK, the bit rate is 40 GHz.
- the signals of the X polarization and the Y polarization were mixed as shown in the following equation (2) while maintaining the orthogonal relationship.
- E x (nT) and E y (nT) and E x ′ (nT) and E y ′ (nT) are input signals of X polarization and Y polarization, and outputs of X polarization and Y polarization, respectively. Signal.
- the signal was input to the receiving block.
- the magnitude of white Gaussian noise was defined by an OSNR (Optical Signal-to-Noise Ratio) based on 12.5 GHz.
- the band-limited distortion was applied by passing through a fifth order Bessel type low pass filter.
- the output signals E x ′′ (nT) and E y ′′ (nT) are expressed by the following equations (3) and (4).
- H bw (t) and H CD (t) are time domain transfer functions representing band limitation and chromatic dispersion, respectively. Moreover, nx (t) and ny (t) represent white Gaussian noise. * Represents a convolution operation. The above signals pass through the butterfly filter 43, the X-side phase compensation circuit 50-1 and the Y-side phase compensation circuit 50-2, and the conversion circuits 49-1 and 49-2, and the maximum likelihood sequence estimation decoder 45- 1 to 45-4.
- A Proposed method according to the first embodiment, that is, a method of performing demodulation using a decision feedback LMS filter including feedback distortion adding filters 47-1 and 47-2 and maximum likelihood sequence estimation.
- B Simple combination method. A system that does not have the feedback distortion addition filters 47-1 and 47-2, and performs demodulation using a normal decision feedback LMS filter and maximum likelihood sequence estimation.
- C Conventional method. A system that does not have the feedback distortion addition filters 47-1 and 47-2, and performs demodulation by a normal determination feedback LMS filter and threshold determination.
- the maximum likelihood sequence estimation decoders 45-1 to 45-4 in the system A and the system B are implemented by a 32-state Viterbi algorithm with a constraint length of 5 symbols.
- the transmission path model for supplying the reference signal to the Viterbi decoder of system A is a band limiting model equivalent to that given on the transmission side, and was implemented as a fixed tap FIR filter.
- FIR filters having the same transfer function are inserted as feedback distortion adding filters 47-1 and 47-2 in feedback loops for supplying target signals to the butterfly filter adaptive control blocks 48-1 and 48-2.
- the constraint length in the proposed method A may be two symbols or more. If the constraint length is 2 to 3 symbols, a limited effect can be obtained regarding the compensation of the band-limited distortion. If the constraint length is 5 symbols or more, an obvious effect is obtained. The longer the restraint length, the more remarkable effect is obtained, so the upper limit is not limited theoretically. However, since the calculation amount increases as the constraint length increases, the constraint length is determined by a trade-off between the obtained effect and the calculation amount. Considering the performance of current devices, the maximum constraint length is about 7-8 symbols.
- the transfer function of the transmission path model of the Viterbi decoder in the system B is a transfer function obtained by correcting the band limit compensation operation performed by the LMS adaptive equalization filter from the band limit given on the transmission side. That is, the transmission path model including the adaptive equalization filter is performed to generate the transmission path model. Further, in the system B and the system C, the feedback distortion addition filter for supplying the target signal is not inserted into the butterfly filter adaptive control blocks 48-1 and 48-2.
- the OSNR tolerance for each bandwidth limit of each method A, B, C was evaluated by simulation.
- the -3 dB full width of the band limiting filter was varied between 0.7 B and 1.5 B in increments of 0.05 B, where B is the symbol rate.
- Noise was given in increments of 1 dB from 15 dB to 6 dB in terms of OSNR.
- FIG. 14 is a diagram showing the OSNR dependence of the bit error rate when the band limitation is 0.9 times the symbol rate ratio (9 GHz).
- the “ ⁇ ” plot is received by method A, the “ ⁇ ” plot is received by method B, and the “ ⁇ ” plot is received by method C.
- FIG. 15 is a diagram illustrating the OSNR when the symbol rate normalized optical signal ⁇ 3 dB bandwidth is plotted on the horizontal axis and the bit error rate is 10 ⁇ 3 on the vertical axis. It can be seen that the increase in required OSNR due to the bandwidth limitation is moderate in method A compared to method B and method C.
- This simulation result shows the advantage of the method A when demodulating a band-limited signal, that is, the proposed method according to the first embodiment.
- This advantage takes advantage of the fact that maximum likelihood sequence estimation can effectively compensate for intersymbol interference due to bandwidth limitations.
- the combination of the adaptive equalization filter and the maximum likelihood sequence estimation is not sufficient to take advantage of this advantage, and the feedback distortion adding filter 47-1.
- the simulation results show the effectiveness of the first embodiment using 47-2.
- FIG. 16 is a block diagram showing a configuration of an optical receiver incorporating a digital signal processing device according to the second embodiment of the present invention. Note that portions corresponding to those in FIG. 12 are denoted by the same reference numerals and description thereof is omitted.
- FIG. 17 is a block diagram of an external environment in the second embodiment, that is, an optical transmission link in which a digital signal processing device is incorporated in a receiving device.
- the optical transmitter 80 generates a polarization multiplexed quaternary phase modulation signal and inputs it to the transmission path 82.
- the optical signal is added with chromatic dispersion, polarization mode dispersion, band-limited distortion, rotation of the polarization state, and noise derived from the spontaneous emission light of the optical amplifier in the transmission line 82 (band restrictions 81 and 83),
- the signal is input to the receiver (optical front end 84, digital signal processing device 85).
- the external environment is almost the same as in the first embodiment. However, in the transmission line 82, in addition to the linear signal distortion and additive Gaussian noise listed in the first embodiment, nonlinear distortion due to self-phase modulation is added to the optical signal.
- the configuration of the butterfly filter 43 is the same as that shown in the first embodiment, and includes four FIR filters.
- the maximum likelihood sequence decoding blocks 70-1 and 70-2 are prepared as one independent Viterbi decoder for each of the X polarization channel and the Y polarization channel.
- the transmission path model of the Viterbi decoder is composed of a linear part that gives band limitation and a non-linear part that gives pattern-dependent distortion caused by self-phase modulation.
- the linear part is implemented with an FIR filter with fixed tap coefficients.
- the non-linear part is a filter controlled by a distortion table that directly specifies the distortion amount depending on the pattern.
- the distortion table is generated by the training signal when the link is established.
- the feedback distortion adding filters 47-1 and 47-2 operate by the same mechanism as the transmission path model built in the Viterbi decoder. That is, it is composed of two filters, a linear part and a nonlinear part, and the transfer characteristics of the filter of the nonlinear part are dynamically controlled according to the input signal sequence.
- the transfer characteristics of the linear filter and the distortion table for controlling the non-linear filter are the same as the transmission path model built in the Viterbi decoder.
- the outputs from the feedback distortion adding filters 47-1 and 47-2 are input as target signals to the butterfly filter adaptive control blocks 48-1 and 48-2, as in the first embodiment.
- the butterfly filter 43 compensates only for linear distortion other than band-limited distortion and nonlinear distortion, and polarization rotation.
- the non-linear distortion passes without being compensated even by a normal linear filter, which is not according to the embodiment of the present invention, but adversely affects the tap coefficient estimation accuracy of the adaptive equalization algorithm. In the method of the second embodiment, such an adverse effect on the tap coefficient estimation accuracy does not occur.
- the distortion caused by the band-limited signal is compensated by the maximum likelihood sequence estimation, and other linear distortions are compensated by the linear filter.
- the decrease in the reception sensitivity due to the restriction of the frequency band of the optical signal can be relieved greatly.
- the decrease in reception sensitivity becomes very gradual. Therefore, it is possible to improve the frequency utilization efficiency of the optical signal, and when constructing a network that passes through the optical bandpass filter in multiple stages, the penalty of narrowing by the optical filter can be reduced. it can.
- the band-limited distortion and the nonlinear distortion are compensated by the maximum likelihood sequence estimation, and the linear distortion other than the band-limited distortion is compensated by the linear filter. It is possible to compensate for the degradation of the received signal quality caused by nonlinear distortion. Compared with a method in which a linear filter and a nonlinear filter are connected in series, it is possible to improve the adaptive equalization convergence accuracy of the linear adaptive filter caused by nonlinear distortion.
- the present invention can be applied to, for example, an optical signal receiver for optical fiber communication.
- an adaptive equalization filter that performs decision feedback and inserting a fixed filter corresponding to the band limitation in the feedback loop, an intersymbol interference corresponding to the amount of the inserted fixed filter remains.
- the parameters of the adaptive filter can be optimized.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Optical Communication System (AREA)
Abstract
Description
本願は、2011年2月7日に日本へ出願された特願2011-024022号に基づき優先権を主張し、その内容をここに援用する。
従来、帯域制限歪による符号間干渉を残存させつつ、それ以外の信号歪みや、偏波クロストークを補償する機能を有する適応等化フィルタの存在は知られていなかった。本発明の一実施形態による適応等化フィルタは、判定帰還を行う適応等化フィルタの帰還ループに帯域制限に対応する固定フィルタを挿入することを特徴としている。これにより、挿入した固定フィルタに対応するシンボル間干渉を残存させることができる効果を奏する。
以下、上述したデジタル信号処理装置を実現するための第1実施形態について説明する。図12は、本発明の第1実施形態によるデジタル信号処理装置を組み込んだ光受信機の構成を示すブロック図である。また、図13は、本第1実施形態における外部環境、すなわち、デジタル信号処理装置を受信装置に組み込んだ光伝送リンクのブロック図である。
(A)本第1実施形態による提案方式、すなわち帰還歪付加フィルタ47-1、47-2を備えた判定帰還LMSフィルタと最尤系列推定とによる復調を行う方式。
(B)単純組み合わせ方式。帰還歪付加フィルタ47-1、47-2を持たない、通常の判定帰還LMSフィルタと最尤系列推定とによる復調を行う方式。
(C)従来方式。帰還歪付加フィルタ47-1、47-2を持たない、通常の判定帰還LMSフィルタと閾値判定とによる復調を行う方式。
次に、上述したデジタル信号処理装置を実現するための第2実施形態について説明する。図16は、本発明の第2実施形態によるデジタル信号処理装置を組み込んだ光受信機の構成を示すブロック図である。なお、図12に対応する部分には同一の符号を付けて説明を省略する。また、図17は、本第2実施形態における外部環境、すなわち、デジタル信号処理装置を受信装置に組み込んだ光伝送リンクのブロック図である。
2 最尤系列復号器(MLSE)
3 信号再生器(Data→Signal)
4 帰還歪付加フィルタ(歪付加)
5 適応等化フィルタ制御ブロック(適応制御)
6、6-2 入力側遅延回路
6-1 主信号遅延回路
6-3 位相逆補償信号遅延回路
7 キャリア位相推定回路
8 位相補償回路
9 位相逆補償回路
10 帯域制限推定回路(帯域制限推定)
20、43 バタフライフィルタ
21-1 X側最尤系列復号器(X側MLSE復号ブロック)
21-2 Y側最尤系列復号器(Y側MLSE復号ブロック)
22-1、46-1 X側信号再生ブロック
22-2、46-2 Y側信号再生ブロック
23-1、47-1 X側帰還歪付加フィルタ(歪付加)
23-2、47-2 Y側帰還歪付加フィルタ
24-1 X側適応等化フィルタ制御ブロック(適応制御)
24-2 Y側適応等化フィルタ制御ブロック
25-1 X入力側遅延回路
25-2 Y入力側遅延回路
26-1 X側キャリア位相推定ブロック
26-2 Y側キャリア位相推定ブロック
27-1 X側位相補償回路
27-2 Y側位相補償回路
28-1 X側位相逆補償回路
28-2 Y側位相逆補償回路
30-1 X側主信号遅延回路
30-2 Y側主信号遅延回路
31-1 X側位相逆補償信号遅延回路
31-2 Y側位相逆補償信号遅延回路
32-1 X入力側遅延回路
32-2 Y入力側遅延回路
33 1タップバタフライフィルタ
34 周波数オフセット補償回路(周波数オフセット補償)
44-1、44-2 キャリア位相推定回路
48-1、48-2 バタフライフィルタ適応制御ブロック(LMS)
52-1、52-2 遅延回路
70-1、70-2 最尤系列復号ブロック
Claims (10)
- 光ファイバ通信用途の光信号受信機に含まれ、光電変換装置およびアナログデジタルコンバータによって光信号から変換されたデジタル信号を処理するデジタル信号処理装置であって、
前記デジタル信号を入力し、動的に制御可能な線形の伝達関数を入力された前記デジタル信号に作用させて出力する線形適応フィルタと、
前記線形適応フィルタの出力信号を入力し、複数の信号系列候補に伝送路モデルの伝達関数を作用させて複数の対照信号を生成し、入力された前記線形適応フィルタの出力信号と前記対照信号との差分を評価して最も尤もらしい送信時間系列を推定する最尤系列推定により受信信号を復号する最尤系列復号器と、
前記最尤系列復号器からの復号データを入力し、前記復号データに対応する信号を生成する信号再生器と、
前記信号再生器からの出力信号を入力し、前記最尤系列復号器の中で用いられる前記伝送路モデルと等価な歪を前記信号再生器の前記出力信号に付加して出力する帰還歪付加フィルタと、
前記帰還歪付加フィルタの出力信号をターゲット信号として入力し、前記線形適応フィルタへ入力される前記デジタル信号を入力し、前記ターゲット信号と入力された前記デジタル信号との差を誤差信号として、LMS(Least Mean Square)アルゴリズムにより、前記線形適応フィルタのタップ係数を更新する適応等化フィルタ制御ブロックと
を備えるデジタル信号処理装置。 - 前記線形適応フィルタへ入力される前記デジタル信号に対して、前記最尤系列復号器の遅延と前記帰還歪付加フィルタの遅延との和に等しい量の遅延を与え、前記適応等化フィルタ制御ブロックへ出力する入力側遅延回路を更に備える請求項1に記載のデジタル信号処理装置。
- 前記線形適応フィルタの前記出力信号を入力し、キャリア周波数・位相オフセットを推定し、位相補償信号、及び前記位相補償信号の逆数である逆補償信号を出力するキャリア位相推定回路と、
前記線形適応フィルタと前記最尤系列復号器との間に配置され、前記線形適応フィルタの前記出力信号を主信号として入力し、前記キャリア位相推定回路から出力された前記位相補償信号を入力し、前記線形適応フィルタの前記出力信号に前記位相補償信号を作用させた上で前記最尤系列復号器に対して出力する位相補償回路と、
前記帰還歪付加フィルタと前記適応等化フィルタ制御ブロックとの間に配置され、前記帰還歪付加フィルタの前記出力信号を主信号として入力し、前記キャリア位相推定回路から出力された前記逆補償信号を入力し、前記帰還歪付加フィルタの前記出力信号に前記逆補償信号を作用させた上で前記適応等化フィルタ制御ブロックに出力する位相逆補償回路と
を更に備える請求項1に記載のデジタル信号処理装置。 - 前記線形適応フィルタの前記出力信号に対して位相推定の遅延に等しい量の遅延を与え、前記位相補償回路に供給する主信号遅延回路と、
前記適応等化フィルタ制御ブロックの前段に設けられ、前記線形適応フィルタへ入力される前記デジタル信号を入力し、前記最尤系列復号器の遅延と前記帰還歪付加フィルタの遅延と前記主信号遅延回路の前記遅延との和に等しい量の遅延を与えて、前記適応等化フィルタ制御ブロックへ供給する入力側遅延回路と、
前記キャリア位相推定回路と前記位相逆補償回路との間に配置され、前記キャリア位相推定回路から出力される前記逆補償信号に対して、前記最尤系列復号器の前記遅延と前記帰還歪付加フィルタの前記遅延との和に等しい量の遅延を与える位相逆補償信号遅延回路と
を備える請求項3に記載のデジタル信号処理装置。 - 前記デジタル信号に基づいて、光伝送リンクにおける伝送の過程で前記光信号が受ける帯域制限幅およびロールオフ形状を推定して、帯域制限を特徴付けるパラメータを生成して出力する帯域制限推定回路をさらに備え、
前記最尤系列復号器は、前記対照信号の生成に前記パラメータを利用し、
前記帰還歪付加フィルタは、前記パラメータに対応する帯域制限歪を前記信号再生器の前記出力信号に付加する
請求項1に記載のデジタル信号処理装置。 - 光ファイバ通信用途の光信号受信機に含まれ、光電変換装置およびアナログデジタルコンバータによって、偏波多重された光信号から変換された2レーンの複素形式のデジタル電気信号を処理するデジタル信号処理装置であって、
2つの入力と2つの出力との間を外部から制御可能な伝達関数を持つ4つの線形適応フィルタを介してストレート及びクロスに接続可能な構成を有し、前記2レーンのデジタル電気信号を入力し、X側及びY側の2レーンのデジタル信号を出力するバタフライフィルタと、
前記バタフライフィルタのX側出力信号を入力し、複数の信号系列候補に伝送路モデルの伝達関数を作用させてX側の複数の対照信号を生成し、入力された前記バタフライフィルタの前記X側出力信号と前記X側の対照信号との差分を評価することで最も尤もらしい送信信号時間系列を推定する最尤系列推定により、X側受信信号を復号するX側最尤系列復号ブロックと、
前記バタフライフィルタのY側出力信号を入力し、複数の信号系列候補に伝送路モデルの伝達関数を作用させてY側の複数の対照信号を生成し、入力された前記バタフライフィルタの前記Y側出力信号と前記Y側の対照信号との差分を評価することで最も尤もらしい送信信号時間系列を推定する最尤系列推定により、Y側受信信号を復号するY側最尤系列復号ブロックと、
前記X側最尤系列復号ブロックのX偏波同相成分及びX偏波直交位相成分の出力データを入力し、X側の信号を再生して出力するX側信号再生ブロックと、
前記Y側最尤系列復号ブロックのY偏波同相成分及びY偏波直交位相成分の出力データを入力し、Y側の信号を再生して出力するY側信号再生ブロックと、
前記X側信号再生ブロックの出力信号を入力し、入力された前記X側信号再生ブロックの前記出力信号に対して、前記X側最尤系列復号ブロックで用いられる前記伝送路モデルと等価な歪を付加して出力するX側帰還歪付加フィルタと、
前記Y側信号再生ブロックの出力信号を入力し、入力された前記Y側信号再生ブロックの前記出力信号に対して、前記Y側最尤系列復号ブロックで用いられる前記伝送路モデルと等価な歪を付加して出力するY側帰還歪付加フィルタと、
前記X側帰還歪付加フィルタの出力信号をX側ターゲット信号として入力し、前記バタフライフィルタへのX側入力信号及びY側入力信号を、入力側X信号及び入力側Y信号として入力し、前記X側ターゲット信号と前記入力側X信号の差、及び前記X側ターゲット信号と前記入力側Y信号との差を誤差信号として、LMS(Least Mean Squared)アルゴリズムにより、前記バタフライフィルタのX側出力へ結線された2つの線形適応フィルタのタップ係数を更新するX側適応等化フィルタ制御ブロックと、
前記Y側帰還歪付加フィルタの出力信号をY側ターゲット信号として入力し、前記入力側Y信号及び前記入力側X信号を入力し、前記Y側ターゲット信号と前記入力側Y信号との差、及び前記Y側ターゲット信号と前記入力側X信号との差を誤差信号として、前記LMSアルゴリズムにより、前記バタフライフィルタのY側出力へ結線された2つの線形適応フィルタのタップ係数を更新するY側適応等化フィルタ制御ブロックと
を備えるデジタル信号処理装置。 - 前記バタフライフィルタへの前記X側入力信号に対して、前記X側最尤系列復号ブロックの遅延と前記X側帰還歪付加フィルタの遅延との和に等しい量の遅延を与え、前記X側適応等化フィルタ制御ブロックの前記入力側X信号、及び前記Y側適応等化フィルタ制御ブロックの前記入力側X信号として供給するX入力側遅延回路と、
前記バタフライフィルタへの前記Y側入力信号に対して、前記Y側最尤系列復号ブロックの遅延と前記Y側帰還歪付加フィルタの遅延との和に等しい量の遅延を与え、前記Y側適応等化フィルタ制御ブロックの前記入力側Y信号、及び前記X側適応等化フィルタ制御ブロックの前記入力側Y信号として供給するY入力側遅延回路と
を更に備える請求項6に記載のデジタル信号処理装置。 - 前記バタフライフィルタの前記X側出力信号を入力し、キャリア位相オフセットを推定し、X側位相補償信号及びX側位相逆補償信号を出力するX側キャリア位相推定ブロックと、
前記バタフライフィルタの前記X側出力信号を主信号として入力し、前記X側キャリア位相推定ブロックからの前記X側位相補償信号を入力し、前記バタフライフィルタの前記X側出力信号に前記X側位相補償信号を作用させた上で、前記X側最尤系列復号ブロックに供給するX側位相補償回路と、
前記X側帰還歪付加フィルタからの前記出力信号を主信号として入力し、前記X側キャリア位相推定ブロックからの前記X側位相逆補償信号を入力し、前記X側帰還歪付加フィルタからの前記出力信号に前記X側位相逆補償信号を作用させた上で前記X側適応等化フィルタ制御ブロックの前記X側ターゲット信号として出力するX側位相逆補償回路と、
前記バタフライフィルタの前記Y側出力信号を入力し、キャリア位相オフセットを推定し、Y側位相補償信号及びY側位相逆補償信号を出力するY側キャリア位相推定ブロックと、
前記バタフライフィルタの前記Y側出力信号を主信号として入力し、前記Y側キャリア位相推定ブロックからの前記Y側位相補償信号を入力し、前記バタフライフィルタの前記Y側出力信号に前記Y側位相補償信号を作用させた上で、前記Y側最尤系列復号ブロックに供給するY側位相補償回路と、
前記Y側帰還歪付加フィルタからの前記出力信号を主信号として入力し、前記Y側キャリア位相推定ブロックからの前記Y側位相逆補償信号を入力し、前記Y側帰還歪付加フィルタからの前記出力信号に前記Y側位相逆補償信号を作用させた上で前記Y側適応等化フィルタ制御ブロックの前記Y側ターゲット信号として出力するY側位相逆補償回路と
を更に備える請求項6に記載のデジタル信号処理装置。 - 前記バタフライフィルタの前記X側出力信号に対して、前記X側キャリア位相推定ブロックの遅延に等しい量の遅延を与え、前記X側位相補償回路へ出力するX側主信号遅延回路と、
前記バタフライフィルタの前記Y側出力信号に対して、前記Y側キャリア位相推定ブロックの遅延に等しい量の遅延を与え、前記Y側位相補償回路へ出力するY側主信号遅延回路と、
前記X側キャリア位相推定ブロックの前記X側位相逆補償信号に対して、前記X側最尤系列復号ブロックの遅延と前記X側帰還歪付加フィルタの遅延との和に等しい量の遅延を与え、前記X側位相逆補償回路へ出力するX側位相逆補償信号遅延回路と、
前記Y側キャリア位相推定ブロックの前記Y側位相逆補償信号に対して、前記Y側最尤系列復号ブロック遅延と前記Y側帰還歪付加フィルタの遅延との和に等しい量の遅延を与え、前記Y側位相逆補償回路へ出力するY側位相逆補償信号遅延回路と、
前記バタフライフィルタへの前記X側入力信号に対して、前記X側主信号遅延回路の前記遅延と前記X側最尤系列復号ブロックの前記遅延と前記X側帰還歪付加フィルタの前記遅延との和に等しい量の遅延を与え、前記X側適応等化フィルタ制御ブロック及び前記Y側適応等化フィルタ制御ブロックの前記入力側X信号として出力するX入力側遅延回路と、
前記バタフライフィルタへの前記Y側入力信号に対して、前記Y側主信号遅延回路の前記遅延と前記Y側最尤系列復号ブロックの前記遅延と前記Y側帰還歪付加フィルタの前記遅延との和に等しい量の遅延を与え、前記Y側適応等化フィルタ制御ブロック及び前記X側適応等化フィルタ制御ブロックの前記入力側Y信号として出力するY入力側遅延回路と
を更に備える請求項8に記載のデジタル信号処理装置。 - 前記アナログデジタルコンバータから出力される前記デジタル電気信号を入力し、X偏波チャネルとY偏波チャネルの間のクロストークを除去し、偏波分離されたX側信号およびY側信号を生成する1タップバタフライフィルタと、
前記1タップバタフライフィルタから出力される前記X側信号および前記Y側信号の周波数オフセットを補償し、周波数オフセット補償されたX側信号およびY側信号を前記バタフライフィルタへ出力する周波数オフセット補償回路と
をさらに具備する請求項6に記載のデジタル信号処理装置。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/983,233 US8977141B2 (en) | 2011-02-07 | 2012-02-07 | Digital signal processing apparatus |
CN201280007549.2A CN103460659B (zh) | 2011-02-07 | 2012-02-07 | 数字信号处理装置 |
JP2012556895A JP5671071B2 (ja) | 2011-02-07 | 2012-02-07 | デジタル信号処理装置 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011024022 | 2011-02-07 | ||
JP2011-024022 | 2011-02-07 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2012108421A1 true WO2012108421A1 (ja) | 2012-08-16 |
Family
ID=46638634
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2012/052724 WO2012108421A1 (ja) | 2011-02-07 | 2012-02-07 | デジタル信号処理装置 |
Country Status (4)
Country | Link |
---|---|
US (1) | US8977141B2 (ja) |
JP (1) | JP5671071B2 (ja) |
CN (1) | CN103460659B (ja) |
WO (1) | WO2012108421A1 (ja) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2015056753A (ja) * | 2013-09-11 | 2015-03-23 | 富士通株式会社 | 非線形歪み補償装置及び方法並びに光受信器 |
JP2016521027A (ja) * | 2013-03-30 | 2016-07-14 | ゼットティーイー コーポレーションZte Corporation | 直交位相シフトキーイング変調された光信号からのデータ復元 |
JP2017516378A (ja) * | 2014-01-08 | 2017-06-15 | アルカテル−ルーセント | 光信号の受信機用の装置、方法およびコンピュータプログラム |
US9787403B2 (en) | 2013-04-04 | 2017-10-10 | Nec Corporation | Digital optical transmitter, optical communication system using the same, and digital optical transmission method |
US10171200B2 (en) | 2013-08-27 | 2019-01-01 | Zte Corporation | Optical communication using super-nyquist signals |
WO2019043748A1 (ja) * | 2017-08-28 | 2019-03-07 | 三菱電機株式会社 | 適応等化フィルタ及び信号処理装置 |
US10608743B2 (en) | 2016-06-17 | 2020-03-31 | Ntt Electronics Corporation | Transmission characteristic compensation apparatus, transmission characteristic compensation method, and communication apparatus |
WO2020175014A1 (ja) * | 2019-02-28 | 2020-09-03 | 日本電信電話株式会社 | 信号処理方法、信号処理装置及び通信システム |
US12126364B2 (en) | 2021-08-18 | 2024-10-22 | Nec Corporation | Delta-sigma modulation apparatus, delta-sigma modulation method, and recording medium |
Families Citing this family (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013183455A (ja) * | 2012-02-29 | 2013-09-12 | Zte (Usa) Inc | ナイキスト波長分割多重システム |
CN103312414B (zh) * | 2012-03-16 | 2016-03-30 | 富士通株式会社 | 一种逆信道装置和包含该装置的发射机、接收机及系统 |
US8923460B1 (en) * | 2013-02-14 | 2014-12-30 | Pmc-Sierra Us, Inc. | Methods and apparatus for signal equalization in dual polarization multiplexed optical communication systems |
EP2804334A1 (en) * | 2013-05-13 | 2014-11-19 | Xieon Networks S.à.r.l. | Method, device and communication system for reducing optical transmission impairments |
US9306618B2 (en) * | 2013-09-16 | 2016-04-05 | Scintera Networks Llc | Filter enhancer and method |
JP6131831B2 (ja) * | 2013-11-06 | 2017-05-24 | 富士通株式会社 | 光受信器および光受信方法 |
US9240754B2 (en) | 2013-12-30 | 2016-01-19 | Qualcomm Technologies International, Ltd. | Frequency fine tuning |
US9442141B2 (en) * | 2014-01-08 | 2016-09-13 | Qualcomm Technologies International, Ltd. | Analogue-to-digital converter |
US9391563B2 (en) | 2013-12-30 | 2016-07-12 | Qualcomm Technologies International, Ltd. | Current controlled transconducting inverting amplifiers |
US20150222470A1 (en) * | 2014-01-31 | 2015-08-06 | Broadcom Corporation | Point to Point Split Mount Impairment Correction System |
AT516086A1 (de) * | 2014-07-23 | 2016-02-15 | Siemens Ag Oesterreich | Verfahren und Vorrichtung zur Ermittlung der Absolutgeschwindigkeit eines Schienenfahrzeugs |
US10038503B2 (en) * | 2014-08-13 | 2018-07-31 | Xilinx, Inc. | Adaptive optical channel compensation |
EP3202056B1 (en) * | 2014-10-02 | 2019-06-19 | B.G. Negev Technologies & Applications Ltd., at Ben-Gurion University | All-optical silicon-photonic constellation conversion of amplitude-phase modulation formats |
EP3219062B1 (en) | 2014-11-13 | 2021-06-02 | Telefonaktiebolaget LM Ericsson (publ) | Digital signal processing of an optical communications signal in a coherent optical receiver |
US9912414B2 (en) | 2014-11-14 | 2018-03-06 | Zte Corporation | Iterative post-equalization for coherent optical receivers |
US20180183542A1 (en) * | 2015-06-29 | 2018-06-28 | Mitsubishi Electric Corporation | Optical transmission device, optical reception device, optical transmission system, and optical transmission method |
US9998235B2 (en) * | 2016-01-08 | 2018-06-12 | Google Llc | In-band optical interference mitigation for direct-detection optical communication systems |
US9887784B1 (en) * | 2016-09-28 | 2018-02-06 | Intel Corporation | Compensation of a frequency disturbance in a digital phase lock loop |
US10128958B1 (en) | 2016-10-24 | 2018-11-13 | Inphi Corporation | Forward and backward propagation methods and structures for coherent optical receiver |
US10887022B2 (en) * | 2017-06-15 | 2021-01-05 | Nokia Of America Corporation | Backward propagation with compensation of some nonlinear effects of polarization mode dispersion |
JP7236392B2 (ja) * | 2017-11-27 | 2023-03-09 | ソニーセミコンダクタソリューションズ株式会社 | 復号装置、復号方法 |
WO2019142912A1 (ja) * | 2018-01-19 | 2019-07-25 | 日本電信電話株式会社 | シンボル判定装置及びシンボル判定方法 |
CN110365428B (zh) * | 2018-04-09 | 2021-09-07 | 富士通株式会社 | 损伤监测装置、损伤监测及补偿系统及方法 |
WO2020039519A1 (ja) * | 2018-08-22 | 2020-02-27 | 三菱電機株式会社 | 光受信機、光信号受信方法及びデータ再生装置 |
US10560289B1 (en) | 2018-09-13 | 2020-02-11 | Viasat, Inc. | Adaptive equalizer system |
JP7252447B2 (ja) * | 2019-05-08 | 2023-04-05 | 日本電信電話株式会社 | シンボル判定装置、及びシンボル判定方法 |
JP7055268B2 (ja) * | 2020-05-28 | 2022-04-18 | Nttエレクトロニクス株式会社 | 適応等化器、適応等化方法及び光通信システム |
US11038549B1 (en) * | 2020-06-08 | 2021-06-15 | Ciena Corporation | Receiver noise loading for control loop stability |
CN113938199A (zh) * | 2020-06-29 | 2022-01-14 | 中兴通讯股份有限公司 | 一种信号的处理方法及装置、存储介质、电子装置 |
CN115174330B (zh) * | 2022-06-22 | 2023-08-25 | 苏州大学 | 一种多载波接入网失真信号的补偿方法及非线性均衡器 |
CN116582187B (zh) * | 2023-07-11 | 2023-09-22 | 深圳市光为光通信科技有限公司 | 基于线性直驱的光电通信模块自适应编码解码方法 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH03284013A (ja) * | 1990-03-30 | 1991-12-13 | Nec Corp | 適応型最尤系列推定方式及び装置 |
JPH04362808A (ja) * | 1991-06-11 | 1992-12-15 | Hitachi Ltd | 等化回路 |
JPH05152894A (ja) * | 1991-07-29 | 1993-06-18 | Oki Electric Ind Co Ltd | 適応等化器 |
JP2000269865A (ja) * | 1999-03-17 | 2000-09-29 | Pioneer Electronic Corp | ディジタル信号受信システムにおける信号処理回路 |
WO2009060920A1 (ja) * | 2007-11-09 | 2009-05-14 | Hitachi Communication Technologies, Ltd. | 光電界送信器及び光電界伝送システム |
JP2010068029A (ja) * | 2008-09-08 | 2010-03-25 | Sumitomo Electric Ind Ltd | 光トランシーバ |
WO2010082344A1 (ja) * | 2009-01-16 | 2010-07-22 | 三菱電機株式会社 | 光変復調システム、光伝送システムおよび光変復調方法 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2546022B2 (ja) * | 1990-03-30 | 1996-10-23 | 日本電気株式会社 | 適応型通信路インパルスレスポンス推定方式 |
JP2986261B2 (ja) * | 1991-10-14 | 1999-12-06 | 沖電気工業株式会社 | 適応最尤系列推定器 |
JP4701928B2 (ja) * | 2005-09-01 | 2011-06-15 | 株式会社日立製作所 | 光電界波形観測装置 |
US8983309B2 (en) * | 2012-02-13 | 2015-03-17 | Ciena Corporation | Constrained continuous phase modulation and demodulation in an optical communications system |
-
2012
- 2012-02-07 US US13/983,233 patent/US8977141B2/en active Active
- 2012-02-07 WO PCT/JP2012/052724 patent/WO2012108421A1/ja active Application Filing
- 2012-02-07 JP JP2012556895A patent/JP5671071B2/ja active Active
- 2012-02-07 CN CN201280007549.2A patent/CN103460659B/zh active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH03284013A (ja) * | 1990-03-30 | 1991-12-13 | Nec Corp | 適応型最尤系列推定方式及び装置 |
JPH04362808A (ja) * | 1991-06-11 | 1992-12-15 | Hitachi Ltd | 等化回路 |
JPH05152894A (ja) * | 1991-07-29 | 1993-06-18 | Oki Electric Ind Co Ltd | 適応等化器 |
JP2000269865A (ja) * | 1999-03-17 | 2000-09-29 | Pioneer Electronic Corp | ディジタル信号受信システムにおける信号処理回路 |
WO2009060920A1 (ja) * | 2007-11-09 | 2009-05-14 | Hitachi Communication Technologies, Ltd. | 光電界送信器及び光電界伝送システム |
JP2010068029A (ja) * | 2008-09-08 | 2010-03-25 | Sumitomo Electric Ind Ltd | 光トランシーバ |
WO2010082344A1 (ja) * | 2009-01-16 | 2010-07-22 | 三菱電機株式会社 | 光変復調システム、光伝送システムおよび光変復調方法 |
Non-Patent Citations (1)
Title |
---|
MI LIN ET AL.: "40 Gb/s Pol-Mux RZ-DQPSK Transmission with Electric Dispersion Compensation", 2010 12TH INTERNATIONAL CONFERENCE ON TRANSPARENT OPTICAL NETWORKS (ICTON), 1 July 2010 (2010-07-01), pages 1 - 3 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9698914B2 (en) | 2013-03-30 | 2017-07-04 | Zte Corporation | Recovering data from quadrature phase shift keying modulated optical signals |
JP2016521027A (ja) * | 2013-03-30 | 2016-07-14 | ゼットティーイー コーポレーションZte Corporation | 直交位相シフトキーイング変調された光信号からのデータ復元 |
US9787403B2 (en) | 2013-04-04 | 2017-10-10 | Nec Corporation | Digital optical transmitter, optical communication system using the same, and digital optical transmission method |
US10171200B2 (en) | 2013-08-27 | 2019-01-01 | Zte Corporation | Optical communication using super-nyquist signals |
JP2015056753A (ja) * | 2013-09-11 | 2015-03-23 | 富士通株式会社 | 非線形歪み補償装置及び方法並びに光受信器 |
JP2017516378A (ja) * | 2014-01-08 | 2017-06-15 | アルカテル−ルーセント | 光信号の受信機用の装置、方法およびコンピュータプログラム |
US10142031B2 (en) | 2014-01-08 | 2018-11-27 | Alcatel Lucent | Apparatus, method and computer program for a receiver of an optical signal |
US10608743B2 (en) | 2016-06-17 | 2020-03-31 | Ntt Electronics Corporation | Transmission characteristic compensation apparatus, transmission characteristic compensation method, and communication apparatus |
WO2019043748A1 (ja) * | 2017-08-28 | 2019-03-07 | 三菱電機株式会社 | 適応等化フィルタ及び信号処理装置 |
JPWO2019043748A1 (ja) * | 2017-08-28 | 2019-12-19 | 三菱電機株式会社 | 適応等化フィルタ及び信号処理装置 |
US10985845B2 (en) | 2017-08-28 | 2021-04-20 | Mitsubishi Electric Corporation | Adaptive equalization filter and signal processing device |
WO2020175014A1 (ja) * | 2019-02-28 | 2020-09-03 | 日本電信電話株式会社 | 信号処理方法、信号処理装置及び通信システム |
JP2020141294A (ja) * | 2019-02-28 | 2020-09-03 | 日本電信電話株式会社 | 信号処理方法、信号処理装置及び通信システム |
JP7128420B2 (ja) | 2019-02-28 | 2022-08-31 | 日本電信電話株式会社 | 信号処理方法、信号処理装置及び通信システム |
US12126364B2 (en) | 2021-08-18 | 2024-10-22 | Nec Corporation | Delta-sigma modulation apparatus, delta-sigma modulation method, and recording medium |
Also Published As
Publication number | Publication date |
---|---|
US8977141B2 (en) | 2015-03-10 |
JP5671071B2 (ja) | 2015-02-18 |
JPWO2012108421A1 (ja) | 2014-07-03 |
CN103460659A (zh) | 2013-12-18 |
US20130308960A1 (en) | 2013-11-21 |
CN103460659B (zh) | 2016-03-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5671071B2 (ja) | デジタル信号処理装置 | |
US10243655B2 (en) | Method and system for optical impairment mitigation for high-speed optical communication systems | |
Kikuchi | Digital coherent optical communication systems: Fundamentals and future prospects | |
Kikuchi | Fundamentals of coherent optical fiber communications | |
Ip et al. | Digital equalization of chromatic dispersion and polarization mode dispersion | |
Winzer et al. | Spectrally efficient long-haul optical networking using 112-Gb/s polarization-multiplexed 16-QAM | |
Spinnler | Equalizer design and complexity for digital coherent receivers | |
US9698914B2 (en) | Recovering data from quadrature phase shift keying modulated optical signals | |
WO2015027903A1 (en) | Optical communication using super-nyquist signals | |
US10148363B2 (en) | Iterative nonlinear compensation | |
US8666251B2 (en) | Electronic dispersion compensation system and method | |
US9608735B2 (en) | MIMO equalization optimized for baud rate clock recovery in coherent DP-QPSK metro systems | |
CN113708845B (zh) | 均衡解复用方法、装置、电子设备及计算机可读存储介质 | |
Chagnon et al. | Half-terabit single-carrier direct-detect transceiver, formats, and DSP: analysis and demonstration | |
Chen et al. | Comparison of advanced detection techniques for QPSK signals in super-Nyquist WDM systems | |
Reza et al. | Blind nonlinearity mitigation of 10G DMLs using sparse Volterra equalizer in IM/DD PAM-4 transmission systems | |
Zhou et al. | Digital signal processing for coherent optical communication | |
Zhang et al. | Training symbol-based equalization for quadrature duobinary PDM-FTN systems | |
Gorshtein et al. | Advanced modulation formats and digital signal processing for fiber optic communication | |
Rosa et al. | 112 Gb/s DP-QPSK coherent optical transmission over 3000 km using an complete set of digital signal processing algorithms | |
Detwiler | Continuous phase modulation for high speed fiber-optic links | |
WO2023067641A1 (ja) | デジタル信号処理回路、方法、受信機、及び通信システム | |
Paskov et al. | Blind adaptive equalization of chromatic dispersion for PDM-QPSK | |
Savory et al. | Ultra Long-Haul QPSK Transmission using a Digital Coherent Receiver | |
JP2024055386A (ja) | 通信システム、受信機、補償量計算装置、及び歪み補償方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201280007549.2 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 12745108 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2012556895 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 13983233 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 12745108 Country of ref document: EP Kind code of ref document: A1 |