JP2013225762A - Coherent optical transmission system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a coherent optical transmission system which stabilizes a one-bit delay at all times and eliminates differential detection penalty, thereby exhibiting excellent stability and high reception sensitivity.SOLUTION: The present invention relates to a coherent optical transmission system which transmits data by a data series of I and Q components of a frame composed of a plurality of bits. The coherent optical transmission system comprises: an optical transmission unit which sends out signal light consisting of continuous light modulated by differential coding to a fiber optic transmission line; an optical receiving unit which branches the signal light from the fiber optic transmission line into two and performs for each of the I and the Q components interaction between the signal light and a signal light delayed by one bit to decode the transmitted data into an electrical signal; parallel band-pass filters having a plurality of frames input as a series and including band-pass filters provided for the zero frequency of the electrical signal and each of integral-multiplied penetrated frequency of the frame period; and a control unit which calculates an intensity ratio between the intensity of a center frequency electrical signal and the intensity of a penetrated frequency electrical signal. The control unit controls the one-bit delay applied by the optical receiving unit so that the intensity ratio becomes minimum for each of the I and the Q components.

Description

  The present invention relates to an optical transmission system in optical fiber communication, and more particularly to a coherent optical transmission system based on a scheme for transmitting data in units of frames.

  In coherent optical communication, information is transmitted by shifting the phase or frequency of signal light to be transmitted, and information is received by a differential detection method based on optical interference. Here, the method of transmitting and receiving information by shifting the phase of signal light is called phase shift keying (PSK: phase shift keying). A method of transmitting and receiving information by shifting the frequency is called frequency shift keying, that is, frequency shift keying (FSK).

  In general, the differential detection method based on optical interference is different from the intensity modulation or direct detection method (IM-DD: Intensity Modulation-Direct Detection). Therefore, the sensitivity can be improved up to 3 dB. Therefore, when coherent optical communication is also used, there is a possibility that the transmission distance of the optical fiber communication can be extended as compared with the intensity modulation or the direct detection method, which has been studied since the 1980s.

  In this differential detection method, when receiving a signal by coherent optical communication, a PSK (Phase Shift Keying) method that receives the signal light by interference between the local light emitted from the local light source and the local light source There is a DPSK (differential phase shift keying) method in which signal light differentially coded on the transmission side is received by delayed interference without using local light.

  Also, in the differential detection method, in order to increase the transmission capacity in communication, multi-level phase modulation, for example, quaternary phase modulation (QPSK), or differential quaternary phase shift keying (DQPSK) is used. shift keying) has been studied since the 1980s (see, for example, Patent Document 1, Non-Patent Document 1, Non-Patent Document 2, and Non-Patent Document 3).

  However, in the above-described DPSK and DQPSK, when the previous bit is delayed by 1 bit and the previous bit is used as the reference light, the differential phase is caused by a sensitivity decrease due to phase noise or the like in the transmission path, so-called differential detection penalty. There is a limitation that a sensitivity improvement of 3 dB at the maximum by detection cannot be achieved. In order to eliminate this restriction, there is a method of performing difference detection including not only the immediately preceding symbol but also the previous symbol (see, for example, Patent Document 2 and Non-Patent Document 4).

Further, in order to control the delay amount of the phase-modulated data and suppress the deterioration of the signal light, the light source light emitted from the light source is modulated using the known pattern signal.
Then, the power difference between the upper frequency component and the lower frequency component of the modulation signal obtained by modulating the light source light with the known pattern signal is calculated.
There is an optical modulation device configured to control the delay amount and suppress the deterioration of signal light by minimizing the power difference (see, for example, Patent Document 3).

Japanese Patent No. 4558271 US Patent Application Publication No. 2007/0201879 JP 2011-197436 A

TG. HODGKINSON, “Receiver Analysis for Synchronous Coherent Optical Fiber Transmission System”, Journal of Lightwave Technology vol. LT-5, no. 4, pp. 573-586 (1987) T.A. CHIKAMA, S.A. WATANABE, T.A. NAITO, H. ONAKA, T.A. KIYONAGA, Y. ONODA, H.C. MIYATA, M.M. SUYAMA, M.M. SEINO, andH. KUWAHARA, “Modulation and Demodulation Technologies in Optical Heterodyne PSK Transmission Systems”, Journal of Lightwave Technology vol. 8, no. 3, pp. 309-322 (1990) S. YAMAZAKI and K.K. EMURA, “Feasibility Study on QPSK Optical-Heterodyne Detection System” Journal of Lightwave Technology, vol. 8, no. 11, pp. 1646-1653 (1990) Harry Leib, “Data-Aided Noncoherent Demodulation of DPSK,” IEEE Transactions on Communications, vol. 43, no. 2/3/4, pp. 722-725 (1995)

In the differential detection method described above, since the method using differential coding does not include a local light source on the reception side, the device on the reception side can be reduced in size as compared with the method using the local light source.
However, in the case of coherent communication using differential coding, when receiving signal light, 1-bit delayed interference of signal light using signal light one bit before as reference light is used.
In order to stabilize the reception sensitivity due to this 1-bit delay interference, it is necessary to control the signal light to be delayed by 1 bit with high accuracy.
In this 1-bit delay control, in Patent Document 3 described above, in order to stabilize the 1-bit delay of the signal light on the receiving side, the 1-bit delay amount of the signal light is corrected by using a signal light of a known pattern. ing.

For this reason, in order to correct the bit delay amount, it is necessary to temporarily stop transmission of signal light corresponding to transmission data and transmit / receive signal light of a known pattern for performing 1-bit delay. There is a problem that data transmission is hindered.
Therefore, when a known pattern is used, the amount of delay cannot be monitored at all times, so that it is difficult to maintain stability at all times.
Furthermore, in order to calculate the upper frequency component of the signal light, a band pass filter in a high frequency band is required, high-speed electric signal processing is required, and there is a problem that the burden on the electronic circuit is increased.

  On the other hand, Patent Document 2 and Non-Patent Document 4 perform delayed interference using not only the immediately preceding bit but also a plurality of bits as reference light in order to eliminate a difference detection penalty due to phase noise superimposed on a transmission line or the like. For this reason, Patent Document 2 and Non-Patent Document 4 require a plurality of delayed bits, a complicated feedback circuit that interferes with a target bit, and a high-speed logic determination circuit. Scale increases. That is, a circuit that stabilizes a delay of a plurality of bits is required in addition to a circuit that performs a 1-bit delay. Since a circuit for accurately delaying a plurality of bits in units of bits is required in addition to stabilizing the 1-bit delay, the circuit becomes large.

  The present invention has been made in view of such circumstances, and with one electronic circuit having a simple configuration, the 1-bit delay is always stabilized and the difference detection penalty is eliminated at the same time. An object is to provide a technique for realizing a highly sensitive coherent optical transmission system.

  The present invention has been made in order to solve the above-described problems. The coherent optical transmission system of the present invention includes a data sequence encoded in units of frames, and the frame includes transmission data composed of a plurality of bits. A coherent optical transmission system in which data transmission is performed, an optical transmission unit that generates a signal light by phase-modulating continuous light by differential coding corresponding to the transmission data, and sends the signal light to an optical fiber transmission line; The signal light is received from the optical fiber transmission line, the signal light is bifurcated into two propagation paths, and the delay between the signal light of one propagation path and the signal light of the other propagation path subjected to 1-bit delay An optical receiver that performs interference, demodulates the transmission data, and outputs the result as an electrical signal; and a plurality of the frames in the electrical signal supplied from the optical receiver. Parallel band-pass filter configured as a plurality of band-pass filters that are input as one series and have a transmission frequency that is a zero frequency of the electrical signal and each frequency that is an integral multiple of the frame frequency. And a control unit that obtains an intensity ratio between the intensity of the electrical signal at the center frequency and the intensity of the electrical signal at the integral multiple of the transmission frequency, and the control unit minimizes the intensity ratio. A delay time control unit for delaying the signal light provided in a propagation path for performing 1-bit delay of the optical receiving unit is controlled.

  In the coherent optical transmission system of the present invention, the optical receiver further includes a phase controller, and the controller is configured to transmit any two transmission frequencies from the transmission frequencies supplied from the parallel bandpass filter. The phase of the electrical signal is detected, and the phase of the center frequency is obtained by extrapolating the phase, the phase difference between the obtained phase and a preset phase is obtained, and the phase difference is minimized. Thus, the phase control unit is controlled.

  The coherent optical transmission system according to the present invention is characterized in that the delay time control unit and the phase control unit are provided in the other propagation path that performs 1-bit delay of the optical reception unit.

  The coherent optical transmission system of the present invention further includes a branching unit provided between the optical receiving unit and the parallel bandpass filter, for branching the electrical signal into a first path and a second path connected to an external circuit. And outputting the electrical signal of the first path to the band-pass filter, and outputting the electrical signal of the second path to an external circuit.

  The coherent optical transmission system of the present invention further includes an AD conversion unit that digitizes an electrical signal output from the optical receiving unit.

  The coherent optical transmission system of the present invention is provided between the optical receiving unit and the AD conversion unit, and amplifies the electric signal having a different signal amplitude to an output signal having a constant amplitude and performs waveform shaping. It further has an amplifying part for adjusting the voltage level.

  In the coherent optical transmission system of the present invention, the branching unit is provided with a plurality of buffers for storing the electrical signal for each frame in the sequence, and the branching unit transmits the electrical signal of the frame at a constant cycle. Is output to the parallel bandpass filter.

  In the coherent optical transmission system of the present invention, the transmission unit includes a modulator composed of a Mach-Zehnder waveguide, and the transmission unit outputs modulation signals whose polarities are inverted with respect to each arm of the Mach-Zehnder waveguide. It is characterized by applying.

  The coherent optical transmission system of the present invention includes a polarization control unit that makes a polarization direction of signal light incident on the optical receiving unit constant between an output end of the optical fiber transmission path and an incident end of the optical receiving unit. It further has these.

  In the coherent optical transmission system of the present invention, the polarization separation for separating the signal light emitted from the output end into the orthogonal first polarization direction and the second polarization direction with respect to the output end of the optical fiber transmission line. And a polarization rotation unit that changes the signal light in the second polarization direction to the first polarization direction, and the polarization separation unit converts the signal light in the first polarization direction to the optical reception unit. On the other hand, the signal light in the second polarization direction is emitted to the polarization rotation unit.

  According to the present invention, the coherent optical transmission system having excellent stability and high reception sensitivity can be realized by simultaneously stabilizing the 1-bit delay and eliminating the difference detection penalty at the same time by an electronic circuit having a simple configuration. be able to.

1 is a schematic block diagram illustrating a configuration example of a coherent optical transmission system (first embodiment) illustrating the principle of the invention. It is a figure which shows the simulation result of the eye pattern of the signal light at the time of transmitting the transmission data defined by the numerical value of this bit period Tb and frame period Tf from the optical transmission part 2. It is a figure which shows the power spectrum of the signal beam | light transmitted to the optical fiber transmission line 3 from the optical transmission part 2 of FIG. FIG. 3 is a diagram illustrating a war spectrum obtained by performing Fourier transform on an electrical signal output from a branching unit 7 after delay interference is performed in a case where the bit delay is 1 bit in the optical receiving unit 4 of FIG. 1. FIG. 3 is a diagram illustrating a war spectrum obtained by performing Fourier transform on an electrical signal output from a branching unit 7 after delay interference is performed in a case where the bit delay is 0.75 bits in the optical receiving unit 4 of FIG. 1. . FIG. 3 is a diagram showing a war spectrum obtained by performing a Fourier transform on an electric signal output from a branching unit 7 after delay interference is performed in a case where the bit delay is 0.5 bit in the optical receiving unit 4 of FIG. 1. . FIG. 4 is a diagram illustrating a phase spectrum of signal light transmitted from the optical transmission unit 2. It is a figure which shows the phase spectrum of the signal beam | light after delay interference output from the branch part. It is a figure which shows the structural example of the optical transmission part 2 in FIG. It is a figure which shows the structural example of the receiving / control part 18 in FIG. It is a figure which shows the structural example of the optical receiver 30 in FIG. It is a figure which shows the structural example of the optical receiver 4 in FIG. It is a figure for demonstrating the polarization control part 50 which hold | maintains the polarization | polarized-light of the signal light which injects into the optical receiving part 4 from the optical fiber transmission line 3 uniformly. It is a figure which shows the structure of the coherent optical transmission system which has an independent receiving system with respect to two orthogonal polarization waves.

<Principle of the present invention>
The present invention stabilizes data reception by holding a 1-bit delayed signal light used as a reference light to be interfered with the signal light in the case of delay interference for demodulating data transmitted as differentially coded signal light in DQPSK. , And improvement in reception sensitivity by suppressing phase fluctuation of signal light. The principle of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration example of a coherent optical transmission system based on DQPSK for explaining the principle of the present invention.

In FIG. 1, a coherent optical transmission system according to the present invention encodes a frame unit and transmits / receives a data series 1, 1 ′ composed of a plurality of frames. An optical transmission unit 2, an optical fiber transmission line 3 , Optical receivers 4, 4 ′, amplifiers (for example, limiter amplifiers) 5, 5 ′, AD converters 6, 6 ′, branching units 7, 7 ′, parallel bandpass filters 8, 8 ′, and controller 9. I have. The frame is composed of a plurality of bits.
In DQPSK transmission, a base of two orthogonal components (orthogonal two components) consisting of cos (In-phase: In-phase component, hereinafter referred to as I) and sin (Quadrature-phase: orthogonal component, hereinafter referred to as Q). A band signal is used.

In the coherent optical transmission system of the present invention, each of the data sequences 1 and 1 ′ encoded in units of frames having a fixed number of bits (bit numbers i = 1,..., N) is respectively an I component, Q It is input to the optical transmitter 2 as component transmission data.
The optical transmitter 2 converts the data series 1 and 1 ′ into signal light, and sends the converted signal light to the optical fiber transmission line 3.
The optical receiver 4 differentially detects the signal light propagated through the optical fiber transmission line 3 for each of the I component and the Q component, and converts them into electrical signals of the I component and the Q component, respectively. This I component electric signal passes through the amplification unit 5 and the AD conversion unit 6 and becomes a series of digital electric signals of “0” and “1”. Similarly, the electrical signal of the Q component becomes a series of digital electrical signals of “0” and “1” through the amplification unit 5 ′ and the AD conversion unit 6 ′.

The branching unit 7 transfers the digital electrical signal sequence supplied from the AD conversion unit 6 to the data processing circuit, which is an external circuit, via the digital signal line. Each frame is grouped and output to the parallel bandpass filter 8.
The branching unit 7 ′ transfers the digital electric signal series supplied from the AD conversion unit 6 ′ to the data processing circuit, which is an external circuit, via the digital signal line, and simultaneously converts the digital electric signal series to the time. According to the order, they are grouped every M frames and output to the parallel bandpass filter 8 '.
Within each group, M frames are rearranged at equal time intervals and then input to the parallel bandpass filter 8 or 8 'for each group. The spectrum of signal light in which frames of a certain bit length are arranged at equal time intervals has a comb arranged around the center frequency at equal frequency intervals.

  In each of the parallel bandpass filters 8 (or 8 '), a plurality of comb components are extracted by the number of parallel channels. The control unit 9 detects the intensity and phase of each comb component extracted by the parallel bandpass filter 8 (or 8 '), and estimates the bit delay amount from the intensity ratio of each comb component. Moreover, the control part 9 estimates the phase in the center frequency of signal light from the phase of each comb component. The control unit 9 controls the optical receiving unit 4 (or 4 ′) based on the estimated bit delay amount and the phase at the center frequency, thereby stabilizing by holding 1 bit delay and receiving sensitivity by suppressing phase fluctuation. Make improvements. Here, since it is necessary for the control unit 9 to control each of the I component and the Q component collectively, the control unit 9 has a single configuration common to the I component and the Q component.

  As described above, the present invention detects the frequency spectrum of light in a comb shape when frames with a constant frame period Tf are continuous, and determines the power ratio (intensity ratio) and frequency component phase of a specific comb. By detecting, the 1-bit delay and the phase shift of the signal light can be estimated, and the 1-bit delay amount and the phase shift amount of the signal light are controlled. The inverse of the frame period Tf is the frame frequency. Hereinafter, the principle of stabilization by holding 1-bit delay and the improvement of reception sensitivity by controlling the phase variation of signal light according to the present invention will be described in detail using the spectrum of signal light of DQPSK transmission.

A. Principle for Stabilization by Holding 1-bit Delay The data structure for storing information to be transmitted in FIG. 1 is arranged in units of frames composed of a plurality of N bits, as in data sequence 1 in FIG. ing.
In this frame, when a bit period that is a time period of 1 bit is defined as Tb and a frame period that is a time period of 1 frame is defined as Tf, Tf = N × Tb. Here, for example, when the bit period Tb = 40 ps (picosecond) and the number of bits in one frame is N = 128, the frame period Tf = 5.12 ns (nanoseconds).

Next, FIG. 2 is a diagram showing the simulation result of the eye pattern of the signal light when the transmission data determined by the numerical values of the bit period Tb and the frame period Tf is transmitted from the optical transmission unit 2. In this figure, the vertical axis shows the optical power normalized to “1”, and the horizontal axis shows the time in the eye pattern display. The optical phase in each bit of the signal light has a binary value of 0 or π, and is NRZ (nonreturn to zero) modulated.
In this eye pattern, the intensity of light in a transmission in which one frame is 128 bits, that is, 40 Gbit / sec, that is, the frame period Tf is 40 ps, is superimposed on the sampling oscilloscope.

  In DQPSK transmission, an orthogonal two component consisting of an I component and a Q component is involved, but the I component, that is, the cos component, and the Q component, that is, the sin component, can be combined into a single sine wave component by the addition theorem of trigonometric functions. . Accordingly, the waveform of the eye pattern is different between DPSK transmission and DQPSK transmission, but the spectrums of the I component and the Q component are equal to each other. In addition, the spectrum after the combination of the I component and the Q component according to the addition theorem is equal to the spectrum of the I component and the Q component, and the following description is valid for both DQPSK transmission and DPSK transmission.

In the following explanation of the principle of the present invention, as an example, 27-1 or 127 pseudo random bit streams (PRBS) are used as bits of a frame.
Then, a 128-bit bit string obtained by adding a bit indicating “0” as the 128th bit to the 127-bit array generated by the PRBS is used as one frame in the signal sequence 1 as transmission data. .

  Here, “0” is added to the 127th bit as the last 128th bit by converting the number of elements of the signal sequence to a power of 2, thereby using a fast Fourier transform used for deriving a frequency spectrum, which will be described later. This is a convenient procedure for facilitating the process, and is not an essential matter in the optical transmission system targeted by the present invention. Further, the frequency spectrum characteristics described below and the effects of the present invention can be easily confirmed in coherent optical transmission using a frame not adding “0” as the 128th bit.

Returning to FIG. 2, when the signal light constituting each bit is continuous, the signal light displayed on the sampling oscilloscope is constant without changing the light intensity, and there is no occurrence of a dip in which the light intensity decreases. However, in order to maintain the quality of the signal light, the optical transmission unit 2 that transmits the signal light is devised with respect to the signal light string that forms the bits.
When transmitting this signal light, if the phase of the continuous light is changed for differential coding, the phase of the signal light is shifted at the boundary between the bits. Due to the phase shift of the signal light, the frequency of the signal light is also shifted.

As a result, the frequency band of the spectrum of the signal light is widened, and crosstalk occurs between two adjacent frequency channels when performing data transmission using BPSK in the wavelength division multiplexing channel. Further, due to the time spread of the signal light due to the frequency chirp, the bit period is changed from 40 psec to 60 psec, for example.
In order to suppress the spread of the frequency band of the spectrum and the time spread due to the frequency chirp, the optical transmitter 2 performs push-pull modulation.

  In this push-pull modulation, only when the phase changes between adjacent bits, the signal light disappears due to interference at the front and rear end portions where the front and rear signal light are adjacent. The extinction of the signal light is performed by applying an electric signal pair that is inverted to each other to a Mach-Zehnder modulator (consisting of a Mach-Zehnder waveguide) that modulates the phase of the signal light in the optical transmitter 2. Is called. As a result, when the phase changes between adjacent bits, as shown in FIG. 2, a dip in which the light intensity attenuates is detected in the time domain in which the light intensity changes, that is, the phase of the signal light changes.

In the present invention, since the signal light having the above-described dip is used, it is assumed that the transmitter side, that is, the optical transmission unit 2 performs push-pull modulation.
Further, when the frames are arranged and transmitted at the present time interval, that is, when they are continuous in a certain period of time, the spectrum of the signal light is detected as a comb shape in which line spectra are arranged at equal intervals.

  Next, FIG. 3 is a diagram illustrating the power spectrum of the signal light transmitted from the optical transmission unit 2 in FIG. 1 to the optical fiber transmission line 3. The horizontal axis shows the frequency shift amount with respect to the center frequency, and the vertical axis shows the logarithmic display of the intensity of the spectrum. In FIG. 3, for example, the used signal light has a bit period Tb = 40 ps, a frame bit number N = 128, a frame period Tf = 5.12 ns, and a frame number M = 16.

At this time, in the signal light in the case where the number of frames M = 16, the bit sequence of the signal light of each frame is arranged as a time series in which no time gap is provided between the frames, that is, between adjacent frames. And Then, the time series is Fourier-transformed to generate a power spectrum graph indicating the correspondence between the spectrum shown in FIG. 3 and the frequency shift amount.
FIG. 3A is a diagram in which a spectrum in a frequency range of ± 100 GHz from the center frequency is plotted with the center frequency of the signal light as the origin. In FIG. 3A, there is no carrier peak, and the envelope of the spectrum of the signal light in the frequency range of ± 100 GHz from the origin can be confirmed.

FIG. 3B is an enlarged view of the range of ± 2 GHz from the origin in FIG. 3A in the frequency shift amount on the horizontal axis. In FIG. 3 (b), it is possible to confirm a roller having a comb shape in which the spectrum of the signal light is arranged as a line spectrum by the frame having the above-described configuration. The frequency interval of the comb which is this line spectrum is equal to the reciprocal of the time period Tf = 12 ns of one frame, and is about 195 Hz.
The frequency interval of the comb is narrowed by increasing the number of bits constituting the frame without changing the bit period Tb, while the bit period Tb is not changed. When the number of bits constituting a frame is reduced, the frame period is shortened and widened.

Next, FIG. 4 shows a Fourier transform of the electrical signal output through the branching unit 7 or 7 ′ after delay interference is performed in the optical receiving unit 4 of FIG. 1 when the bit delay is 1 bit. It is a figure which shows the power spectrum obtained by doing. The horizontal axis shows the frequency shift amount with respect to the center frequency, and the vertical axis shows the logarithmic display of the intensity of the spectrum. FIG. 4B is an enlarged view of the comb shape in the range of ± 2 GHz with respect to the center frequency in FIG.
In FIG. 4, as in FIG. 3, the used signal light has a bit period Tb = 40 ps, a frame bit number N = 128, a frame period Tf = 5.12 ns, and a frame number M = 16.

FIG. 4 shows the comb shape of the power spectrum of the electric signal after passing through the branching section 7 or 7 ′ when the 1-bit delay is held. As shown in the figure, when the 1-bit delay is stably and accurately maintained, the carrier peak at the center frequency does not appear. That is, the power of the line spectrum at the center frequency cannot be larger than the line spectrum other than the center frequency.
Therefore, the power of the comb component (line spectrum) of the center frequency that is the origin is detected to be smaller than the power of the adjacent comb component.

In the case of FIG. 4, the power of the comb component at the origin is the frame period Tf = 5.12 ns, where the bit period Tb = 40 ps, the number of bits of the frame is N = 128, and the number of frames M = 16. Thus, the frequency shift is about 0.65 times that of the adjacent comb component at about 193 MHz.
Here, when the bit delay shifts from 1 bit, the power of the comb component changes, and the power ratio of the comb component changes when the frequency shift adjacent to the origin comb component is about 193 MHz.

For example, FIG. 5 shows a case where the optical signal output from the branching unit 7 or 7 ′ is subjected to Fourier transform after delay interference is performed in the optical receiving unit 4 of FIG. 1 when the bit delay is 0.75 bits. It is a figure which shows the obtained war spectrum. FIG. 5B is an enlarged view of the comb shape in the range of ± 2 GHz with respect to the center frequency in FIG.
FIG. 6 shows a war spectrum obtained by performing a Fourier transform on the electrical signal output from the branching unit 7 after the delay interference in the case where the bit delay is 0.5 bit in the optical receiving unit 4 of FIG. FIG. FIG. 6B is an enlarged view of the comb shape in the range of ± 2 GHz with respect to the center frequency in FIG.

  The power of the comb component at the origin in the power spectra of FIGS. 4, 5, and 6, as the amount of bit delay deviates from 1 bit, the comb component at the origin is compared with the case of 1 bit delay. It turns out that power becomes relatively large. That is, the power of the comb component at the origin is about 3.5 times when the bit delay is 0.75 with respect to the power of the comb component at a frequency shift adjacent to the origin of about 193 MHz. When the bit delay is 0.5, it is about 10 times.

Therefore, from the comb-shaped spectral characteristics shown in FIG. 3 to FIG. 6 described above, the ratio between the power of the comb component at the origin, that is, the center frequency, and the power of the comb component adjacent to the frequency shift amount (hereinafter simply referred to as the power ratio). It is understood that the bit delay amount can be estimated by calculating.
For this reason, the bit delay amount can be periodically estimated and controlled so as to be within a preset accuracy, and the bit delay amount can be stabilized in response to environmental changes such as temperature.
When the bit delay amount in the case of 1 bit is equal to 1 bit in advance, the power ratio of the comb component adjacent to the origin and the origin takes the minimum value, and the larger the absolute value of the deviation amount from 1 bit, the larger the power ratio Will increase. Here, the power ratio is a numerical value obtained by dividing the power of the comb component of the center frequency of the origin by the power of the adjacent comb component.

In BPSK, the bit delay amount is within a control range of ± 0.5 bits with respect to 1-bit delay. Based on the accuracy in the current optical component design and manufacturing technology, it is easy to control the bit delay amount generated in the optical component used in the optical receiver 4 within a range of ± 0.5 bits. The minimum value of the power ratio depends on fluctuations in the environmental conditions such as the characteristics and temperature of the electrical and optical components used in the optical transmitter 2.
However, by controlling the optical receiver 4 so that the power ratio is 1 or less, the fluctuation of the bit delay amount can be suppressed within ± 0.1 bits, that is, within ± 10%. As described above, according to the present invention, it is possible to stabilize the transmission quality in the coherent optical transmission system by maintaining the bit delay amount substantially at 1 bit.

Next, a method for measuring the power of the center frequency comb component and the adjacent frequency shift comb component will be described below.
As shown in FIG. 1, the parallel bandpass filter 8 is composed of three bandpass filters having different frequencies ν 0, ν 1 and ν 2, and transmits only a single comb component corresponding to each frequency. Similarly, the parallel bandpass filter 8 ′ is composed of three bandpass filters having different frequencies ν0, ν1 and ν2, and transmits only a single comb component corresponding to each frequency.
In the following description, the bit period Tb = 40 ps, the number of bits of the frame is N = 128, the frame period Tf = 5.12 ns, and the frequency shift is approximately 193 MHz for the condition of the number of frames M = 16. In this case, the frequency ν0 is 0 Hz (that is, zero frequency), the frequency ν1 is about 193 MHz, and the frequency ν2 is about 386 MHz.

  The frequency ν1 is the frequency of the comb component adjacent to the origin on the shift frequency axis, and the frequency ν2 is the frequency shift amount next to the frequency ν1, that is, the frequency of the comb component adjacent to the origin next to the frequency ν1. is there. Here, the frequency ν1 is equal to the comb frequency interval of the target signal light, and the frequency ν2 is equal to twice the comb frequency interval. Here, it is necessary to set the transmission frequency width of each bandpass filter so that only a single comb component is transmitted and the transmission frequency bands do not overlap each other. For this reason, the transmission frequency width of each bandpass filter needs to be set smaller than the comb frequency interval.

However, if the transmission frequency width is too narrow, the frequency of the signal light fluctuates due to wavelength fluctuations of the light source. As a result of the fluctuation of the frequency of the signal light, a component of the signal light that is not included in the transmission frequency width is generated, and the transmission power of the single comb component becomes unstable.
For this reason, in the present invention, for example, the transmission frequency width of each of the bandpass filters constituting each of the parallel bandpass filters 8 and 8 'is set to half of the comb frequency interval.
The configuration of the parallel bandpass filter 8 described above includes a bandpass filter 80 having the frequency ν0 as the center of the transmission frequency width, a bandpass filter 81 having the frequency ν1 as the center of the transmission frequency width, and a frequency ν2 as the center of the transmission frequency width. The band-pass filter 82 is configured. Similarly, the configuration of the parallel bandpass filter 8 ′ includes a bandpass filter 80 ′ having the frequency ν0 as the center of the transmission frequency width, a bandpass filter 81 ′ having the frequency ν1 as the center of the transmission frequency width, and the frequency ν2 as the transmission frequency width. And a band-pass filter 82 'as the center of the.

With this configuration, the control unit 9 can obtain the power of the comb component of the center frequency of the signal light in the I component by the power of the electric signal having the transmission frequency width centered on the frequency ν 0 from the band pass filter 80. On the other hand, the control unit 9 obtains the power of the comb component at the center frequency of the signal light in the Q component by the power of the electric signal having the transmission frequency width centered on the frequency ν 0 from the band pass filter 80 ′.
Further, the control unit 9 controls the power of the comb component having a frequency adjacent to the center frequency of the signal light from each of the bandpass filters 81 and 82 by the power of the electric signal having a transmission frequency width centered on the frequencies ν1 and ν2, respectively. Is obtained. Similarly, the control unit 9 generates a frequency comb adjacent to the center frequency of the signal light from each of the bandpass filters 81 ′ and 82 ′ by the power of the electric signal having a transmission frequency width centered on the frequencies ν1 and ν2. The power of the component is obtained.
In addition, the control unit 9 uses the phase of the electrical signal having the transmission frequency width ν1 and ν2 of the transmission frequency from each of the bandpass filters 82 and 80 ′ to obtain the phase at the center frequency of the transmission signal light. To do. Here, each of the frequencies ν1 and ν2 is set as an integer multiple of the reciprocal of the frame period Tf. In the present embodiment, for example, the frequency ν1 is set to one time the inverse of the frame period Tf, and the frequency ν2 is set to twice.

B. Principle of improving reception sensitivity by suppressing phase fluctuation Next, the principle of improving reception sensitivity by suppressing phase fluctuation will be described with reference to FIGS. FIG. 7 is a diagram showing the phase spectrum of the signal light transmitted from the optical transmitter 2. FIG. 8 is a diagram showing the phase spectrum of the signal light after delayed interference output from the branching unit 7 or 7 ′ to the parallel bandpass filters 8, 8 ′ and the external data processing circuit.
7 and 8, the horizontal axis represents the frequency shift amount, the display range thereof is the same as in FIGS. 3 to 6, and the vertical axis represents the phase. Based on the correspondence relationship between the phase and the frequency shift amount in FIG. 7, the phase at the center frequency of the transmitted signal light is obtained. Further, the phase at the center frequency of the signal light subjected to the delay interference in the bit delay is obtained from the correspondence relationship between the phase and the shift frequency in FIG. Therefore, in each of FIGS. 7 and 8, the phase of each of the comb component of the frequency ν1 which is the shift frequency adjacent to the center frequency and the comb component of the frequency ν2 adjacent to the frequency ν1 is Fourier-transformed. The phase at the origin is derived by extrapolating each phase.

That is, in the two-dimensional coordinate system based on the frequency shift amount and the phase shown in FIGS. 7 and 8, the coordinate values of the frequency ν1 and the phase at the frequency ν1 and the frequency ν2 and the phase at the frequency ν2 are connected. By generating a straight line and detecting the phase of the frequency ν0 on the straight line, the phase of the frequency ν0 is estimated.
Then, the difference between the phase of the center frequency obtained in FIG. 8 and the phase of the center frequency obtained in FIG. 7 is obtained, and the phase of the center frequency of the signal light is emitted from the optical transmitter 2 by the difference amount and the positive / negative polarity. It is possible to determine how much the delay amount is deviated from the point in time, or whether the phase difference is + or −.

Here, the phase of the comb component of each of the frequencies ν1 and ν2 for adopting the linear extrapolation of the I component is the phase of the electric signal transmitted through the bandpass filters 81 and 82 having the transmission frequencies ν1 and ν2. It is obtained by measuring. Similarly, the phase of the comb component of each of the frequencies ν1 and ν2 for adopting the linear extrapolation of the Q component is the electric signal transmitted through the bandpass filters 81 ′ and 82 ′ having the transmission frequencies ν1 and ν2. It is obtained by measuring the phase. The power measurement of the above-described comb component of the center frequency and the comb component of the adjacent frequency may be performed using any one of the parallel bandpass filters 8 and 8 ′, and both the parallel bandpass filters 8 and 8 ′. You may go in. When both are performed, the influence of noise superimposed on the signal light can be further reduced by obtaining an average value of the power of each comb component measured by each of the parallel bandpass filters 8 and 8 ′. .
In order to improve the phase derivation accuracy at the center frequency of the signal light, the number of bandpass filters constituting each of the parallel bandpass filters 8 and 8 ′ is increased to four or more, and comb components at different frequencies are further increased. The phase may be measured and information detected from each bandpass filter may be used.

In addition, when the number of band-pass filters constituting each of the parallel band-pass filters 8 and 8 ′ is four or more, an extrapolation method for approximating a nonlinear polynomial having higher accuracy than linear extrapolation is adopted. Is possible.
As a result, the phase at the center frequency of the signal light can be kept constant, noise due to phase fluctuations at the center frequency of the signal light can be removed, and reception sensitivity can be improved. In addition, transmission by DQPSK involves two orthogonal components consisting of cos (in-phase, I) and sin (quadture, Q). Here, since each of the cos component and the sin component can be synthesized into a single sine wave component by the addition theorem of the trigonometric function, the above description of the DPSK transmission also holds true for the DQPSK transmission.

Moreover, although the optical fiber transmission line 3 is a transmission line comprised using the optical fiber, the relay apparatus is provided in the middle and may be connected with another optical fiber transmission line. Due to the distribution and exchange of the signal light between the optical transmission lines, a waiting time occurs in the transmission of the signal light, and the time interval between frames in the transmitted signal light may not be constant.
In addition, if the transmission traffic for transmitting the frames is sparse, a waiting time is generated between the frames constituting the data sequence 1 while the frames are sequentially transmitted. For this reason, as in the case where the distribution and exchange described above occur, the time interval between frames of the transmitted signal light is not constant. Therefore, in these states, the frequency interval of the line spectrum as the comb component is not constant in time.

It is also conceivable that optical characteristics such as chromatic dispersion of the optical fiber constituting the optical fiber transmission line 3 fluctuate with time due to a change in environmental conditions, and the frequency interval of the comb component is not constant.
Therefore, in order to keep the frequency interval of the comb component constant in time, after being converted into an electric signal in the optical receiving unit 4, the branching unit 7 adjusts the time difference between the frames to thereby set the time interval between the frames. Keep it constant.
On the other hand, in the case of an optical transmission system in which the time interval between frames is always kept constant in the optical fiber transmission line 3, the branching unit 7 does not need to have a function of adjusting the time interval between frames.

In the principle already explained, since the time gap between frames is zero, the time interval between frames is equal to the time period Tf of one frame.
However, the time interval between frames does not have to be zero. For example, a gap may be set for a certain bit. In this case, the frequency interval of the comb component is narrowed by the number of bits included in the gap. Here, the time gap between frames may be determined in consideration of the ease of configuration of the optical transmission system and the cost. Further, if the time gap is constant, it need not be an integer multiple of 1 bit, and may be a fraction.

<First Embodiment>
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 used for explaining the principle is used as a drawing. Hereinafter, only differences in configuration and operation of the first embodiment from the description of the principle will be described.
In FIG. 1, considering application to OTN, one frame of data sequence 1 includes 4 rows of 4080 bytes and includes the total number of bits N = 4080 × 8 × 4 = 130560. For example, when the bit rate 43.018413559 Gbps (Bits Per Second) of the payload capacity of OTU3 is applied, the bit period Tb of 1 bit is about 23.25 ps, and the frame period Tf of 1 frame is about 3.035 μs.

  In the present embodiment, the time gap between adjacent frames is set to zero, and the frequency interval of the comb is about 329.5 MHz. The center value of the transmission frequency of each of the bandpass filters 80, 81, 82 in each of the parallel bandpass filters 8, 8 ′ and each of the bandpass filters 80 ′, 81 ′, 82 ′ is ν 0 = 0 Hz (DC Component), .nu.1 = 329.5 kHz, .nu.2 = 659.0 kHz. Here, the width of the transmission frequency of each of the bandpass filters 81, 82, 81 ′, and 82 ′ is half of the frequency interval of the comb component (the generation period of the line spectrum that is the comb component, hereinafter referred to as the comb frequency interval) Using 2) as a guide, for example, it is set as half the comb frequency interval.

The DC component in the signal light after the interference corresponds to the comb component at the origin. Accordingly, since each of the bandpass filters 80 and 80 ′ transmits a DC component and there is no negative frequency component, for example, the comb frequency is set to 1/4 (1/4) of the comb frequency interval. Set as one quarter of the interval.
As a result, each of the bandpass filters 80, 81, and 82 has a transmission frequency width corresponding to the center values ν1, ν2, and ν3 of the transmission frequency set to 80 kHz, 160 kHz, and 160 kHz. Similarly, in each of the bandpass filters 80 ', 81' and 82 ', the transmission frequency widths corresponding to the center values ν1, ν2 and ν3 of the transmission frequency are set to 80 kHz, 160 kHz and 160 kHz, respectively. .

In addition, the numerical values of the transmission frequencies of the bandpass filters 80, 81, and 82 are examples in the present embodiment, and are not limited to these, so that the accuracy of identifying the line spectrum as a comb component is improved. Adjust and set timely according to the bit frequency and frame frequency.
With the above-described configuration, in this embodiment, each of the bandpass filters 80, 81 and 82 and the bandpass filters 80 ′, 81 ′ and 82 ′ constituting the parallel bandpass filters 8 and 8 ′ has a transmission frequency region of MHz. (Megahertz) or less, there is no need for high-speed signal processing, and the electronic circuit performs high-speed operation and high-accuracy operation, so that the circuit scale can be reduced and power consumption can be suppressed.

Next, FIG. 9 is a diagram illustrating a configuration example of the optical transmission unit 2 in FIG. In FIG. 9, the optical transmission unit 2 includes an optical modulation unit 10, a light source 11, an incident optical fiber 12, an outgoing optical fiber 13, an optical modulation driving unit 14, a connection optical fiber 15, an optical branching unit 16, a connection optical fiber 17, and reception. A control unit 18, an optical branching unit 19, and a connection optical fiber 20 are provided.
The optical modulator 10 is a phase modulator using a nested MZ waveguide configured by connecting a first Mach-Zehnder (MZ) waveguide and a second MZ waveguide in parallel as a propagation path of signal light that is a light wave. is there. The optical modulation unit 10 generates four-level (for example, π / 4, 3π / 4, 5 / 4π, and 7π / 4) phase modulation signals between 0 and 2π in radians. For example, in the optical modulation unit 10, the first MZ waveguide composed of the arm waveguides 10A and 10B generates the I component of the signal light, and the second MZ waveguide composed of the arm waveguides 10C and 10D generates the Q component of the signal light. To do.

The optical modulation unit 10 applies push-pull drive to each of the arm waveguides 10A and 10B in the first MZ waveguide by applying an electric control signal that is complementary to the phase modulation units 10A1 and 10B1 provided, that is, whose polarity is inverted. By doing so, I component signal light with suppressed frequency chirp is generated. Similarly, the light modulation unit 10 applies, to each of the arm waveguides 10A and 10B in the first MZ waveguide, an electrical control signal that is complementary to the phase modulation units 10C1 and 10D1 provided, that is, whose polarity is inverted, By performing push-pull driving, Q component signal light with suppressed frequency chirp is generated.
Each of these arm waveguides 10A, 10B, 10C, and 10D is configured using, for example, a silicon waveguide, a lithium niobate waveguide, an indium phosphide-based compound semiconductor waveguide, or the like.

  If frequency chirp can be suppressed, waveguides using other materials may be used for each of the arm waveguides 10A, 10B, 10C, and 10D. When the silicon waveguide is used, for example, the phase modulation unit (10A1, 10B1, 10C1, 10D1) can be configured using a waveguide having a PN junction. A forward bias or a reverse bias is applied to the PN junction, and the carrier density is changed to control the refractive index to perform phase modulation. When causing the phase modulation units (10A1, 10B1, 10C1, 10D1) to perform a high-speed phase modulation operation, it is desirable to apply a reverse bias to the PN junction.

The light source 11 uses a semiconductor laser that continuously oscillates in a single longitudinal mode, and has a wavelength locker for stabilizing the oscillation wavelength (or oscillation frequency). Here, in order to use the light source 11 in the C band and L band of optical fiber communication, a semiconductor laser having an indium phosphide-based compound semiconductor as an active layer can be used.
In the light source 11, a tunable semiconductor laser may be used in order to correspond to any one wavelength of the C band and the L band.

Continuous light emitted from the light source 11 is incident on the light modulator 10 via an incident optical fiber 12 using a polarization-maintaining single-mode optical fiber. Further, an optical branching portion 19 is inserted in the incident optical fiber 12. The light branching unit 19 branches continuous light emitted from the light source 11 to the connection optical fiber 20, for example, branches 10% of the light amount emitted from the light source 11 to the connection optical fiber 20. Exit. Here, the quantity of the branched light is not limited to 10%, but is determined according to the characteristics of the coherent optical transmission system to be constructed.
In addition, the light modulation unit 10 has an output end connected to one end of an output optical fiber 13 using a polarization-maintaining single-mode optical fiber. The other end of the outgoing optical fiber 13 is the outgoing end of the optical transmitter 2. The other end of the outgoing optical fiber 13 that is the outgoing end of the optical transmitter 2 is connected to one end of the connecting optical fiber 15. The other end of the connection optical fiber 15 is connected to the incident end of the optical branching section 16.

The light branching unit 16 has a first outgoing end connected to the incident end of the optical fiber transmission line 3 and a second outgoing end connected to one end of the connection optical fiber 17. As a result, the signal light emitted from the optical modulation unit 10 is guided to the optical fiber transmission line 3 via each of the outgoing optical fiber 13, the connection optical fiber 15, and the optical branching unit 16.
Here, the light modulation unit 10 emits signal light obtained by phase shift modulation of continuous light as outgoing light according to each of the data series 1 and 1 ′ to the incident end of the outgoing optical fiber 13.
The line width of the line spectrum of the comb component described later is determined by the line width of the oscillation spectrum of the continuous light of the semiconductor laser in the light source 11.

The optical modulation driving unit 14 sequentially inputs the bits in each frame of the data series 1 in time series, converts the bits of each frame into an output electrical signal corresponding to differential coding, and converts the output electrical signal to the optical modulation unit 10. Output for.
In addition, the light modulation unit 10 performs phase shift modulation on the continuous light according to the output electric signal from the light modulation driving unit 14 to obtain signal light. Here, in each of the arm waveguides 10A and 10B, depending on the configuration of the waveguide as the first MZ waveguide, either one of the phase modulation unit 10A1 of the arm waveguide 10A and the phase modulation unit 10B1 of the arm waveguide 10B. It is necessary to apply an electrical signal having the same phase as that of the bit in the frame of the data series 1 and to apply an electrical signal having an opposite phase to that of the bit in the frame of the data series 1.

Similarly, in each of the arm waveguides 10C and 10D, depending on the configuration of the waveguide as the second MZ waveguide, either one of the phase modulation unit 10C1 of the arm waveguide 10C and the phase modulation unit 10D1 of the arm waveguide 10D. It is necessary to apply an electrical signal having the same phase as the bit in the frame of the data sequence 1 ′, and to apply an electrical signal having an opposite phase to the bit in the frame of the data sequence 1 ′.
When performing the process of generating the signal light of the I component and the Q component, the light modulation driving unit 14 has two systems of electrical signals having the same phase and the opposite phase for each bit in each frame of the data series 1, 1 ′. For each of the first MZ waveguide and the second MZ waveguide is output to the light modulation unit 10 as the output electric signal described above.

One end of the connection optical fiber 17 is connected to the second emission end of the branching section 16, and the other end is connected to the incident end of the reception / control section 18.
The optical branching unit 16 branches the signal light emitted from the other end of the connection optical fiber 15 with respect to one end of the connection optical fiber 17, for example, the amount of signal light emitted from the other end of the connection optical fiber 15. 10% of the branch is emitted to the reception / control unit 18 through the connection optical fiber 17. Here, the quantity of the branched light is not limited to 10%, but is determined according to the characteristics of the coherent optical transmission system to be constructed.

Next, FIG. 10 is a diagram illustrating a configuration example of the reception / control unit 18 in FIG. 9. The reception / control unit 18 in FIG. 10 includes an optical reception unit 30, an amplification unit (limiter amplifier) 31, an AD conversion unit 32, a buffer unit 33, a parallel bandpass filter 34, and a control unit 35.
The optical receiver 30 has the first incident end connected to the other end of the connection optical fiber 17, the second incident end connected to the other end of the connection optical fiber 20, and the output terminal connected to the amplifier 31.
The light receiving unit 30 receives the signal light from the optical branching unit 16 guided through the connection optical fiber 17 and the light from the optical branching unit 19 guided through the connection optical fiber 20. The optical receiving unit 30 combines the signal light from the optical branching unit 16 and the light from the optical branching unit 19 into combined light, differentially detects a component due to interference of the combined light, and outputs a differential electric signal. Output.

Next, FIG. 11 is a diagram illustrating a configuration example of the optical receiving unit 30 in FIG. In FIG. 11, the optical receiving unit 30 includes an interference unit 40, incident optical fibers 41 and 42, and a differential receiving unit 43.
The interference unit 40 is a 2 × 2 multiplexing element including a 2 × 2 waveguide having a first incident end and a second incident end, and a first exit end and a second exit end.
The incident optical fiber 41 has one end connected to the other end of the connection optical fiber 17 and the other end connected to the first incident end of the interference unit 40.
The incident optical fiber 42 has one end connected to the other end of the connection optical fiber 20 and the other end connected to the second incident end of the interference unit 40.
The differential receiving unit 43 has a first incident end connected to the first outgoing end of the interference unit 40 and a second incident end connected to the second outgoing end of the interference unit 40.
Further, the differential receiver 43 differentially detects the interference light emitted from the first emission end of the interference unit 40 and the interference light emitted from the second emission end of the interference unit 40, and performs a differential electrical signal. Is output.

Returning to FIG. 10, the amplifying unit 31 includes a limiter amplifier, and amplifies input signals having different signal amplitudes to output signals having constant amplitudes and performs waveform shaping in order to avoid erroneous conversion of differential electrical signals. Has been introduced. Therefore, the amplification unit 31 may be omitted if only noise that can be ignored is superimposed. 10 and 11, some optical / electrical connection paths such as the optical fiber transmission path 3 connected to the optical transmitter 2 are omitted.
The AD conversion unit 32 converts the differential electrical signal output from the amplification unit 31 into a digital electrical signal by AD conversion processing, and outputs the digital electrical signal to the buffer unit 33 at the next stage.
The buffer unit 33 temporarily stores a plurality of, for example, M frames that are sequentially input, and outputs the frames to the parallel bandpass filter 34 as a sequence arranged at this time interval. In the following description, M = 16, but is not limited to this value, and may be set according to the measurement accuracy of the power and phase of each comb component.

  The parallel bandpass filter 34 has the same configuration as the parallel bandpass filters 8 and 8 'in FIG. That is, the parallel bandpass filter 34 includes a bandpass filter 341 having a transmission frequency ν0, a bandpass filter 342 having a transmission frequency ν1, and a bandpass filter 342 having a transmission frequency ν2. The parallel bandpass filter 34 is configured by arranging three bandpass filters 340, 341, and 342 having different transmission frequencies in parallel with respect to the input.

The parallel band-pass filter 34 outputs electric signals of comb components of the frequencies ν 0, ν 1 and ν 2 extracted by the band-pass filters 340, 341 and 342 to the control unit 35. Here, each of the bandpass filters 340, 341, and 342 has the same configuration as the bandpass filters 80, 81, and 82, respectively. Further, each of the bandpass filters 340, 341, and 342 has the same frequencies ν0, ν1, and ν2 that are transmission frequencies as the bandpass filters 80, 81, and 82, respectively.
Similar to the control unit 9 in FIG. 1, the control unit 35 obtains the power and phase of each comb component from the electric signals of the respective comb components of the input frequencies ν 0, ν 1 and ν 2.
Then, the control unit 35 calculates the power ratio of the comb component of the frequency ν0 to the comb component of the frequency ν1, and estimates the bit delay amount from this power ratio. Further, the control unit 35 stabilizes the 1-bit delay generated by the optical transmission unit 2 by outputting a control signal to the optical transmission unit 2 so that the power ratio is minimized.

The control unit 35 sends a control signal to the light modulation drive unit 14 of the light transmission unit 2. Here, similarly to the control unit 9 in FIG. 1, the control unit 35 preliminarily stores a delay time control table indicating the correspondence between the control signal output to the light modulation driving unit 14 and the power ratio, as necessary. Write and store in the internal storage.
Then, the control unit 35 searches the delay time control table based on the power ratio, selects a corresponding electric signal, and outputs it to the light modulation driving unit 14.
Thereby, the light modulation drive unit 14 outputs a control signal for controlling the delay time to the phase modulation units 10A1, 10B1, 10C1, and 10D1 provided in the arm waveguides 10A, 10B, 10C, and 10D, respectively. .

  Further, the control unit 35 extrapolates the phase of the comb component at the frequency ν 0, that is, at the center frequency, from the phase of the comb component at the frequency ν 1 and the phase of the comb component at the frequency ν 2, as in the control unit 9 in FIG. To derive. Similarly to the control unit 9 in FIG. 1, the control unit 35 includes, in each of the arm waveguides 10 </ b> A, 10 </ b> B, 10 </ b> C, and 10 </ b> D in the nested MZ waveguide, the phase of the comb component of the center frequency and the preset phase. A control signal to be output to the phase control unit 10CD1 provided in either the first MZ waveguide or the second MZ waveguide is output so that the phase difference of π / 2 becomes π / 2. Here, similarly to the control unit 9 in FIG. 1, as necessary, the control unit 35 outputs a control signal (voltage value or current for controlling the phase by changing the refractive index) to the light modulation driving unit 14. Value) and a phase time control table showing a correspondence between the phase difference and written in the internal storage unit in advance. Then, the control unit 35 reads the control signal corresponding to the phase difference from the phase time control table, and outputs the read control signal to the phase control unit, whereby the phase difference between the I component and the Q component is obtained. Is controlled to be π / 2 radians.

  In FIG. 9, the phase control unit 10CD1 is provided, for example, after the emission unit of the second MZ waveguide. The phase control unit 10CD1 may be arranged at any stage before the incident part of the first MZ waveguide, after the output part, before the incident part of the second MZ waveguide, or after the output part. For example, when the nested MZ waveguide is formed of, for example, a silicon waveguide, the phase control unit can be formed using a waveguide having a PN junction. In this silicon waveguide, a forward or reverse bias is applied to the PN junction to perform refractive index control, and phase control corresponding to a change in refractive index is performed.

The reception / control unit 18 calculates a delay time of 1-bit delay and a phase at the center frequency of the signal light for each of the two systems of the I component and the Q component, and each of the first MZ waveguide and the second MZ waveguide The structure which performs control of these may be sufficient. When the control of each of the first MZ waveguide and the second MZ waveguide is performed independently, the reception / control unit 18 branches the data processing circuit in the configuration from the optical reception unit 4 to the control unit 9 in FIG. The configuration is the same as the configuration in which the branch portions 7 and 7 ′ to be removed are removed.
Further, in the signal light incident from the optical receiving unit 30, the buffer unit 33 is omitted from the receiving / control unit 18 when it is guaranteed that the frames are arranged at regular intervals in time. be able to.

Further, as described above, the optical fiber transmission line 3 is a transmission line configured using, for example, a polarization-maintaining single mode optical fiber.
The optical fiber transmission line 3 may be provided with, for example, an optical amplifying unit, a clock regenerator, an optical switch, etc. as a relay device in the transmission line. In the optical fiber transmission line 3, the incident end of the light receiving unit 4 is connected to the emission end.

Next, FIG. 12 is a diagram illustrating a configuration example of the optical receiver 4 in FIG. In FIG. 12, the optical receiver 4 includes a delay interference unit 45, an incident optical fiber 46, and a differential receiver 47.
The optical receiver 4 causes the signal light incident from the exit end of the optical fiber transmission path 3 to enter the delay interference unit 45 via the incident optical fiber 46.
The delay interference unit 45 includes a first MZ waveguide and a second MZ waveguide therein for detecting the difference between the I component and the Q component of the signal light incident from the optical fiber transmission path 3, and includes a first asymmetric MZ waveguide. Delay interference signals are generated in the waveguide and the second asymmetric MZ waveguide, respectively.
The delay interference unit 45 includes a first asymmetric MZ waveguide and a first asymmetric 2MZ waveguide connected in parallel to the 1 × 2 waveguide. The incident end of the delay interference unit 45 is the incident end of the 1 × 2 waveguide. The 1 × 2 waveguide branches the signal light incident from the incident end into each of the first asymmetric MZ waveguide and the second asymmetric waveguide.

The first asymmetric waveguide has arm waveguides 45A and 45B, and the signal light incident from the first emission end of the 1 × 2 waveguide is converted into an arm waveguide that forms the asymmetric MZ waveguide. Branches to 45A and 45B. Since each of the arm waveguides 45A and 45B performs differential detection by interference using the signal light and the previous signal light, the signal light is branched into two and delayed to generate a 1-bit delayed interference signal. The configuration is an asymmetric MZ waveguide.
Similarly, the second asymmetric waveguide includes arm waveguides 45C and 45D, and the signal light incident from the second emission end of the 1 × 2 waveguide is guided by the arm guide that forms the asymmetric MZ waveguide. Branches to each of the waveguides 45C and 45D. Since each of the arm waveguides 45C and 45D performs differential detection by interference using the signal light and the previous signal light, the signal light is branched into two and delayed to generate a 1-bit delayed interference signal. The configuration is an asymmetric MZ waveguide.

As described above, the delay interference unit 45 includes the first asymmetric MZ waveguide including the arm waveguide 45A and the arm waveguide 45B, and the second asymmetric MZ waveguide including the arm waveguide 45C and the arm waveguide 45D. Contains. In order to generate a 1-bit delay in this first asymmetric MZ waveguide, asymmetry is introduced so that the difference in optical propagation time between the arm waveguides 45A and 45B is 1 bit. For example, in order to generate a 1-bit delayed interference signal obtained by delaying the signal light that interferes with the signal light by 1 bit, a delay time control unit 45B1 and a phase control unit 45B2 described later are provided for the arm waveguide 45B. Yes.
Similarly, in order to generate a 1-bit delay in the second asymmetric MZ waveguide, asymmetry is introduced so that the difference in optical propagation time between the arm waveguides 45C and 45D is 1 bit. For example, in order to generate a 1-bit delayed interference signal obtained by delaying the signal light that interferes with the signal light by 1 bit, a delay time control unit 45D1 and a phase control unit 45D2 described later are provided for the arm waveguide 45D. Yes.

  The difference in the light propagation time between the arm waveguides 45A and 45B or the difference in the light propagation time between the arm waveguides 45C and 45D varies depending on environmental conditions such as temperature, and also changes depending on the wavelength used. To do. For this reason, at a specific wavelength and a specific temperature used for optical transmission, the arm propagation 45A and 45B or between the arm waveguides 45C and 45D has an optical propagation time difference equal to 1-bit delay. A first asymmetric MZ waveguide composed of the waveguides 45A and 45B or a first asymmetric MZ waveguide composed of the arm waveguides 45C and 45D is designed. Each of the arm waveguides 45A, 45B, 45C, and 45D can be configured using a silicon waveguide, a lithium niobate waveguide, an indium phosphide-based compound semiconductor waveguide, or the like.

That is, the light receiving unit 4 is connected to the incident end of the incident end of the incident optical fiber 46 using a single mode optical fiber or a polarization-maintaining single mode optical fiber. The incident optical fiber 46 has an incident end connected to an output end of the optical fiber transmission line 3.
Each of the first and second asymmetric MZ waveguides of the delay interference unit 45 is connected to the above-described arm waveguides 45A and 45B and arm waveguides 45C and 45D as two-branch waveguides on the output side. The light wave emitted from each of the exit ends of the arm waveguides 45A and 45B or the light wave emitted from the exit end of each of the arm waveguides 45A and 45B is designed to have a π phase difference.

  The differential receiver 47 takes the difference between the two I-component photodetectors and the signal light output from each of the two photodetectors, and outputs the first difference as a first differential electrical signal. And a dynamic amplification unit. The differential receiver 47 takes the difference between the two light detectors of the Q component and the signal light output from each of the two light detectors, and outputs the difference as a second differential electrical signal. 2 differential amplifiers.

  That is, in the two I-system photodetectors of the differential receiver 47, one photodetector receives the signal light from the arm waveguide 45A, and the other photodetector is 1 from the arm waveguide 45B. Receive bit delayed interference signal. The first differential amplifying unit of the differential receiver 47 is configured such that the difference between the electrical signal based on the signal light output from one of the photodetectors and the electrical signal based on the 1-bit delayed interference signal output from the other photodetector. The difference is output to the amplifying unit 5 as a first differential electric signal for the I component.

  Similarly, in the two-system photodetectors of the Q component of the differential receiver 47, one photodetector receives the signal light from the arm waveguide 45C, and the other photodetector is from the arm waveguide 45D. A 1-bit delayed interference signal is received. The second differential amplifying unit of the differential receiver 47 is configured such that the difference between the electrical signal based on the signal light output from one of the photodetectors and the electrical signal based on the 1-bit delayed interference signal output from the other photodetector. The difference is output to the amplifying unit 5 ′ as a second differential electric signal for the Q component.

In the asymmetric MZ waveguide, the delay time τ is set to one or both of the arm waveguides 45A and 45B in order to control the difference in the optical propagation time between the arm waveguides 45A and 45B to be always equal to 1 bit. A delay time control unit 45B1 for controlling and a phase control unit 45B2 for controlling the phase φ are provided. In FIG. 12, as already described, the delay time control unit 45B1 and the phase control unit 45B2 are interposed in the arm waveguide 45B, and the delay time control unit 45D1 and the phase control unit 45D2 are interposed in the arm waveguide 45D. It is provided.
Here, the control unit 9 sends control signals for controlling the delay time control units 45B1 and 45D1 and the phase control units 45B2 and 45D1 to the delay interference unit 45 through a control electric signal system schematically indicated by arrows. Output.

As shown in FIG. 12, each of the delay time control units 45B1 and 45D1 and the phase control units 45B2 and 45D2 is configured to apply a signal electrically independently from an independent output port.
Each of the delay time controllers 45B1 and 45D1 has a function of controlling the refractive index of the arm waveguides 45B and 45D. For example, when an arm waveguide (45A, 45B, 45C, 45D) is configured by a silicon waveguide, the silicon waveguide is a waveguide having a PN junction, and a forward or reverse bias is applied to the PN junction. Then, the refractive index control corresponding to the control electric signal is performed, and the control for holding the 1-bit delay is performed. This PN junction becomes the delay time control unit 45B1 or 45D1.

On the other hand, the phase controller 45B2 performs refractive index control corresponding to the control electric signal so that the phase difference between the output lights to each of the two systems of photodetectors corresponding to the I component becomes π / 2. Similarly, the phase controller 45D2 outputs the phase difference between the output lights to each of the two systems of photodetectors corresponding to the Q component, that is, the signal light emitted from the arm waveguide 45C and the arm light 45D. The refractive index control corresponding to the control electric signal is performed so that the phase difference from the 1-bit delayed interference signal is π / 2.
Each of the phase control units 45B2 and 45D2 can use the same configuration as the delay time control unit 45B1 or 45D1, and may be set to a waveguide length suitable for phase control.

Further, when a lithium niobate waveguide is used for each of the arm waveguides 45A, 45B, 45C, and 45D, it is possible to perform 1-bit delay and phase control using a change in birefringence due to the Pockels effect. It is.
In addition, when an indium phosphide compound semiconductor waveguide is used for each of the arm waveguides 45A, 45B, 45C, and 45D, the refractive index can be controlled by the Stark effect or the Pockels effect, and the signal light is delayed by 1 bit. And phase control.

  Each of the delay time control unit 45B1 and the phase control unit 45B2 may be installed on any one of the arm waveguides 45A and 45B in the first asymmetric MZ waveguide as shown in FIG. Similarly, each of the delay time control unit 45D1 and the phase control unit 45D2 may be installed on any one of the arm waveguides 45C and 45D in the second asymmetric MZ waveguide as shown in FIG.

  However, the voltage required for controlling each of the delay time control units 45B1 and 45D1, and the phase control units 15B2 and 45D2, or the power of the light emitted from the output ends of the arm waveguides 45A, 45B, 45C, and 45D, respectively. It is also necessary to perform control such as equalization. In this case, a delay time control unit 45B1 and a phase control unit 45B2 are provided for both arm waveguides 45A and 45B, and a delay time control unit 45C1 and a phase control unit 45D2 are provided for both arm waveguides 45C and 45D. You may comprise so that it may provide.

Each of the AD conversion units 6 and 6 ′ is input through the amplification units 5 and 5 ′, respectively, and the first differential electrical signal of I component output from the differential reception unit 77 and the second difference of the Q component. Each of the electrical signals is converted into a digital electrical signal. At this time, there is a possibility that the AD conversion units 6 and 6 ′ may perform erroneous conversion when converting the digital electric signal due to noise superimposed on the first differential electric signal and the second differential electric signal, respectively.
For this reason, each of the amplifying units 5 and 5 ′ is composed of a limiter amplifier. In order to avoid erroneous conversion of the first differential electric signal and the second differential electric signal, input signals having different signal amplitudes are used. It is introduced to amplify the output signal with a constant amplitude and perform waveform shaping. Therefore, if only noise that can be ignored is superimposed, the amplifying units 5 and 5 ′ may be omitted.

In FIG. 12, only a part of the connection path for the amplification units 5, 5 ′ and the control unit 9 is displayed. That is, each of the AD conversion units 6 and 6 ′ converts the first differential electric signal and the second electric signal to generate the first digital electric signal and the second digital electric signal, respectively.
Then, each of the AD converters 6 and 6 ′ outputs the first digital electric signal and the second digital electric signal to the branching units 7 and 7 ′, respectively.
Returning to FIG. 1, the branching unit 7 branches the input first digital electric signal of the I component into two paths, outputs the first digital electric signal of one branched path to the external data processing circuit, and The first digital electric signal of the branched path is output to the parallel band pass filter 8.
The branching unit 7 ′ branches the input second digital electric signal of the Q component into two paths, outputs the second digital electric signal of one branched path to an external data processing circuit, and the other branched path Are output to the parallel bandpass filter 8 '.
The data processing circuit performs a decoding process on the first and second digital electric signals and supplies the received data obtained to the receiver.

Further, when the branching unit 7 or 7 ′ has a function of making the time width of the time interval between adjacent frames constant, that is, having a time width of equal intervals, the following control is performed.
That is, the branching unit 7 receives the first digital electric signal of the other path after the branching, and a plurality of, for example, M frames of data set in advance in the internal buffer memory from the AD conversion unit 6. Store sequentially.
When the branching unit 7 detects that the M frames are stored, the branching unit 7 arranges the M frames at equal time intervals, for example, sets the time interval between the frames to zero, and sets the M frames. It outputs to the parallel band pass filter 8 as one series.

Similarly, the branching unit 7 ′ receives the second digital electric signal of the other path after branching, a plurality of, for example, M frames of data set in advance in the internal buffer memory, from the AD converting unit 6 ′. Store each time it is entered.
Then, when the branching unit 7 ′ detects that M frames are stored, the M frames are arranged at equal time intervals, for example, the time interval between the frames is set to zero, and the M frames are arranged. As a series, it is output to the parallel bandpass filter 8 '.

Here, the number of buffers included in each of the branching units 7 and 7 ′ has been described as M = 16. However, the number of buffers is not limited to this value, and may be set according to the measurement accuracy of the power and phase of each comb component. Good.
Further, in the case of a transmission system in which the signal light incident on the optical receiving unit 4 is guaranteed to be arranged at regular intervals in time, a buffer is provided in each of the branching units 7 and 7 ′. The function of providing a memory and making the time width of the time interval between adjacent frames constant can be omitted.

As described above, the parallel band-pass filter 8 extracts the frequency-shifted comb component electrical signal corresponding to each transmission frequency from each of the band-pass filters 80, 81, and 82, and the frequencies ν1, ν2, and An electrical signal having a frequency shift of ν 3 is output to the control unit 9.
Similarly, as described above, the parallel bandpass filter 8 ′ extracts the electric signal of the comb component of the frequency shift corresponding to each transmission frequency from each of the bandpass filters 80 ′, 81 ′ and 82 ′. , Output frequency-shifted electrical signals of frequencies ν1, ν2 and ν3 to the control unit 9, respectively.
The control unit 9 obtains the power and phase of each comb component for each I component and Q component from the electrical signals of frequency shifts of the frequencies ν1, ν2 and ν3 supplied from each of the parallel bandpass filters 8 and 8 ′. .

That is, the control unit 9 divides the power of the comb component of the frequency ν0 that is the center frequency by the power of the comb component of the adjacent frequency ν1 for each I component and Q component, and the power ratio between the frequency ν0 and the frequency ν1. Is calculated. Then, the control unit 9 calculates the bit delay amount of the signal light in the delay interference unit 45 based on the power ratio between the frequency ν0 and the frequency ν1.
Then, the control unit 9 uses the delay time control unit 45B1 or 45D1 for each of the I component and the Q component so that the power ratio between the frequency ν0 and the frequency ν1 is minimized. 1-bit delay is stabilized by controlling.

  Here, the control unit 9, for example, a delay time control table that indicates the correspondence between the power ratio and the control electrical signal that controls the delay amount corresponding to the power ratio, for each polarity indicated by the phase change amount of the frequency ν 1. Are previously written and stored in the internal storage unit corresponding to each of the I component and the Q component. The control unit 9 selects a power ratio corresponding to the polarity of the phase change of the frequency ν1 and a delay time control table of the control electric signal for each of the I component and the Q component. Then, the control unit 9 extracts a control electrical signal corresponding to the calculated power ratio from the selected delay time control table, and associates the extracted control electrical signal with each of the I component and the Q component to receive light. Output to unit 4. Here, the control electric signal for controlling the delay amount causes the delay time control units 45B1 and 45D1 to change the refractive indexes of the arm waveguides 45B and 45D, respectively, so that the current power ratio becomes a desired value. Voltage value or current value. The voltage value or the current value changes even if the absolute value of the bit delay amount is the same due to the polarity indicated by the phase change amount of the frequency ν1 described later.

  Further, the polarity indicated by the phase change amount of the frequency ν1 means that the phase of the electric signal component of the frequency ν1 output from each of the bandpass filters 81 and 81 ′ is, for example, 0.1 bit delayed by experiment or simulation in advance. To 1.9 bit delay including 1 bit delay. For each 1-bit delay, a phase in a delay range from a 0.1-bit delay to a 0.9-bit delay, a phase in a delay range from a 1.1-bit delay to a 1.9-bit delay, and 1 The difference from the phase of the bit delay is obtained in advance. Then, it is determined whether the phase difference in the delay of less than 1 bit and the phase difference in the delay of more than 1 bit are + and −, and a delay control table is created for each polarity.

  Then, the control unit 9 reads the phase at the time of 1-bit delay that is written and stored in the internal storage unit in advance. Further, the control unit 9 obtains the phase of the frequency ν1 in the digital electric signal supplied from each of the bandpass filters 81 and 81 'corresponding to M frames by Fourier transform. Further, the control unit 9 takes the difference between the read phase at the time of 1-bit delay and the phase obtained by Fourier transform, and extracts the delay control table corresponding to the polarity of this difference from the internal storage unit. Here, the description has been made with the frequency ν1, but the polarity may be obtained using the frequency ν2.

  Further, the control unit 9 obtains not only the frequency ν1 but also the phase of the frequency ν2 for each of the I component and the Q component by Fourier transform, and the phase of the frequency ν1 and the comb component of the frequency ν1, the frequency ν2 and the frequency ν2 Based on the phase of the comb component, a linear function in the two-dimensional coordinate system of the frequency shift amount and phase shown in FIG. 7 and FIG. 8 is derived by extrapolation processing. And the control part 9 calculates | requires the phase of a frequency shift to zero, ie, a center frequency, by this calculated | required linear function for every I component and Q component.

  Here, the correspondence between the center frequency and the phase of the signal light emitted from the optical transmission unit 2 is calculated and written and stored in the internal storage unit of the control unit 9. In the internal storage unit of the control unit 9, a phase control table is written and stored in advance for each of the I component and the Q component. This phase control table shows the correspondence between the phase difference and the control electric signal for controlling the phase difference. Here, the phase difference indicates a difference between the phase of the center frequency of the signal light emitted from the optical transmission unit 2 and the phase of the center frequency obtained by the control unit 9 from a linear function based on the frequencies ν1 and ν2. A correspondence relationship between a phase difference and a control electric signal in which this phase difference is 0 is set as a phase control table in advance through experiments or simulations. Here, the control electric signal for controlling the phase is a voltage value or current that causes the phase control units 45B2 and 45D2 to change the refractive indexes of the arm waveguides 45B and 45D, respectively, so that the current phase difference is zero. Value.

Next, the control unit 9 inputs digital electric signals of frequencies ν1 and ν2 from the bandpass filters 81 and 82 and from the bandpass filters 81 ′ and 82 ′, respectively. Find the phases of frequencies ν1 and ν2. Further, the control unit 9 reads the phase at the center frequency of the signal light emitted from the optical transmission unit 2 for each of the I component and the Q component from the internal storage unit, and calculates the phase obtained from the digital electrical signal from the read phase. Subtract to find the phase difference.
Then, the control unit 9 reads out the voltage value or current value as the control electric signal corresponding to the obtained phase difference for each of the I component and the Q component from the phase control table, and sends the read control electric signal to the optical receiving unit 4. Output.
Therefore, the control unit 9 has a non-volatile memory as a digital circuit including an arithmetic circuit for calculating a power ratio and a phase, an internal storage unit for storing the delay control table and the phase control table as the calibration data, and a delay. The time control units 45B1 and 45D1 and the phase control units 45B2 and 45D2 are each composed of a signal source for generating a voltage or current to be a control electric signal.

In the present embodiment, with the configuration described above, for each of the I component and the Q component, the 1-bit delay is always stabilized by the bit delay control, and the differential detection penalty by the control of the phase of the signal light (phase fluctuation control). Can be simultaneously performed using the line spectrum of the comb component, and it is possible to realize a coherent optical transmission system that has excellent reception accuracy and high reception sensitivity.
Based on the configuration of the present embodiment, the bit period or the number of bits of one frame is changed, that is, the frame period is changed, so that not only OTN but also coherent compatible with other optical transmission systems such as Ethernet (registered trademark). An optical transmission system can be constructed.
Further, in the present embodiment, the bit delay amount and the phase at the center frequency obtained by the control unit 9 are fed back to the optical transmission unit 2 to be used for stabilizing the optical transmission unit 2. Can do.
Moreover, in this embodiment, you may feed back with respect to the optical transmission part which belongs to another coherent optical transmission network connected to the optical fiber transmission line 3. FIG.

<Second Embodiment>
The second embodiment of the present invention will be described below with reference to the drawings. The configuration of the second embodiment is the same as that of the first embodiment, and will be described with reference to FIG. Only the configuration and operation of the second embodiment different from those of the first embodiment will be described below.
The polarization of the signal light emitted from the optical transmission unit 2 in FIG. 1 is constant, but the polarization of the signal light incident on the optical reception unit 4 is in the process of propagating through the optical fiber transmission line 3. There is a possibility that it will fluctuate with fluctuations in the polarization characteristics of the optical fiber transmission line 3.

  For this reason, when the refractive index of each of the arm waveguides 45A, 45B45C and 45D constituting the delay interference unit 45 in the optical receiving unit 4 depends on the polarization, it depends on the polarization of the signal light incident on the optical receiving unit 4 The bit delay time of each of the arm waveguides 45A and 45B (bifurcated waveguide) of the first asymmetric MZ waveguide, and each of the arm waveguides 45C and 45D (bifurcated waveguide) of the second asymmetric MZ waveguide The bit delay time varies. Similarly, the phase difference between the signal lights propagating through each of the arm waveguides 45A and 45B of the first asymmetric MZ waveguide according to the polarization of the signal light incident on the optical receiver 4 (the level of the signal light). The phase difference between the signal lights propagating through the arm waveguides 45C and 45D of the second asymmetric MZ waveguide varies.

In order to eliminate the above-described bit delay and fluctuation in phase difference, one solution is to keep the polarization of the signal light incident on the optical receiver 4 constant.
Next, FIG. 13 is a diagram for explaining the polarization controller 50 that keeps the polarization of the signal light incident on the optical receiver 4 from the optical fiber transmission line 3 constant. In FIG. 13, only the incident side of the light receiving unit 4 is extracted from the emission end of the optical fiber transmission line 3 and illustrated. As shown in FIG. 1, the output terminal of the optical receiver 4 is connected to the input terminals of the amplifiers 5 and 5 ′.
The exit end of the optical fiber transmission line 3 is connected to the entrance end of the polarization controller 50. The polarization controller 50 includes, for example, a polarization conversion element and a polarization monitor using a quarter (¼) wavelength plate and a half (½) wavelength plate, and an incident signal. It has a function to keep the polarization of light constant at all times.

In this embodiment, each of the arm waveguides 45A, 45B, 45C, and 45D in the delay interference unit 45 is configured to be optimized for TE (Transverse Electric wave) polarization.
In this case, the polarization control unit 50 controls the polarization of the signal light incident from the optical fiber transmission line 3 so that the polarization of the outgoing signal light is held at a preset TE polarization.
The polarization control unit 50 has one end of a connection optical fiber 51 that uses a polarization-maintaining single mode optical fiber connected to the output end.
The other end of the connection optical fiber 51 is connected to the incident end of the optical receiver 4. Here, a polarization-maintaining single-mode optical fiber is used as the incident optical fiber 46 (FIG. 12) that connects the incident end of the optical receiving unit 4 and the incident end of the delay interference unit 45.

With the configuration described above, according to the present embodiment, the polarization state of the signal light incident on the optical receiver 4 is kept constant, the polarization characteristics of the optical fiber transmission line 3 fluctuate, and the polarization of the signal light Even when the signal changes, it is possible to configure a coherent optical transmission system that is stabilized while maintaining a 1-bit delay and improved in receiving sensitivity by suppressing phase fluctuation of signal light.
In the present embodiment, the configuration for optimizing the TE polarization of the delay interference unit 45 is adopted. However, the configuration is not limited to this TE polarization, and the polarization direction is orthogonal to the TE polarization. It is good also as a structure optimized with respect to TM (Transverse Magnetic wave) polarization to be performed or another polarization.

<Third Embodiment>
The third embodiment of the present invention will be described below with reference to the drawings. The configuration of the third embodiment is the same as that of the first embodiment, and will be described with reference to FIG. Only the configuration and operation of the third embodiment different from those of the first embodiment will be described below.
The polarization of the signal light emitted from the optical transmission unit 2 in FIG. 1 is constant, but the polarization of the signal light incident on the optical reception unit 4 is in the process of propagating through the optical fiber transmission line 3. There is a possibility that it will fluctuate with fluctuations in the polarization characteristics of the optical fiber transmission line 3.

  For this reason, when the refractive index of each of the arm waveguides 45A, 45B, 45C and 45D constituting the delay interference unit 45 in the optical receiver 4 depends on the polarization, the polarization of the signal light incident on the optical receiver 4 In accordance with the bit delay time of each of the arm waveguides 45A and 45B (bifurcated waveguide) of the first asymmetric MZ waveguide, and the arm waveguides 45C and 45D (bifurcated waveguide) of the second asymmetric MZ waveguide. Each bit delay time varies. Similarly, the phase difference between the signal lights propagating through each of the arm waveguides 45A and 45B of the first asymmetric MZ waveguide according to the polarization of the signal light incident on the optical receiver 4 (the level of the signal light). The phase difference between the signal lights propagating through the arm waveguides 45C and 45D of the second asymmetric MZ waveguide varies.

In the configuration described above, by configuring an independent reception system for two orthogonal polarizations (for example, TE polarization and TM polarization), it is stabilized while maintaining a 1-bit delay, and phase fluctuation suppression Thus, a coherent optical transmission system with improved reception sensitivity can be configured.
Next, FIG. 14 is a diagram showing a configuration of a coherent optical transmission system having an independent reception system for two orthogonal polarizations. In FIG. 14, the optical transmission unit 2, the AD conversion unit 6, the branching units 7 and 7 ′, the parallel bandpass filters 8 and 8 ′, and the control unit 9 are omitted, but the TE polarized wave is illustrated. As an independent receiving system corresponding to each of the polarizations of TM and TM, the output stages of the I-component amplification units 5_1 and 5′_1 and the Q-component amplification units 5_2 and 5′_2 are as follows: Branch units 7 and 7 ′, parallel band pass filters 8 and 8 ′, and control units 9 and 9 ′ (not shown) are provided as in FIG. Here, each of the amplification units 5_1, 5′_1, 5_2, and 5′_2 has the same configuration as that of the amplification unit 5.

Therefore, for example, the digital electrical signals output from each of the amplification units 5_1, 5′_1, 5_2, and 5′_2 are respectively connected to the AD conversion units 6 and 6 ′, the branching units 7 and 7 ′, and the parallel bandpass. The signals are sequentially transmitted to the filters 8, 8 ′ and the control unit 9, and the control unit 9 controls the optical receiving unit 4 by outputting a control electrical signal based on the digital electrical signals of the I component and the Q component.
On the other hand, the digital electric signal output from the amplification unit 5 ′ is sequentially transmitted to the AD conversion unit 6 ′, the branching unit 7 ′, the parallel bandpass filter 8 ′, and the control unit 9 ′, and the control unit 9 ′ is transmitted to the optical receiving unit. 4 'is controlled by outputting a control electric signal based on the digital electric signal.
Here, each of the optical receiving unit 4 ′, the amplifying unit 5 ′, the AD converting unit 6 ′, the branching unit 7 ′, the parallel bandpass filter 8 ′, and the control unit 9 ′ includes the optical receiving unit 4, the amplifying unit 5, The configuration is the same as that of the AD conversion unit 6, branching unit 7, parallel bandpass filter 8, and control unit 9.
In addition, a selection circuit is provided for the output terminals to the data processing circuit of the branching units 7 and 7 ′, and it is selected which output of the branching unit 7 or 7 ′ is supplied to the data processing circuit. When selecting the output, the selection circuit compares the strengths of the digital electric signals of the amplifying units 5_1 and 5′_1 and the amplifying units 5_2 and 5′_2, and outputs the output having the higher strength to the data The processing circuit selects a process.

Further, in the present embodiment, as in the second embodiment, each of the optical receiving units 4 and 4 ′ is configured to be optimized with respect to the TE polarization of the delay interference unit 45. This optical receiver 4 ′ has the same configuration as the optical receiver 4.
In FIG. 14, the polarization separation unit 60 is connected at the incident end to the output end of the optical fiber transmission line 3. The polarization separation unit 60 includes a first emission end that emits a linearly polarized light wave parallel to the TE polarization corresponding to the delay interference unit 45, and a linear polarization parallel to the TM polarization. And a second emission end for emitting the light wave. For example, a polarization beam splitter using a polarization prism that separates light waves according to the polarization direction can be used as the polarization separation unit 60.

In addition, the polarization separation unit 60 has a first emission end connected to the incident end of the optical reception unit 4 via the connection optical fiber 21. One end of the connection optical fiber 61 is connected to the first emission end of the polarization separation unit 60, and the other end is connected to the incident end of the optical reception unit 4. Further, the polarization separation unit 60 has a second output end connected to the incident end of the polarization rotation unit 63 via a connection optical fiber 62. One end of the connection optical fiber 62 is connected to the second emission end of the polarization separation unit 60, and the other end is connected to the incident end of the polarization rotation unit 23.
Each of the connecting optical fibers 61 and 62 uses a polarization-maintaining single mode optical fiber in order to propagate the light wave while maintaining the polarization.

The polarization rotation unit 63 rotates the polarization of the light wave from the polarization separation unit 60, that is, the TM polarization by 90 degrees, and outputs it as TE polarization from the output end. The polarization rotation unit 63 is configured by using, for example, a half (1/2) wavelength plate set so as to rotate the polarization direction by 90 degrees.
Further, the polarization rotator 63 is connected to the incident end of the optical receiver 4 ′ via the connection optical fiber 24 at the output end. The connection optical fiber 64 is a polarization-maintaining single mode optical fiber, one end of which is connected to the output end of the polarization rotating unit 63 and the other end is connected to the incident end of the optical receiving unit 4 ′. Yes.

With the configuration described above, according to the present embodiment, regardless of whether the polarization of the signal light emitted from the optical fiber transmission line 3 is TE polarization or TM polarization, depending on the polarization characteristics of the optical fiber, the optical receiver Each of 4 and 4 ′ can be made incident as TE-polarized signal light.
As a result, according to the present embodiment, the polarization state of the signal light incident on each of the optical receivers 4 and 4 ′ is kept constant, the polarization characteristics of the optical fiber transmission line 3 fluctuate, and the signal light Even when the polarization changes, a coherent optical transmission system that is stabilized while maintaining a 1-bit delay and improved in receiving sensitivity by suppressing phase fluctuation of signal light can be configured.
Further, in the present embodiment, a configuration that is optimized for the TE polarization of the delay interference unit 45 of each of the optical reception units 4 and 4 ′ is adopted, but is not limited to this TE polarization, It is good also as a structure optimized with respect to TM polarized wave or another polarized wave.
In addition, the configuration of the present embodiment is more complicated than that of the second embodiment, but it is not necessary to provide the polarization control unit 50 in the second embodiment, so that the configuration of the present embodiment is different from that of the first embodiment. Faster polarization control can be performed against polarization fluctuations.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes design and the like within a scope not departing from the gist of the present invention.

  DESCRIPTION OF SYMBOLS 1,1 '... Signal sequence 2 ... Optical transmission part 3 ... Optical fiber transmission line 4, 4', 30 ... Optical reception part 5, 5_1, 5_2, 5'_1, 5'_2, 31 ... Amplification part 6, 6 ' , 32 ... AD conversion unit 7, 7 '... Branching unit 8, 8', 34 ... Parallel bandpass filter control unit ... 9, 35 10 ... Optical modulation unit 10A, 10B, 10C, 10D, 45A, 45B, 45C, 45D ... arm waveguides 45B1 and 45D1 ... delay time controllers 45B2 and 45D2 ... phase controllers 10A1, 10B1, 10C1, and 10D1 ... phase modulators 11 ... light sources 12, 41, 42, 46 ... incident optical fibers 13 ... outgoing optical fibers 14 ... optical modulation drive unit 15, 17, 20, 51, 61, 62, 64 ... connection optical fiber 16, 19 ... optical branching unit 18 ... reception / control unit 30 ... light 33 ... buffer unit 40 ... Wataru portion 43, 47 ... differential receiver 45 ... delay interferometer 50 ... polarization controller 60 ... polarization splitter 63 ... polarization rotation unit 80, 81, 82 ... band-pass filter

Claims (11)

A coherent optical transmission system comprising two data series corresponding to each of an I component and a Q component encoded in a frame unit, wherein the frame is transmitted by transmission data composed of a plurality of bits,
An optical transmitter that modulates continuous light by differential coding corresponding to the transmission data to generate signal light, and sends the signal light to an optical fiber transmission line;
The signal light is received from the optical fiber transmission line, each of the I component and the Q component in the signal light is bifurcated into two propagation paths, and the other propagation that is delayed by one bit with the signal light of one propagation path An optical receiver that performs delayed interference with signal light on the path for each I component and Q component, restores the transmission data for each I component and Q component, and outputs each of the I component and Q component electrical signals;
Provided in correspondence with each of the I component and the Q component, and inputs a plurality of the frames in the electric signal supplied from the optical receiver as one series, and the zero frequency of the electric signal and an integer of the frame frequency A parallel bandpass filter composed of a plurality of bandpass filters provided for each transmission frequency, each of the double frequency as a transmission frequency,
A control unit for obtaining an intensity ratio between the intensity of the electric signal at the center frequency supplied from each of the parallel bandpass filters corresponding to each of the I component and the Q component and the intensity of the electric signal at the integral multiple of the transmission frequency; Prepared,
The control unit delays the signal light provided in the propagation path for performing 1-bit delay in the optical receiving unit so that the intensity ratio corresponding to each of the I component and the Q component is minimized. A coherent optical transmission system that controls a delay time control unit provided for each component. The optical receiver further comprises a phase controller;
The control unit detects the phase of the electric signal of any two transmission frequencies from the transmission frequencies supplied from the parallel bandpass filter for each electric signal corresponding to the I component and the Q component, The phase of the center frequency is obtained by extrapolating this phase, the phase difference between the obtained phase and a preset phase is obtained, and the phase difference corresponding to the I component and the Q component is minimized. The coherent optical transmission system according to claim 1, further comprising: controlling the phase control unit. The optical transmitter is also provided with other bandpass filters having the same configuration as each of the bandpass filter, the optical receiver, and the controller, another optical receiver, and another controller.
The other control unit is
The phase difference of each of the I component and the Q component in the signal light transmitted to the optical receiving unit by the electric signal of the center frequency supplied from the other band pass filter from the optical receiving unit is π The coherent optical transmission system according to claim 2, wherein control is performed to hold at / 2 radians. The delay time control unit and the phase control unit are provided for each of the I component and the Q component in the propagation channel and the other propagation channel provided corresponding to each of the I component and the Q component. 4. The coherent optical transmission system according to claim 2, wherein the coherent optical transmission system is provided in the other propagation path that performs bit delay.   The AD conversion part which digitizes the electric signal output from the said optical receiver is provided for every I component and Q component, The Claim 1 characterized by the above-mentioned. Coherent optical transmission system. The electric signal provided between the optical receiving unit and each of the AD component of the I component and the Q component, and amplifies the electric signal having a different signal amplitude to an output signal having a constant amplitude to perform waveform shaping The coherent optical transmission system according to claim 1, further comprising an amplifying unit that adjusts a voltage level of the coherent optical transmission system. Provided between the optical receiver and each of the parallel bandpass filters for the I component and the Q component, and the electrical signals corresponding to the I component and the Q component are respectively connected to the first path and an external circuit. A branch part that branches to the second route,
7. The electrical signal of the first path is output to the band-pass filter, and the electrical signal of the second path is output to an external circuit. 8. Coherent optical transmission system. For each of the branch portions of the I component and the Q component, a plurality of buffers for storing the electrical signal is provided for each frame in the sequence,
The branch is
The coherent optical transmission system according to claim 7, wherein the electrical signal of the frame is output to each of the parallel bandpass filters of the I component and the Q component at a constant period. The transmitter has a modulator comprising a waveguide in which a first Mach-Zehnder waveguide and a second Mach-Zehnder waveguide are arranged in parallel;
9. The transmitter according to claim 1, wherein the transmitter applies a modulation signal whose polarities are inverted to each other in each arm in the first Mach-Zehnder waveguide and the second Mach-Zehnder waveguide. A coherent optical transmission system according to claim 1. A polarization controller for making a polarization direction of signal light incident on the optical receiver constant between the outgoing end of the optical fiber transmission line and an incident end of the optical receiver. The coherent optical transmission system according to any one of claims 1 to 9. A polarization separation unit that separates the signal light emitted from the emission end into the first polarization direction and the second polarization direction orthogonal to the emission end of the optical fiber transmission line;
A polarization rotator that changes the signal light in the second polarization direction to the first polarization direction, and
The polarization separation unit is
10. The signal light in the first polarization direction is output to the optical receiving unit, while the signal light in the second polarization direction is output to the polarization rotation unit. The coherent optical transmission system according to one item.

Images