JP2012100128A - Demodulation method for coherent light time division multiplex signals - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve a demodulation method capable of detecting a multi-level signal at a high S/N ratio from an ultra high-speed coherent light time division multiplex signal obtained by modulating an amplitude and a phase of a light pulse simultaneously and performing acceleration by time division multiplex.SOLUTION: A receiving part has: a local oscillator; a light phase synchronization circuit synchronizing phases of a light pulse signal after transmission and local oscillation light with each other; a light demultiplexing circuit demultiplexing a high-speed light pulse into 1/N; and a homodyne/heterodyne detection circuit demodulating and detecting a coherent light signal. By using a CW light source as the local oscillator, a coherent light time division multiplex pulse signal is light-demultiplexed, and then, the light pulse is converted into CW light by using a dispersed medium and a light phase modulator. Further, coherent detection is performed on the local oscillation light by using the CW light. Thereby, a multi-level signal is demodulated at a high S/N ratio.

Description

  The present invention is for detecting a multi-level signal of high S / N from a received optical pulse in ultra-high speed coherent optical time division multiplexing transmission in which information is simultaneously put on the amplitude and phase of an optical pulse and high-speed transmission is performed by time division multiplexing. The present invention relates to a demodulation method.

  In recent optical communications, there is a great interest in a coherent multilevel transmission system that uses not only the amplitude but also the phase of light for information transmission. In this transmission method, since information of 2 bits or more can be transmitted with one symbol, high-speed transmission is possible at a low symbol rate. As a result, it is possible to realize characteristics such as expansion of frequency utilization efficiency, improvement of resistance to chromatic dispersion and polarization dispersion, and power saving.

  On the other hand, due to the processing limit of the digital signal processor used for modulation / demodulation of coherent optical signals, especially the speed limit of A / D and D / A conversion, the symbol rate of the coherent optical transmission system is currently limited to about 10 Gsymbol / s. It is. In order to overcome this, instead of the conventionally used coherent CW light, the amplitude and phase of a coherent RZ (Return-to-Zero) light pulse are subjected to multilevel modulation, and this is time-multiplexed in the optical domain. A method for realizing coherent light transmission exceeding the processing limit of an electronic device has been proposed (Non-Patent Documents 1 to 3).

An example of a schematic configuration of a conventionally used coherent optical time division multiplexing transmission system is shown in FIG. On the transmission side, the amplitude and phase of the coherent RZ optical pulse are subjected to multilevel modulation at the symbol rate R, and the symbol rate is increased N times by optical time division multiplexing. Here, if the multi-level degree of multi-level modulation is 2 ^M , the transmission rate is N × M × R. When the polarization multiplexing is combined, the transmission speed can be doubled.

  On the receiving side, coherent detection is performed on the coherent optical signal after transmission using a local signal. Coherent detection includes a method of performing homodyne or heterodyne detection of a local light and a transmission signal whose phases are synchronized, and a method using asynchronous local light. In the former method, as shown in FIG. 1, an optical phase synchronization circuit is used to synchronize the phase between the transmission signal and the local oscillation signal via a reference optical signal called a pilot tone. In the latter method, the phase error is estimated by digital signal processing after conversion into an electrical signal.

  In the coherent light time division multiplex transmission realized so far, an optical pulse train of repetition R is used as local light. The repetition of the local pulse light is synchronized with the clock signal extracted from the signal after transmission by the clock extraction circuit. By coherently detecting the local pulse light and the signal light pulse having the symbol rate N × R, not only the demodulation / detection of the coherent signal but also the demultiplexing from the symbol rate N × R to R can be realized simultaneously. . In direct detection or delay detection for conventional OOK (On-Off Keying), DPSK (Differential Phase Shift Keying), DQPSK (Differential Quadrature Phase Shift Keying) signals, etc., the symbol is preliminarily processed by an optical demultiplexing circuit before receiving the signal. While it is necessary to reduce the rate to 1 / N, this transmission has the advantage that the optical demultiplexing circuit is not required because the coherent detection circuit also has a demultiplexing function.

Japanese Patent Application No. 2010-38885

C. Zhang, Y. Mori, K. Igarashi, K. Katoh, and K. Kikuchi, “Demodulation of 1.28-Tbit / s polarization-multiplexed 16-QAM signals on a single carrier with digital coherent receiver,” in Proc. Optical Fiber Communication Conf. (OFC 2009), San Diego, CA, Mar. 2009, Paper OTuG3. C. Schmidt-Langhorst, R. Ludwig, D.-D. Grob, L. Molle, M. Seimetz, R. Freund, and C. Schubert, “Generation and coherent time-division demultiplexing of up to 5.1 Tb / s single -channel 8-PSK and 16-QAM signals, ”in Proc. Optical Fiber Communication Conf. (OFC 2009), San Diego, CA, Mar. 2009, Postdeadline paper PDPC6. K. Kasai, T. Omiya, P. Guan, M. Yoshida, T. Hirooka, and M. Nakazawa, “Single-Channel 400-Gb / s OTDM-32 RZ / QAM Coherent Transmission Over 225 km Using an Optical Phase- Locked Loop Technique, ”IEEE Photon. Technol. Lett., Vol. 22, no. 8, pp. 562-564, April 2010.

  However, as a current problem, the demodulation band of the data signal is limited to several tens of GHz due to the band of the electronic device in the demodulation circuit. As a result, the limitation of the demodulation band by the electronic device causes deterioration of S / N (signal power to noise power ratio) and waveform distortion, which makes it difficult to accurately demodulate highly coherent RZ optical signals. It was. In particular, the RZ pulse signal has a wide band and this deterioration is remarkable.

  Furthermore, there is an upper limit to the optical power that can be input to the heterodyne or homodyne detector. Therefore, if the power input to the coherent detection circuit is constant, the optical power per local light emission pulse decreases as the multiplicity N increases, and it becomes difficult to perform coherent detection at a high S / N. . As a result, there has been a problem that the increase in multiplicity is restricted and becomes an obstacle to speeding up.

  The present invention is to solve these problems, and it is an object of the present invention to newly provide a demodulation circuit capable of detecting an ultrafast coherent optical pulse signal with high S / N.

  In order to achieve such an object, the coherent time division multiplexed signal demodulation method according to the present invention newly includes an optical demultiplexing circuit for demultiplexing a high-speed optical pulse by 1 / N times in the conventional coherent detection system, and CW light is used as local light.

  Further, the optical pulse demultiplexed by the optical demultiplexing circuit is converted into CW light, and coherent detection of the CW signal light and CW local light is performed. This conversion from optical pulse to CW light can be realized by dispersing an optical pulse with a dispersive medium and then applying an appropriate inverse frequency chirp to the optical pulse by an optical phase modulator. At that time, by setting the chirp rate K of the optical phase modulator to satisfy the relationship of D = 1 / K with respect to the dispersion amount D of the dispersive medium, the pulse waveform is converted into a frequency spectrum by the inverse Fourier transform. I can do it. That is, it can be converted into a line spectral component. If this chirp rate K is sufficiently large, the spectrum width after conversion can be narrowed, that is, ideal conversion to CW becomes possible. As a result, demodulation with high S / N becomes possible.

  By performing coherent detection of the CW local light and the coherent optical signal converted into CW, it is possible to demodulate the multilevel signal with a higher S / N compared to the conventional method using the local pulse light. Become. As a result, the multiplicity N can be increased without sacrificing S / N, and the speed of coherent light time division multiplex transmission can be increased.

This is a general configuration of coherent optical time division multiplexing transmission. 1 is a configuration of a demodulation circuit for coherent optical time division multiplexing transmission according to an embodiment of the present invention. This is a configuration of a pulse → CW conversion circuit. IF band data signal spectrum, (a) shows the case where the signal light and local light are both pulsed, and (b) shows the case where the signal light is converted from pulse to CW and the local light uses CW. . This is a configuration of an optical phase modulator for applying a strong chirp used for pulse-to-CW conversion (1). This is a configuration of an optical phase modulator for applying strong chirp used for pulse-to-CW conversion (2). This is a configuration of a pulse → CW conversion experimental system. It is a pulse signal before conversion, (a) is an optical spectrum, and (b) is a time waveform. It is the pulse signal after conversion, (a) is an optical spectrum, (b) is a time waveform.

  An example of an embodiment of the present invention is shown in FIG. The coherent optical time division multiplexed pulse signal (symbol rate N × R) after transmission is first converted to a symbol rate of 1 / N times by the optical demultiplexing circuit 16 using a clock signal (frequency R) synchronized with the optical pulse signal. Demultiplexed. As the optical demultiplexing circuit, for example, an electro-absorption (EA) light modulator or the like can be used. In order to extract a clock signal having a frequency R from an optical signal having a symbol rate N × R, it is composed of a voltage-controlled oscillator (VCO) and a phase comparator for comparing the VCO frequency and signal timing. A PLL circuit or the like is used.

  The demultiplexed optical pulse is converted into CW light by using a pulse → CW conversion circuit 17. For the conversion from pulse to CW, the same clock signal as that used for the optical demultiplexing is used. As the pulse → CW conversion circuit, a combination of a dispersion medium and an optical phase modulator is used as will be described in detail later.

  In parallel, the reference optical signal (pilot tone) transmitted through the optical fiber transmission line simultaneously with the optical pulse signal is used to adjust the phase of the local CW light source via the optical phase synchronization circuit 9. Synchronize with phase. The optical phase synchronization circuit includes a phase comparator that compares the phase of the local light and the phase of the reference optical signal, and a negative feedback circuit that feeds back the error signal to the frequency variable mechanism of the local light source. The phase compensation may be performed by signal processing without using the phase synchronization.

  The coherent optical signal converted into CW light is input to the heterodyne or homodyne detection circuit 14 together with the local CW light phase-synchronized with the signal, and a multilevel signal converted into an intermediate frequency (IF) by coherent detection is obtained. Is output. For the coherent detection circuit, for example, a 90-degree optical hybrid circuit and a balanced photodetector are used. The IF signal is finally demodulated by the decoder 15. In the decoder, the IF signal is converted into a digital signal by an A / D converter, and demodulated by digital signal processing using an FPGA (Field Programmable Gate Array) or software.

Here, the configuration and operation principle of the pulse-to-CW conversion circuit used in this demodulation method will be described in detail. An example of the configuration of the pulse → CW conversion circuit is shown in FIG. The optical pulse is first given group velocity dispersion by the dispersive medium 17a, and then optical phase modulation (frequency chirp) is applied at the clock frequency R by the optical phase modulator 17b. As the dispersive medium, a single mode fiber, a diffraction grating, a fiber Bragg grating, or the like is used. In addition, LiNbO ₃ or an InP semiconductor modulator can be used as the optical phase modulator.

  Here, the IF band data spectrum (FIG. 4A) when the signal light is a pulse and the local light is also a pulse and the IF band data spectrum when the signal light is pulse-to-CW conversion and the local light is also a CW (FIG. 4B) is shown. Comparing (a) and (b), it can be seen that the signal component in the demodulation band is larger in (b) and demodulation with a high S / N is possible. In (a), since the signal light spreads to a high frequency, the S / N is bad.

の関係で結ばれている。分散性媒質の分散量をＤ＝β_２Ｌ（β_２：２次分散、Ｌ：長さ）とすると、分散性媒質を通過した光パルスのスペクトルおよび波形は、

と表される。この光パルスに対し光位相変調器によりφ＝Ｋｔ^２／２の位相変調を印加すると、時間領域では、

と書けることから、周波数領域では畳み込み積分を用いて

と表される。したがって式（２）および（５）より、

が得られる。ここでＤとＫがＤ＝１／Ｋの関係を満たすとき、式（６）は、

と書くことが出る。式（７）の積分は、−ω／Ｋをｔとみなすと、Ｕ_ｉｎ（ω）の逆フーリエ変換に対応している。すなわち式（７）は、

と表すことが出来る。式（８）は、出力される信号のスペクトルを見るとその形状は入力信号の波形に比例し、時間ｔがｔ＝−ω／Ｋの尺度で周波数ωに変換されていることがわかる。 The principle of operation of the pulse-to-CW conversion circuit can be explained as follows using mathematical formulas. The time waveform of an optical pulse input to this circuit is u _in (t), and its frequency spectrum is U _in (ω). u _in (t) and U _in (ω) are Fourier transforms

It is tied in a relationship. When the dispersion amount of the dispersive medium is D = β ₂ L (β ₂ : second-order dispersion, L: length), the spectrum and waveform of the light pulse that has passed through the dispersive medium are:

It is expressed. When the optical pulse with respect to applying a phase modulation of φ = Kt ^2/2 by the optical phase modulator, in the time domain,

In the frequency domain, using convolution integral

It is expressed. Therefore, from equations (2) and (5),

Is obtained. Here, when D and K satisfy the relationship of D = 1 / K, equation (6) becomes

You can write. The integration of equation (7) corresponds to the inverse Fourier transform of U _in (ω), assuming −ω / K as t. That is, equation (7) is

Can be expressed as When looking at the spectrum of the output signal, the shape of the equation (8) is proportional to the waveform of the input signal, and it can be seen that the time t is converted to the frequency ω on the scale of t = −ω / K.

を入力する場合を考える。パルス→ＣＷ変換回路の出力における信号のスペクトルは式（８）を用いて

と表される。このスペクトルの半値全幅Δνは

で与えられる。すなわち、Ｋを十分大きくとることによりスペクトル幅を狭窄化することが出来、これにより信号をＣＷ光に近づけることが可能となる。さらにその結果として、式（１０）からもわかる通り、中心周波数ω＝０におけるスペクトル強度｜Ｕ_ｏｕｔ（ω＝０）｜^２もＫ倍に増大し、局発ＣＷ光とのコヒーレント検波においてビート信号のＳ／ＮをＫ倍に増大できることがわかる。 As a specific example, Gaussian light pulse

Suppose you enter. The spectrum of the signal at the output of the pulse-to-CW conversion circuit is given by

It is expressed. The full width at half maximum Δν of this spectrum is

Given in. That is, by making K sufficiently large, the spectrum width can be narrowed, whereby the signal can be brought close to CW light. As a result, as can be seen from the equation (10), the spectral intensity | U _out (ω = 0) | ^{2 at} the center frequency ω = 0 also increases K times, and the beat signal is obtained in the coherent detection with the local CW light. It can be seen that the S / N can be increased K times.

For example, considering a Gaussian pulse (T ₀ = Δt / (2√ (ln2)) = 6 ps) with a pulse width Δτ = 10 ps, the spectrum width is Δν = 73.5 GHz. For example, when a chirp with a chirp rate of K = 2 ps ^-2 is applied using an optical phase modulator, the spectrum width is narrowed to 36.75 GHz. If it is difficult to realize such a strong chirp with a normal phase modulator configuration using one optical phase modulator in one direction, the folded configuration of the optical phase modulator as shown in FIG. A configuration in which a plurality of optical phase modulators having a folded configuration as shown in the figure are cascade-connected may be used (Patent Document 1).

  Next, an example of a pulse → CW conversion experiment is shown. FIG. 7 is a block diagram showing the configuration of the experimental system. The CW signal light output from the coherent CW light source 1 is shaped into a pulse signal by the optical phase modulator 17b driven by the clock frequency R and the dispersive medium 25. Next, this pulse signal has a dispersion characteristic whose sign is inverted from that of the dispersive medium 25. The CW signal is converted by the dispersive medium 17a and the optical phase modulator 17b driven at the clock frequency R. Here, a phase difference of π is added between clock signals for driving both optical phase modulators. 8 and 9 are the spectrum waveform and time waveform of the optical signal before and after conversion. It can be seen that the optical spectrum is narrowed by this conversion circuit, and the pulse waveform is converted to a CW signal.

  As described above in detail, in coherent optical time division multiplex pulse transmission, a coherent optical signal is optically demultiplexed and then converted to CW light, and further, coherent detection is performed using CW light as local light, It becomes possible to demodulate a multilevel signal with a high S / N compared to a system using local pulsed light. As a result, the multiplicity can be increased without sacrificing the S / N, so that ultrafast coherent optical time division multiplexing transmission can be realized.

DESCRIPTION OF SYMBOLS 1 Coherent CW light source 2 Optical pulse generation circuit 2a Optical comb generator 2b Optical filter 3 Modulator 4 Optical time division multiplexing circuit 5 Polarization multiplexing circuit 6 Pilot tone generation circuit 7 Optical fiber transmission line 8 Pilot tone extraction optical filter 9 Light Phase synchronization circuit 10 Polarization demultiplexing circuit 11 Clock extraction circuit 12 Optical pulse generation circuit 13 Locally generated CW light source 14 Heterodyne / homodyne detection circuit 15 Decoder 16 Optical demultiplexing circuit 17 Pulse → CW conversion circuit 17a Dispersive medium 17b Optical phase modulation 18 Optical circulator 19 Optical phase modulator 20 Phase shifter 21 Electric amplifier 22 Isolator 23 Optical delay circuit 24 Reflecting mirror 25 Dispersive medium

Claims (5)

  In a coherent optical time division multiplexing pulse transmission system that transmits an optical pulse whose amplitude and phase are multi-level modulated at an N-times symbol rate by optical time division multiplexing, the receiving unit is a local oscillator, an optical pulse signal after transmission, An optical phase synchronization circuit for synchronizing the phase with the local light, an optical demultiplexing circuit for demultiplexing a high-speed optical pulse by 1 / N times, and a homodyne or heterodyne detection circuit for demodulating and detecting a coherent optical signal, A coherent time-division multiplexed signal demodulation method using a CW light source as a local oscillator.   2. The method of demodulating a coherent optical time division multiplexed signal according to claim 1, wherein the phase synchronization of the optical pulse signal and the local light is performed by software instead of the optical phase synchronization circuit.   An optical pulse demultiplexed by an optical demultiplexing circuit is converted into CW (continuous) light, and a multilevel signal having a high S / N is demodulated by coherent detection of the CW signal light and CW local light. The demodulation method of the coherent optical time division multiplexed signal according to claim 1 or 2.   The optical pulse is converted into CW light by smoothing the optical pulse with a dispersive medium and then applying an appropriate frequency chirp to the optical pulse by an optical phase modulator. Demodulation method for division multiplexed signals.   5. The coherent optical time division multiplexed signal demodulation method according to claim 4, wherein the dispersion amount D of the dispersive medium and the chirp rate K of the optical phase modulator have a relationship of D = 1 / K.

Images