JP5713776B2 - Pre-equalized optical transmitter - Google Patents

Description

  The present invention relates to a pre-equalized optical transmitter for compensating for degradation of channel transmission characteristics, which is a cause of signal quality degradation in optical communication, by pre-equalization.

  Conventionally, as a compensation technique for transmission characteristic deterioration in a communication channel, transmission / reception equalization in the electric domain or optical domain has been mentioned. In the electric domain, transmission equalization (pre-equalization), reception equalization, etc. In the optical region, various compensation techniques such as dispersion compensating fibers have been studied.

Among them, the transmission equalization technique in the electric domain is attracting attention because it is a system that can obtain good characteristics without causing noise enhancement.
In particular, in the compensation using the dispersion compensating fiber, the installation cost and the installation location of the fiber become problems, but there is an advantage that the installation cost and the installation location of the dispersion compensation fiber can be reduced by using pre-equalization. is there.

  Also, there are NRZ (Non-Return to Zero) method and RZ (Return to Zero) method, but the RZ method is more resistant to intersymbol interference than NRZ, and has a maximum amplitude. When equal, the average power can be set small.

However, in the RZ system, since a transmission symbol always passes “0”, modulation in the time domain using a pulse carver is required after modulation to an NRZ optical transmission signal.
However, in pre-equalization transmission, signal modulation in the time domain is accompanied when generating an NRZ optical pre-equalization signal, so it is impossible to apply RZ transmission signal modulation using a pulse carver.

Therefore, in order to overcome the above problem, an RZ optical pre-equalization technique as shown in FIG. 7 has been proposed (for example, see Non-Patent Document 1).
In FIG. 7, a conventional pre-equalized optical transmitter includes a high-frequency circuit for a DSP (Digital Signal Processor) 31, D / A (Digital / Analog) converters 32 and 33, and analog signals after D / A conversion. The analog processing unit 34 for performing RZ modulation according to the above, drivers 35 and 36, a light source 37, and an optical modulator 38 are provided.

First, the DSP 31 performs pre-equalization processing and generates an NRZ electric pre-equalization signal.
The high frequency mixer in the analog processing unit 34 synthesizes both sidebands of the NRZ electrical pre-equalization signal to perform modulation in the frequency domain, and converts the NRZ electrical pre-equalization signal to the RZ electrical pre-equalization signal. Convert to
Finally, the optical modulator 38 realizes pre-equalization by the RZ method by modulating the RZ electrical pre-equalization signal into the RZ optical pre-equalization signal.

R. I. Killey, “Mitigation of Transmission Impulses in Long-Haul Submarine Links Using DSP-Based Electronic Predistortion”, IEEE LEOSpM 243-244.

  According to the technology described in Non-Patent Document 1, the conventional pre-equalized optical transmitter can transmit the RZ optical pre-equalized signal, but an additional circuit for analog processing is required to perform RZ modulation. The analog processing circuit requires a precise analog circuit composed of at least four high-frequency mixers, six high-frequency adders, and four high-frequency delay elements, which increases the circuit size, power consumption, and cost.

  The present invention has been made to solve the above-described problems, and by generating an RZ optical pre-equalization signal without using an additional circuit for analog processing, the circuit size, power consumption, and cost can be reduced. An object of the present invention is to obtain a pre-equalized optical transmitter capable of realizing reduction.

The pre-equalized optical transmitter according to the present invention includes an NRZ input signal sequence having one symbol length, a first sub-impulse response sampled alternately by over-sampling from an impulse response having a reverse characteristic of a signal quality degradation factor, A pre-equalization transmitter that receives two sub-impulse responses as inputs and compensates for deterioration factors of the signal quality by pre-equalization, the input NRZ input signal sequence and the input first sub-impulse response The first FIR filter for the Even component that generates and outputs the first pre-equalization signal of the Even component by the convolution operation with the first FIR filter, and the NRZ input signal that is input in parallel with the first FIR filter A second pre-equalization signal of the Odd component is generated by convolution calculation of the sequence and the input second sub-impulse response, and is output. Converting a second FIR filter for Odd component force, each of said input from the 1FIR filter first pre-equalized signal and said input from the 2FIR filter second pre-equalized signal into an analog signal The D / A converter to be output and the serial signal stored in one symbol length by alternately selecting the first pre-equalization signal and the second pre-equalization signal input from the D / A converter a selector for outputting a light source, based on the serial signal inputted from the selector, in which and a light modulator for modulating the RZ modulation format into an optical signal light from the light source.

  According to the present invention, the circuit size, power consumption, and cost can be reduced by eliminating the need for an additional circuit for analog processing.

It is a block diagram which shows the circuit structure of the pre-equalization optical transmitter which concerns on Embodiment 1 of this invention. It is explanatory drawing which shows the tap coefficient setting operation | movement by Embodiment 1 of this invention. It is explanatory drawing which shows in part a block diagram the operation | movement which produces | generates a pre-equalization signal using the tap coefficient setting of FIG. It is explanatory drawing which shows the generation | occurrence | production operation | movement of the conventional RZ pre-equalization signal as a comparative example with Embodiment 1 of this invention. It is explanatory drawing which shows the operation | movement of FIG. It is a block diagram which shows the circuit structure of the pre-equalization optical transmitter which concerns on Embodiment 2 of this invention. It is a block diagram which shows the circuit structure of the conventional pre-equalization optical transmitter.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing a circuit configuration of a pre-equalized optical transmitter 100 according to Embodiment 1 of the present invention.
In FIG. 1, a pre-equalized optical transmitter 100 is arranged in parallel with Ich side FIR (Finite Impulse Response) filters 1 and 2 arranged in parallel, and Qch side FIR filters 3 and 4 arranged in parallel. Ich D / A converters 5 and 6, Qch D / A converters 7 and 8 arranged in parallel, Ich selector 9, Qch selector 10, Ich driver 11, Qch A driver 12 on the side, a light source 13, and an optical modulator 14 that optically modulates light emitted from the light source 13 to generate an RZ light pre-equalization signal. The optical modulator 14 includes a phase adjuster 14a (π / 2) for enabling synthesis of orthogonal two-axis (Ich and Qch) RZ optical pre-equalization signals.

  The pre-equalization optical transmitter 100 converts electrical signals (real part component and imaginary part component on the polar coordinate plane) corresponding to Ich and Qch of the constellation (I-Q constellation) of the pre-equalization signal (pre-equalization transmission signal). In this case, the oversampling signal component is generated by parallel processing using the FIR filters 1 to 4 and the parallel signals are selected by the selectors 9 and 10 to generate the RZ pre-equalization signal.

In FIG. 1, a configuration is shown by way of example of “double oversampling” in which pre-equalization is performed by sampling at a rate twice that of the symbol rate of the pre-equalization optical transmitter 100.
Therefore, the Ich and Qch of the pre-equalized optical transmitter 100 parallel the double oversampling Even (even) time pre-equalization signal and Odd (odd) time pre-equalization signal before the selectors 9 and 10. In order to perform processing, FIR filters 1 to 4 in parallel are provided.

After the parallel pre-equalization processing by the parallel FIR filters 1 and 2 (even, odd) on the Icn side, the D / A converters 5 and 6 generate parallel analog signals.
Similarly, after parallel pre-equalization processing by the parallel FIR filters 3 and 4 (even, odd) on the Qcn side, the D / A converters 7 and 8 generate parallel analog signals.

Hereinafter, the selector 9 on the Icn side generates an RZ electrical pre-equalization signal with double oversampling by selecting an analog signal with an Even time and an Odd time with a period of half of the transmission symbol rate.
Similarly, the selector 10 on the Qcn side generates an RZ electrical pre-equalization signal with double oversampling by selecting an analog signal of Even time and Odd time in a period of half of the transmission symbol rate.

  Further, the Ich side and Qch side drivers 11 and 12 adjust the RZ electrical pre-equalization signals from the selectors 9 and 10 to the drive amplitude of the optical modulator 14.

Finally, the optical modulator 14 generates an RZ optical pre-equalization signal by optical signal modulating the emitted light from the light source 13 based on the Ich and Qch RZ electrical pre-equalization signals.
Although FIG. 1 shows the configuration at the time of oversampling twice, it is needless to say that the same configuration can be applied to oversampling of three times or more.

Next, the generation operation of the RZ optical pre-equalization signal according to the first embodiment of the present invention shown in FIG. 1 will be described with reference to FIGS.
FIG. 2 is an explanatory diagram showing a tap coefficient setting operation according to the first embodiment of the present invention, and FIG. 3 is a partial block diagram showing an operation for generating a pre-equalization signal using the tap coefficient setting of FIG. FIG.

In FIGS. 2 and 3, focusing on only the Ich side of the RZ pre-equalized optical transmitter, the setting values of the tap coefficients of the parallel FIR filters 1 and 2 at the time of double oversampling are shown.
In FIG. 2, the impulse response 15 to be compensated is shown as an impulse response signal having a reverse characteristic of signal quality degradation factors (transmission path wavelength dispersion, optical filter, etc.).

Further, the sub impulse response 16 (see dotted line pulse) of the even component is shown as an impulse characteristic for the FIR filter 1 (even) in which the even time component of the impulse response 15 is extracted.
Similarly, the sub impulse response 17 (see solid line pulse) of the Odd component is shown as an impulse characteristic for the FIR filter 2 (odd) in which the Odd time component of the impulse response 15 is extracted.

  In the first embodiment of the present invention, in order to generate an RZ optical pre-equalization signal, the tap coefficients of the parallel FIR filters 1 and 2 corresponding to the number of oversampling are indicated by the sub impulse responses 16 and 17, respectively. In addition, a value obtained by sampling the compensating impulse response 15 at intervals corresponding to the number of oversampling is used.

That is, as the tap coefficients of the parallel FIR filters 1 and 2 (even, odd), sub-impulse responses sampled by shifting at an oversampling interval from the impulse response having the inverse characteristic of the signal quality degradation factor are used.
In FIG. 2, when the impulse response 15 includes the sub-impulse responses 16 and 17 and the oversampling number is “2”, the tap coefficients of the parallel FIR filters 1 and 2 are sampled alternately. Yes.

  Here, as an example of the deterioration factor of the signal quality, the chromatic dispersion of the transmission path and the optical filter are considered, but for the deterioration factor for which the transfer function can be expressed, in any other cases, Similar to the above, pre-equalization processing for generating an RZ optical pre-equalization signal is possible.

In FIG. 3, the tap coefficient setting operation of FIG. 2 is shown together with the pre-equalization circuit (FIR filters 1 and 2, D / A converters 5 and 6, selector 9) on the Ich side.
In FIG. 3, the NRZ input signal sequence 18 has a symbol length T, and the FIR filters 1 and 2 perform convolution operations on the Even component and the Odd component, respectively.

At this time, the sub impulse responses 16 and 17 sampled alternately as shown in FIG. 2 are used as tap coefficients of the FIR filters 1 and 2 for the Even component and the Odd component, respectively.
The FIR filter 1 for Even generates a pre-equalization signal 19 of the Even component by convolution operation of the NRZ input signal sequence 18 and the sub impulse response 16 (Even tap coefficient).

Similarly, the Odd FIR filter 2 generates an Odd component pre-equalized signal 20 by convolution of the NRZ input signal sequence 18 and the sub impulse response 17 (Odd tap coefficient).
The D / A converters 5 and 6 convert the pre-equalized signals 19 and 20 of the Even component and the Odd component generated from the FIR filters 1 and 2 into analog signals.

  The selector 9 performs oversampling on the parallel analog signals from the D / A converters 5 and 6 at the switching timing of the half cycle T / 2 of one symbol length, and outputs the analog signals for the Even component and the Odd component. By alternately selecting, the RZ pre-equalization signal 21 is generated.

  That is, the RZ pre-equalization signal 21 is selected by the selector 9 at the time of oversampling twice, and is acquired as a pulse signal in which an Even component and an Odd component are arranged in time series as shown in FIG.

In this way, by using the impulse responses 16 and 17 having the inverse characteristics of the signal quality degradation factors (wavelength dispersion, optical filter, etc.) as the tap coefficients of the parallel FIR filters 1 and 2 in increments of oversampling, An RZ pre-equalization signal 21 is generated.
Similarly, in the Qch-side pre-equalization circuit (FIR filters 3, 4, D / A converters 7, 8, and selector 10), the Qch-side RZ pre-equalization signal is generated as described above.

  Note that the RZ pre-equalization signal 21 is an output signal from the parallel FIR filters 1 and 2 as shown in FIG. 3 regardless of the symbol modulation scheme (BPSK, QPSK, 16QAM, etc.) of the NRZ input signal sequence 18. Can be generated by applying a configuration in which the selector 9 is arranged within one symbol length T.

  In addition, here, the number of FIR filters 1 and 2 in parallel for one channel (the number of oversampling) is set to “2”, but the present invention is not limited to this, and the number of oversampling is set to “3” or more. It may be set.

Next, as a comparative example with the first embodiment, the operation of the conventional RZ pre-equalized optical transmitter will be described with reference to FIG. 4 and FIG.
FIG. 4 is an explanatory diagram conceptually showing the operation of generating a conventional RZ pre-equalization signal, and shows a conventional pre-equalization modulation configuration.
FIG. 5 is an explanatory diagram showing the operation of the serial FIR filter 22 in FIG. 4 when the RZ signal is input.

In FIG. 4, the conventional RZ pre-equalized optical transmitter includes a serial FIR filter 22, and an input signal to the serial FIR filter 22 includes an RZ input signal 23.
The serial FIR filter 22 holds a double oversampling tap coefficient, and uses the impulse response 15 (compensation impulse response 15 in FIG. 2) having the inverse characteristic of the signal quality deterioration factor as the tap coefficient. .

  In FIG. 5, the series FIR filter 22 is a multiplication circuit 23 of the tap coefficient of the Even component (see the dotted line pulse) included in the impulse response 15 and the multiplication of the tap coefficient of the Odd component (see the solid line pulse) included in the impulse response 15. A circuit 24, an adder 25 for adding the calculation results of the multiplication circuits 23 and 24, and multipliers 26a and 26b (see black frames) in which “0 value” of the RZ signal is alternately input in the Even time and Odd time. It is equipped with.

When the RZ signal is input to the serial FIR filter 22, the multiplication circuits 26a and 26b to which the value of the RZ signal is input are different between the Even time and the Odd time, as shown in FIG.
As a result, as shown in FIG. 5, in the serial FIR filter 22, the Even component of the impulse response 15 (the sub impulse response 16 of the Even component in FIG. 2) of the inverse characteristic of the signal quality degradation factor is convoluted at the Even time. In the Odd time, the Odd component (sub-impulse response 17 of the Odd component in FIG. 2) of the impulse response 15 having the inverse characteristic of the signal quality degradation factor is convoluted.

  The above operation by the serial FIR filter 22 is a time series of the same convolution operation processing as the convolution operation in the Even component and Odd component of the parallel FIR filters 1 to 4 according to the first embodiment (FIGS. 1 to 3) of the present invention. Is equivalent to running on

  That is, in FIG. 1 to FIG. 3, an NRZ input signal is input to FIR filters 1 to 4 arranged in parallel, and an RZ pre-equalization signal can be generated by pre-equalization processing by the FIR filters 1 to 4 and the selectors 9 and 10. It is shown that.

  Therefore, as in the first embodiment (FIG. 1) of the present invention, the RZ pre-equalization signal can be generated by the circuit configuration using the parallel FIR filters 1 to 4 and the selectors 9 and 10, and the conventional configuration Compared to (FIGS. 4, 5, and 7), an additional circuit for RZ is not required, and the circuit size, power consumption, and cost can be reduced.

  As described above, the pre-equalized optical transmitter according to the first embodiment (FIGS. 1 to 3) of the present invention is arranged in parallel to compensate for a signal quality deterioration factor by pre-equalization. FIR filters 1 to 4 for generating an equalized signal, D / A converters 5 to 8 for converting the pre-equalized signal into an analog signal, selectors 9 and 10 for converting the parallel signal into a serial signal containing one symbol length, A light source 13 and an optical modulator 14 that modulates the light from the light source 13 into an optical signal based on the serial signal.

The D / A converters 5 to 8 are inserted between the FIR filters 1 to 4 and the selectors 9 and 10 and convert the respective pre-equalized signals from the FIR filters 1 to 4 arranged in parallel into analog signals. .
The selectors 9 and 10 convert the analog signals from the D / A converters 5 to 8 into serial signals.

The FIR filters 1 to 4 arranged in parallel use sub-impulse responses 16 and 17 sampled by shifting at an oversampling interval from the impulse response 15 having the inverse characteristic of the signal quality deterioration factor as the tap coefficients.
This makes it possible to generate an RZ optical pre-equalization signal without using an additional circuit for analog processing, and the circuit size, power consumption, and cost can be reduced.

Embodiment 2. FIG.
In the first embodiment (FIG. 1), the D / A converters 5 to 8 are inserted between the FIR filters 1 to 4 and the selectors 9 and 10. However, as shown in FIG. And D / A converters 5A and 7A may be inserted with selectors 9A and 10A.
FIG. 6 is a block diagram showing a circuit configuration of a pre-equalized optical transmitter 100A according to Embodiment 2 of the present invention. Components similar to those described above (see FIG. 1) are denoted by the same reference numerals as those described above. Or, “A” is added after the reference numerals, and the detailed description is omitted.

  6, the pre-equalization optical transmitter 100A includes FIR filters 1 to 4 arranged in parallel, a selector 9A that selects digital data from the Ich-side FIR filters 1 and 2, a Qch-side FIR filter 3, 4, a selector 10A that selects digital data from 4, a D / A converter 5A that converts digital data from the selector 9A into an analog signal, a D / A converter 7A that converts digital data from the selector 10A into an analog signal, Drivers 11 and 12, a light source 13, and an optical modulator 14 are provided.

In this case, the selectors 9A and 10A select parallel digital signals from the FIR filters 1 to 4, and the D / A converters 5A and 7A convert the parallel digital signals into analog signals.
Note that the tap coefficient set value selection processing for the FIR filters 1 to 4 and the processing operation of the optical modulation unit 14 are the same as those in the first embodiment.
FIG. 6 also shows a circuit configuration at the time of oversampling twice as described above, but it can also be configured for oversampling twice or more.

As described above, according to the second embodiment (FIG. 6) of the present invention, selectors 9A and 10A are inserted between FIR filters 1 to 4 and D / A converters 5A and 7A. Each pre-equalized signal from the FIR filters 1 to 4 is converted into a serial signal.
The D / A converters 5A and 7A convert the serial signals from the selectors 9A and 10A into analog signals.
Also in the circuit configuration described above, the RZ optical pre-equalization signal that is finally generated from the optical modulator 14 is the same as described above, and has the same operational effects as described above.

  1-4 FIR filter, 5-8, 5A, 7A A / D converter, 9, 10, 9A, 10A selector, 11, 12 driver, 13 Light source, 14 Optical modulator, 16, 17 Sub impulse response, 18 Input signal Series, 19 Even component pre-equalization signal, 20 Odd component pre-equalization signal, 21 RZ pre-equalization signal, 100, 100A pre-equalization optical transmitter.

Claims (2)

An NRZ input signal sequence having one symbol length and a first sub-impulse response and a second sub-impulse response sampled alternately by oversampling from an impulse response having a reverse characteristic of a signal quality degradation factor are input, and the signal quality A pre-equalization transmitter that compensates for deterioration factors by pre-equalization,
A first FIR filter for the Even component that generates and outputs a first pre-equalization signal of the Even component by convolution of the input NRZ input signal sequence and the input first sub-impulse response;
A second pre- equalized signal having an Odd component is generated by convolution with the input NRZ input signal sequence and the input second sub-impulse response, arranged in parallel with the first FIR filter. A second FIR filter for the output Odd component ;
Each of the first 1FIR said first pre-equalized signal is input from the filter and the first 2FIR said second pre-equalized signal inputted from the filter into an analog signal, and a D / A converter for outputting,
A selector that outputs a serial signal within one symbol length by alternately selecting the first pre-equalization signal and the second pre-equalization signal input from the D / A converter ;
A light source;
An optical modulator that modulates light from the light source into an optical signal in an RZ modulation format based on the serial signal input from the selector ;
A pre-equalized optical transmitter characterized by comprising: An NRZ input signal sequence having one symbol length and a first sub-impulse response and a second sub-impulse response sampled alternately by oversampling from an impulse response having a reverse characteristic of a signal quality degradation factor are input, and the signal quality A pre-equalization transmitter that compensates for deterioration factors by pre-equalization,
A first FIR filter for the Even component that generates and outputs a first pre-equalization signal of the Even component by convolution of the input NRZ input signal sequence and the input first sub-impulse response;
A second pre-equalized signal having an Odd component is generated by convolution with the input NRZ input signal sequence and the input second sub-impulse response, arranged in parallel with the first FIR filter. A second FIR filter for the output Odd component;
By alternately selecting the first pre-equalization signal input from the first FIR filter and the second pre-equalization signal input from the second FIR filter, a serial signal within one symbol length is output. A selector,
A D / A converter that converts the serial signal input from the selector into an analog signal and outputs the analog signal;
A light source;
An optical modulator that modulates light from the light source into an optical signal in an RZ modulation format based on the serial signal input from the D / A converter;
A pre-equalized optical transmitter characterized by comprising:

Images