JP2015119437A - Coherent cfp optical transmitter and loss property compensation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To compensate for a loss property of an internal transmission line of a coherent CFP transceiver without requiring work for setting a tap coefficient in accordance with a live-wire insertion/ejection type optical module.SOLUTION: A sinusoidal wave electric signal 10P of each frequency for loss measurement is outputted from D/A converters 13A-13D of a DSP-LSI 10, an amplitude 32S is detected by drive circuits 21A-21D of an optical module 20, and a control circuit 30 calculates a tap coefficient 33S indicating frequency characteristics inverse to the loss property formed from these amplitude values 32S for each of internal transmission lines LA-LD. On the basis of these tap coefficients 33S, a signal processing part 11 of the DSP-LSI 10 compensates for the frequency characteristics of each analog electric signal 10S lost in the internal transmission lines LA-LD by performing digital filter processing on a digital electric signal 11S corresponding to the analog electric signal 10S propagated on the internal transmission lines LA-LD.

Description

  The present invention relates to an optical communication technique, and more particularly to a loss characteristic compensation technique for compensating for a loss characteristic caused by an internal transmission line of a coherent CFP optical transmitter.

  In recent years, communication traffic has increased rapidly due to the expansion of mobile broadband services due to the rapid spread of mobile information terminals such as smartphones, and the expansion of Internet services such as cloud computing and video distribution. In order to cope with this situation, as one of the research and development for practical application of the next generation 100 Gigabit optical network, research and development of pluggable (hot pluggable) coherent CFP (100G Form-factor Pluggable) transceiver Is progressing.

  CFP is one of MSA (Multi-Source Agreement) of a pluggable optical transceiver having an interface of 100 Gb / s (see Non-Patent Documents 1 and 2). In 2010, standardization of 100 GbE (Gigabit Ethernet (registered trademark)) was completed by the IEEE 802 Committee, and CFP transceivers compliant with these standards are used for short-distance communications between data centers and servers.

  FIG. 8 is a block diagram showing a configuration of a conventional CFP transceiver. As shown in FIG. 8, in a conventional CPF transceiver, a 10 Gb / s × 10 electrical signal from a MAC (Media Access Control) -IC is converted into a 25 Gb / s × 4 electrical signal by a gearbox. Thereafter, lasers (LDs) oscillated at different wavelengths are individually driven by 25 Gb / s electrical signals from the gearbox to generate optical signals of 25 Gb / s × 4 wavelengths. The signal is input to the transmission side optical fiber via the multiplexer.

  On the other hand, the optical signal transmitted through the receiving side optical fiber is divided for each wavelength by a demultiplexer, and then a 4-channel photodiode (PD) and a transimpedance amplifier (TIA) corresponding to each wavelength. Is converted into an electric signal of 25 Gb / s × 4. Thereafter, these 25 Gb / s × 4 electric signals are converted in transmission speed and number of channels in the gearbox and sent to the MAC-IC. The MAC-IC is mounted on a printed circuit board and connected to the CFP via a pluggable connector.

FIG. 9 is an explanatory diagram showing the type of CPF. CPFs are categorized as shown in FIG. 9 according to their dimensions and power consumption.
FIG. 10 is a block diagram showing a configuration of a conventional CFP2 / 4 transceiver. As shown in FIG. 10, in the conventional CFP2 / 4 transceiver, in order to reduce the number of I / O pins of the module, the interface of the electrical signal with the MAC-IC is 25 Gb / s × 4. This eliminates the need for a gearbox, and instead performs waveform shaping with CDR (Clock and Data Recovery) for timing. The interface on the line side (optical fiber side) is the same as CFP, and multiplexes and transmits an optical signal modulated with 25 Gb / s On / Off keying by four wavelengths.

  In recent years, application of a digital coherent optical communication technology used for a long-distance optical communication transceiver of about several hundred km to several thousand km to CFP is being studied. Digital coherent technology is a technology in which digital signal processing (DSP) cultivated in the wireless field is applied to coherent optical fiber communication. Coherent detection and DSP enable improvement of reception sensitivity and optical waveform distortion compensation by dispersion (wavelength dispersion or polarization mode dispersion).

  FIG. 11 is a block diagram showing a configuration of a conventional coherent CFP transceiver. As shown in FIG. 11, in a conventional coherent CFP transceiver, a 25 Gb / s × 4 (or 10 Gb / s × 10) electrical signal from a MAC-IC is converted into a 25 Gb / s × 4 signal by a DSP-LSI. Is done. In general, the line side transmission unit of a DSP-LSI is a digital / analog converter (DAC), and therefore any transmission end DSP (format conversion, Nyquist filter, frequency characteristic compensation of optical / electrical components, etc.) in the DSP-LSI, etc. ) To output an analog signal. Further, the DAC may simply output a binary signal. After that, the 25 Gb / s × 4 electrical signal from the DSP-LSI is amplified by the drive circuit (DRV.), Converted to an optical signal by the optical modulator (Modulator), and input to the transmission side optical fiber. The

  On the other hand, the optical signal transmitted through the receiving side optical fiber is mixed by the local light from the LD and the optical demodulator (De-Modulator), and optical / electrically converted by the PD and TIA. The line side receiver of the DSP-LSI is an analog / digital converter (ADC), and converts a 25 Gb / s × 4 analog electric signal from the TIA into a digital electric signal. Thereafter, waveform distortion compensation is performed in the DSP-LSI, and an electric signal of 25 Gb / s × 4 (or 10 Gb / s × 10) is output.

As shown in FIG. 11, in the case of CFP, the DSP-LSI can be housed in the module. In the case of CFP2 / 4, as shown in FIG. 9, the DSP is considered from the viewpoint of dimensions and power consumption. -LSI cannot be stored in the module.
FIG. 12 is a block diagram showing a configuration of a conventional coherent CFP transceiver (CFP2 / 4). For this reason, as shown in FIG. 12, a DSP-LSI is mounted on a printed circuit board instead of in a module, and in a hot-swap optical module corresponding to CFP2 / 4, a drive circuit, an optical modulator / demodulator, a PD , TIA, and LD only.

CFP MSA, Hardware Specification, Revision 1.4, 7 June 2010 CFP MSA, CFP2 Hardware Specification, Revision 1.0, 31 July 2013

  As shown in FIG. 12 described above, in the conventional coherent CFP transceiver (CFP2 / 4), the DSP-LSI is mounted on a printed circuit board and connected to the hot-swap optical module via a pluggable connector. . Therefore, the high frequency wiring from the DSP-LSI to the pluggable connector on the printed circuit board and the DRV. Because the total length of the internal transmission line consisting of the transmission line up to / TIA is several inches, the loss in the internal transmission line in the coherent CFP transceiver becomes large and the electric signal quality deteriorates significantly, and the characteristics of the entire system Cause significant deterioration.

  For such a loss in the internal transmission line, it is possible to compensate for deterioration due to the frequency characteristics of the internal transmission line by using the digital equalizer function of the DSP-LSI. For example, DAC and DRV. The transmission line loss between the two can be pre-equalized using a DSP-LSI transmission end equalizer (Tx EQL). Further, the transmission path loss between the TIA and the ADC can be equalized using a receiving end equalizer (Rx EQL) of the DSP-LSI. Generally, these transmission / reception end equalizers are composed of FIR (Finite Impulse Response) type or IIR (Infinite Impulse Response) type digital filters, and loss is determined by setting tap coefficients according to the frequency characteristics of transmission line loss. Can be compensated.

Therefore, if the tap coefficient corresponding to the loss characteristic of the internal transmission line existing in the coherent CFP transceiver (CFP2 / 4) is set in the transmission end equalizer or the reception end equalizer of the DSP-LSI, the loss in the internal transmission line is reduced. Can be compensated.
However, since the transmission line design in the hot-swap optical module differs depending on the CFP2 / 4 vendor and product, the loss characteristics are not uniform. For this reason, it is necessary for the operator to individually set the tap coefficient of the digital filter in accordance with the optical module used in the coherent CFP transceiver (CFP2 / 4), which increases the setting work load. It was.

  The present invention is for solving such problems, and compensates for the loss characteristics of the internal transmission line of the coherent CFP transceiver without requiring a tap coefficient setting operation according to the hot-swap type optical module. It aims at providing the loss characteristic compensation technology which can be done.

  In order to achieve such an object, the coherent CFP optical transmitter according to the present invention digitally processes input transmission data to generate a plurality of digital electric signals indicating symbols used for optical modulation, and generates these digital electric signals. A DSP-LSI that converts an electrical signal into an analog electrical signal by an individual D / A converter and outputs the analog signal, and is electrically connected to the DSP-LSI via a pluggable connector. A digital circuit for generating a digital coherent optical signal by amplifying an analog electric signal from the D / A converter and digitally modulating continuous light based on the obtained optical modulation driving signal. Output hot-swap type optical module and internal transmission from the D / A converter to which the analog electric signal propagates to the drive circuit A control circuit for calculating a tap coefficient for compensating the loss characteristic of the internal transmission line for each path, and the DSP-LSI receives the analog electric signal from each D / A converter at the time of loss measurement. Instead, the sine wave electric signals having different frequency for loss measurement designated in advance are sequentially switched and output, and the optical module is adapted to the D / A corresponding to the drive circuit by the drive circuit at the time of the loss measurement. The amplitude value of the sine wave electric signal from the converter is detected and output for each loss measurement frequency, and the control circuit outputs the sine wave electric signal output from the drive circuit for each internal transmission line. Based on the loss characteristics related to the internal transmission line composed of the amplitude values, the tap coefficient having a frequency characteristic for compensating the loss characteristic is calculated and output, and the DSP- SI performs digital filter processing for each of the internal transmission lines, based on the tap coefficient of the internal transmission line output from the control circuit, and the digital electric signal corresponding to the analog electric signal propagating through the internal transmission line. Thus, the frequency characteristic of each analog electric signal lost in the internal transmission line is compensated.

  Also, in one configuration example of the coherent CFP optical transmitter according to the present invention, the DSP-LSI sequentially outputs a sine wave electric signal having at least two different specific frequencies selected from the loss measurement frequencies. When calculating the tap coefficient, the control circuit interpolates or extrapolates the amplitude value at each loss measurement frequency of the loss characteristic based on the amplitude value at the specific frequency output from the drive circuit. This is obtained by insertion.

  In one configuration example of the coherent CFP optical transmitter according to the present invention, the control circuit is mounted inside the DSP-LSI.

  In addition, the loss characteristic compensation method according to the present invention generates a plurality of digital electric signals indicating symbols used for optical modulation by digital signal processing of input transmission data, and each of these digital electric signals is converted into an individual D / A. A DSP-LSI that converts the signal into an analog electric signal by a converter and outputs the signal, and a drive circuit that is electrically connected to the DSP-LSI via a pluggable connector and provided for each D / A converter. A hot-swap optical module that amplifies analog electrical signals from the D / A converter and generates and outputs a digital coherent optical signal by digitally modulating continuous light based on the obtained optical modulation drive signal. A loss characteristic compensation method used in a coherent CFP optical transmitter comprising: the D / A conversion signal through which the analog electrical signal propagates. A control circuit for calculating a tap coefficient for compensating for the loss characteristic of the internal transmission line for each internal transmission line from the detector to the drive circuit, and the DSP-LSI is configured to measure each D / A during loss measurement. Instead of the analog electrical signal from the converter, a step of sequentially switching and outputting a sine wave electrical signal of a different frequency for loss measurement specified in advance, and the optical module, at the time of the loss measurement, by each of the drive circuits, Detecting and outputting an amplitude value of a sine wave electric signal from the D / A converter corresponding to the drive circuit for each loss measurement frequency; and a control circuit transmitting the analog electric signal For each internal transmission line from the A / A converter to the drive circuit, the amplitude value of the sine wave electric signal output from the drive circuit is received, and the internal value composed of these amplitude values Calculating and outputting a tap coefficient having a frequency characteristic that compensates for the loss characteristic based on the loss characteristic related to the transmission line; and the DSP-LSI is output from the control circuit for each of the internal transmission lines. Based on the tap coefficient of the internal transmission line, the digital electric signal corresponding to the analog electric signal propagating through the internal transmission line is digitally filtered, thereby causing the frequency of each analog electric signal lost in the internal transmission line. Compensating the characteristic.

According to the present invention, in response to a measurement start instruction to the control circuit, a sine wave electric signal is output, and the loss characteristic in the internal transmission line from the D / A converter to the drive circuit is measured, and each analog electric signal is measured. The tap coefficient for compensating the frequency characteristic is automatically calculated and set in the digital filter of the DSP-LSI.
Therefore, according to the hot-swap type optical module used in the coherent CFP transceiver, the loss characteristic caused by the internal transmission line of the coherent CFP transceiver can be obtained without requiring the operator to set the tap coefficient in advance. Can compensate. Therefore, in the coherent CFP transceiver, it is possible to provide excellent electrical characteristics while reducing the setting work burden.

It is a block diagram which shows the structure of the coherent CFP optical transmitter concerning 1st Embodiment. It is a sequence diagram which shows the loss measurement operation | movement of the coherent CFP optical transmitter concerning 1st Embodiment. It is a graph which shows the loss characteristic which consists of an amplitude value for every frequency for loss measurement. It is a graph which shows the frequency characteristic which consists of a loss value for every frequency for loss measurement. It is a graph which shows the production | generation method of the loss characteristic using extrapolation. It is a graph which shows the production | generation method of the loss characteristic using the interpolation and extrapolation. It is a block diagram which shows the structure of the coherent CFP optical transmitter concerning 3rd Embodiment. It is a block diagram which shows the structure of the conventional CFP transceiver. It is explanatory drawing which shows the classification of CPF. It is a block diagram which shows the structure of the conventional CFP2 / 4 transceiver. It is a block diagram which shows the structure of the conventional coherent CFP transceiver. It is a block diagram which shows the structure of the conventional coherent CFP transceiver (CFP2 / 4).

Next, embodiments of the present invention will be described with reference to the drawings.
[First Embodiment]
First, a coherent CFP optical transmitter 1 according to a first embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram illustrating a configuration of the coherent CFP optical transmitter according to the first embodiment.

  The coherent CFP optical transmitter 1 converts transmission data 1S input from the MAC-IC into a plurality of electric signals indicating symbols used for optical modulation by digital signal processing, and converts continuous light into digital based on these electric signals. It has a function of generating a digital coherent optical signal 20S by optical modulation and outputting it to the transmission optical fiber OFS. The coherent CFP optical transmitter 1 may be used alone as an optical transmitter, or may be used with a coherent CFP optical receiver to constitute a coherent CFP transceiver.

As shown in FIG. 1, the coherent CFP optical transmitter 1 includes a DSP-LSI 10 and an optical module 20 as main circuit configurations.
In the present embodiment, a case where a digital coherent optical signal is generated by optically modulating transmission data 1S based on a QPSK (Quadrature Phase Shift Keying) method will be described as an example. Is not limited to QPSK. For example, in addition to phase shift modulation methods such as BPSK (Binary Phase Shift Keying) and DPSK (Differential Phase Shift Keying), QAM (Quadrature Modulation: Quadrature The present invention can be applied in the same manner when a phase amplitude modulation method such as Amplitude Modulation is used.

  The DSP-LSI is composed of an LSI mounted on a printed circuit board, and is used for optical modulation by subjecting input transmission data 1S to digital signal processing such as FEC encoding, symbol mapping, and pre-equalization by the signal processing unit 11. It has a function of generating four digital electric signals 11S representing symbols, converting these digital electric signals 11S into analog electric signals 10S by individual D / A converters 13A to 13D, and outputting them. In the case of QPSK, transmission data (binary signal) is symbol-mapped to four 2-bit symbols 00, 10, 11, 01, and the phase is shifted by π / 2 on the IQ plane corresponding to these symbols. Two digital electric signals 11S are respectively generated.

  The optical module 20 includes a hot-swap type optical module that is electrically connected to the DSP-LSI 10 via a pluggable connector CN, and includes driving circuits 21A to 21D provided for the D / A converters 13A to 13D. By amplifying the analog electrical signals 10S from the D / A converters 13A to 13D, respectively, and based on the obtained light modulation drive signal 21S, the optical modulator 23 digitally modulates the continuous light 24S from the light source 24. The digital coherent optical signal 20S is generated and output to the transmission optical fiber OFS.

  In the coherent CFP optical transmitter 1 as described above, the present invention includes an arithmetic processing unit such as a CPU, and internal transmission lines from D / A converters 13A to 13D through which analog electric signals 10S propagate to drive circuits 21A to 21D. A control circuit 30 having a function of calculating a tap coefficient 33S for compensating the loss characteristics related to the internal transmission lines LA to LD is provided for each of LA to LD, and based on the loss characteristics measured for each of the internal transmission lines LA to LD. Then, the tap coefficient 33S is calculated by the control circuit 30, and the frequency characteristic of the digital electric signal 11S is compensated by the digital filter (equalizer) 12 of the DSP-LSI 10 based on the tap coefficient 33S.

  Here, as a specific configuration relating to loss measurement, the DSP-LSI 10 may have different loss measurement frequencies designated in advance in place of the analog electrical signal 10S from the D / A converters 13A to 13D during loss measurement. The sine wave electric signal 10P is sequentially switched and output. When the optical module 20 measures the loss, the sine wave electric signals from the D / A converters 13A to 13D corresponding to the drive circuits 21A to 21D in the respective drive circuits 21A to 21D. The amplitude value 32S of the signal 10P is detected for each loss measurement frequency and output to the control circuit 30. The control circuit 30 outputs the amplitude value output from each of the drive circuits 21A to 21D for each of the internal transmission lines LA to LD. Loss characteristics relating to the internal transmission lines LA to LD are generated from 32S.

  In addition, as a specific configuration related to the compensation of the frequency characteristics, the control circuit 30 has a frequency characteristic for compensating the loss characteristics for each of the internal transmission lines LA to LD based on the loss characteristics related to the internal transmission lines LA to LD. The tap coefficient 33S is calculated and output, and the DSP-LSI 10 performs the internal transmission for each internal transmission line LA to LD based on the tap coefficient 33S of the internal transmission line LA to LD output from the control circuit 30. The digital electrical signal 11S corresponding to the analog electrical signal 10S propagating through the lines LA to LD is digitally filtered by the digital filter 12 of the signal processing unit 11, thereby causing each analog electrical loss lost in the internal transmission lines LA to LD. The frequency characteristics of the signal 10S are compensated.

[Operation of First Embodiment]
Next, the operation of the coherent CFP optical transmitter 1 according to the present embodiment will be described with reference to FIG. FIG. 2 is a sequence diagram illustrating a loss measurement operation of the coherent CFP optical transmitter according to the first embodiment.
The control circuit 30 performs the loss measurement operation shown in FIG. 2 when, for example, the optical module 20 is inserted into the printed circuit board on which the DSP-LSI 10 is mounted via the pluggable connector CN in response to an external measurement start instruction. Run.

  First, the control circuit 30 selects an unselected loss measurement frequency from each preset loss measurement frequency (step 100), and control parameters such as the selected loss measurement frequency, output amplitude value, and output time. Based on the above, a control signal 31S for instructing the output of the sine wave electric signal 10P is generated (step 101) and output to each of the D / A converters 13A to 13D of the DSP-LSI 10 (step 102).

  At this time, the frequency for loss measurement is selected from frequencies corresponding to the frequency band of the analog electrical signal 10S. For example, when the interface of the coherent CFP optical transmitter 1 is 100 Gbps and QPSK, the analog electrical signal 10S is 25 GHz, so the selection may be made in 1 GHz increments from 1 GHz to 25 GHz. In general, since the D / A converter for digital coherent optical communication has an operation speed of 50 Gs / s, it is possible to output a sine wave up to 25 GHz which is a Nyquist frequency of 50 Gs / s.

  The D / A converters 13A to 13D of the DSP-LSI 10 set the sine wave output units 14A to 14D provided therein according to the control signal 31S from the control circuit 30 (step 103). The sine wave electrical signal 10P having the loss measurement frequency is output for a specified output time with a specified output amplitude value (step 104).

  For the sine wave output units 14A to 14D, a test function mounted on a general D / A converter may be used. The output amplitude value of the sine wave electric signal 10P is set to the same level at each loss measurement frequency. The output time is a time length that can secure the required time until the drive circuits 21A to 21 detect the amplitude value 32S of the sine wave electric signal 10P and output it to the control circuit 30. In addition, although each sine wave output part 14A-14D can shorten the time which loss measurement requires by outputting the sine wave electric signal 10P in parallel according to the control signal 31S, it is not limited to this. .

  On the other hand, the drive circuits 21A to 21D of the optical module 20 have the amplitude value 32S of the sine wave electric signal 10P output from the D / A converters 13A to 13D by the amplitude detectors 22A to 22D provided therein. Is detected (step 105), and the obtained amplitude value 32S is output to the control circuit 30 (step 106). The amplitude value 32S may be measured at the input stage of the drive circuits 21A to 21D, but may be measured at the output stage of the drive circuits 21A to 21D. In this case, the amplifier in the drive circuits 21A to 21D It is also possible to compensate for the frequency characteristics.

  The control circuit 30 stores the amplitude value 32S output from each of the drive circuits 21A to 21D for each of the internal transmission lines LA to LD (step 107), and confirms that all the loss measurement frequencies have been selected (step 108). If there is an unselected loss measurement frequency (step 108: NO), the process returns to step 100.

  On the other hand, when all the loss measurement frequencies have been selected (step 108: YES), the control circuit 30 generates loss characteristics regarding the internal transmission lines LA to LD from the amplitude value 32S of each loss measurement frequency (step 108). 110) For each of the internal transmission lines LA to LD, a tap coefficient 33S having a frequency characteristic that compensates for the loss characteristic is calculated (step 111). For the tap coefficient 33S, a general tap coefficient calculation method may be used.

Subsequently, the control circuit 30 outputs the tap coefficient 33S of each internal transmission line LA to LD to the signal processing unit 11 of the DSP-LSI 10 (step 112).
The signal processing unit 11 of the DSP-LSI 10 sets the tap coefficient 33S of each internal transmission line LA to LD output from the control circuit 30 in the digital filter 12 (step 113).

  Therefore, after this, when transmitting actual transmission data, the signal processing unit 11 outputs the digital electrical signal 11S corresponding to the analog electrical signal 10S propagating through the internal transmission lines LA to LD based on the tap coefficients 33S. The digital filter 12 performs digital filter processing and then outputs to the D / A converters 13A to 13D. Thereby, the frequency characteristic of each analog electric signal 10S lost in the internal transmission lines LA to LD is compensated.

  FIG. 3 is a graph showing loss characteristics including amplitude values for each loss measurement frequency. In FIG. 3, the horizontal axis represents the frequency for loss measurement of the sine wave electric signal 10P, and the vertical axis represents the amplitude value 32S of the sine wave electric signal 10P detected by the drive circuits 21A to 21D. Here, a case where a 1000 mVpp sine wave electric signal 10P is output from the D / A converters 13A to 13D is shown as an example.

  In the internal transmission lines LA to LD having a length of several inches such as the coherent CFP optical transmitter 1 shown in FIG. 1, the sine wave electric signal 10P having a frequency as low as about 100 MHz is hardly attenuated, and the drive circuit 21A. To 21D. On the other hand, as shown in FIG. 3, since the loss in the internal transmission lines LA to LD increases as the frequency increases, the input amplitude to the drive circuits 21A to 21D decreases, for example, 500 mVpp at 15 GHz. It has decreased to 2.

In this way, if the amplitude value V (f) for each loss measurement frequency f in the drive circuits 21A to 21D is measured, the loss in the internal transmission lines LA to LD at, for example, fLOW = 100 MHz is sufficiently small. Assuming 0 dB, the frequency characteristic regarding the loss LOSS in the internal transmission lines LA to LD can be calculated by the following equation (1).
LOSS (f) = 20LOG {V (f) / V (fLOW)} (1)

FIG. 4 is a graph showing frequency characteristics composed of loss values for each loss measurement frequency. In FIG. 4, the horizontal axis represents the loss measurement frequency of the sine wave electric signal 10P, and the vertical axis represents the loss LOSS in the internal transmission lines LA to LD. For example, it is −6 dB at 15 GHz, and it can be seen that the frequency is reduced to ½ as in FIG. 3.
The control circuit 30 calculates a frequency characteristic that compensates for such a loss characteristic, for example, an inverse frequency characteristic that is substantially constant at each loss measurement frequency (a frequency characteristic indicated by a broken line in FIGS. 3 and 4), Based on this compensation frequency characteristic, the tap coefficient 33S is calculated.

[Effect of the first embodiment]
As described above, according to the present embodiment, the DSP-LSI 10 causes the D / A converters 13A to 13D to replace the analog electrical signal 10S with a sine wave having a different loss measurement frequency specified in advance when measuring the loss. The electrical signal 10P is sequentially switched and output. When the optical module 20 measures the loss, in each of the drive circuits 21A to 21D, the sine wave electrical signal 10P from the D / A converters 13A to 13D corresponding to the drive circuits 21A to 21D. The amplitude value 32S of the sine wave electric signal 10P output from the drive circuits 21A to 21D is detected by the control circuit 30 for each of the internal transmission lines LA to LD. A tap having a frequency characteristic that receives and compensates for the loss characteristic based on the loss characteristic relating to the internal transmission lines LA to LD that includes these amplitude values 32S The DSP-LSI 10 calculates and outputs Equation 33S for each internal transmission line LA to LD based on the tap coefficient 33S of the internal transmission line LA to LD output from the control circuit 30. The digital electrical signal 11S corresponding to the analog electrical signal 10S propagating through the LD is digitally filtered to compensate for the frequency characteristics of each analog electrical signal 10S lost in the internal transmission lines LA to LD. It is.

Thereby, according to the measurement start instruction with respect to the control circuit 30, the sine wave electric signal 10P is output, and the loss characteristics in the internal transmission lines LA to LD from the D / A converters 13A to 13D to the drive circuits 21A to 21D are obtained. The tap coefficient 33S for compensating the frequency characteristic of each analog electric signal 10S is automatically calculated and set in the digital filter 11 of the DSP-LSI 10 after being actually measured.
Therefore, in accordance with the hot-swap type optical module 20 used in the coherent CFP optical transmitter 1, the inside of the coherent CFP optical transmitter 1 is not required without a setting operation in which an operator sets the tap coefficient 33S in advance. Loss characteristics caused by the transmission lines LA to LD can be compensated. For this reason, in the coherent CFP optical transmitter 1, it is possible to provide excellent electrical characteristics while reducing the setting work load.

[Second Embodiment]
Next, a coherent CFP optical transmitter 1 according to a second embodiment of the present invention will be described with reference to FIGS. FIG. 5 is a graph showing a method for generating loss characteristics using extrapolation. FIG. 6 is a graph illustrating a method for generating loss characteristics using interpolation and extrapolation.

  In the first embodiment, an example has been described in which each frequency that covers the entire frequency band of the analog electrical signal 10S is selected as the loss measurement frequency, and the amplitude value 32S is detected for each loss measurement frequency. . In the present embodiment, at least two or more loss measurement frequencies are selected from these loss measurement frequencies to detect the amplitude value 32S, and the amplitude values of the remaining loss measurement frequencies are obtained from the obtained amplitude values 32S. A case where 32S is estimated by extrapolation or interpolation will be described.

  The high frequency characteristics of the loss related to the wiring on the printed board on which the DSP-LSI 10 is mounted can be approximated by a linear log scale. Therefore, it is not necessary to actually output the sine wave electric signal 10P and detect the amplitude value 32S up to 25 GHz which is the Nyquist frequency of the D / A converters 13A to 13A.

  As an example, as shown in FIG. 5, an amplitude value 32S is measured in 1 GHz increments from a loss measurement frequency 1 GHz to 18 GHz corresponding to a range of 70% of the baud rate of the analog electrical signal 10S, and a loss measurement frequency higher than that. 19 GHz to 25 GHz may be estimated by extrapolation from the actually measured amplitude value 32S.

As another example, as shown in FIG. 6, the amplitude value 32S is measured only at two points of the loss measurement frequencies 1 GHz and 10 GHz, and the loss measurement frequencies 2 GHz to 9 GHz are interpolated from the measured amplitude value 32S. The frequency for loss measurement 11 GHz to 25 GHz may be estimated by extrapolation from the actually measured amplitude value 32S.
Thereby, it is possible to reduce the frequency for loss measurement for detecting the amplitude value 32S, and to shorten the time required for the loss characteristic measurement operation.

[Third Embodiment]
Next, a coherent CFP optical transmitter 1 according to a third embodiment of the present invention will be described with reference to FIG. FIG. 7 is a block diagram illustrating a configuration of a coherent CFP optical transmitter according to the third embodiment.

  In the first embodiment, the case where the control circuit 30 is configured independently of the DSP-LSI 10 as shown in FIG. 1 has been described as an example. In the present embodiment, the case where the control circuit 30 is mounted inside the DSP-LSI 10 will be described.

In general, the DSP-LSI can realize a functional unit that controls operations of various circuit units by a program and executes predetermined arithmetic processing. Therefore, the control circuit 30 can be mounted in the DSP-LSI 10 by realizing the control circuit 30 with such a functional unit.
As a result, the circuit configuration of the coherent CFP optical transmitter 1 can be simplified, and downsizing of the apparatus can be realized.

  In the description of each embodiment, the transmission speed of the D / A converter and the input / output electric signal of the A / D converter has been described as 25 Gb / s for the sake of understanding, but in an actual system, Due to the redundancy of FEC (Forward error correction), the input / output electric signal transmission speed of the D / A converter and A / D converter is 30 Gb / s, and the D / A converter and A The operating speed of the / D converter also becomes a frequency corresponding to it.

[Extended embodiment]
The present invention has been described above with reference to the embodiments, but the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. In addition, each embodiment can be implemented in any combination within a consistent range.

  DESCRIPTION OF SYMBOLS 1 ... Coherent CFP optical transmitter, 1S ... Transmission data, 10 ... DSP-LSI, 10S ... Analog electric signal, 11 ... Signal processing part, 11S ... Digital electric signal, 12 ... Digital filter, 13A-13D ... D / A conversion 14A-14D ... sine wave output unit, 20 ... optical module, 20S ... coherent optical signal, 21A-21D ... drive circuit, 21S ... optical modulation drive signal, 22A-22D ... amplitude detector, 23 ... optical modulator, DESCRIPTION OF SYMBOLS 24 ... Light source, 24S ... Continuous light, 30 ... Control circuit, 31S ... Control signal, 32S ... Amplitude value, 33S ... Tap coefficient, CN ... Pluggable connector, LA-LD ... Internal transmission line, OFS ... Transmission system optical fiber.

Claims (4)

The input transmission data is digitally processed to generate a plurality of digital electrical signals indicating symbols used for optical modulation, and these digital electrical signals are converted into analog electrical signals by individual D / A converters and output. DSP-LSI,
Obtained by amplifying the analog electrical signals from the D / A converter by a drive circuit that is electrically connected to the DSP-LSI via a pluggable connector and provided for each D / A converter. A hot-swap optical module that generates and outputs a digital coherent optical signal by digitally modulating continuous light based on the optical modulation driving signal;
For each internal transmission line from the D / A converter through which the analog electrical signal propagates to the drive circuit, a control circuit that calculates a tap coefficient for compensating the loss characteristic of the internal transmission line, and
The DSP-LSI sequentially switches and outputs a sinusoidal electric signal having a different frequency for loss measurement specified in advance, instead of the analog electric signal, from each D / A converter at the time of loss measurement,
The optical module detects and outputs the amplitude value of the sine wave electric signal from the D / A converter corresponding to the drive circuit for each loss measurement frequency at the time of the loss measurement. ,
The control circuit receives, for each internal transmission line, the amplitude value of the sine wave electric signal output from the drive circuit, and based on the loss characteristic related to the internal transmission line composed of these amplitude values, the control circuit calculates the loss characteristic. Calculate and output tap coefficients with frequency characteristics to compensate,
For each internal transmission line, the DSP-LSI digitally converts a digital electrical signal corresponding to an analog electrical signal propagating through the internal transmission line based on a tap coefficient of the internal transmission line output from the control circuit. A coherent CFP optical transmitter characterized by compensating the frequency characteristics of each analog electric signal lost in the internal transmission line by filtering. The coherent CFP optical transmitter of claim 1,
The DSP-LSI sequentially switches and outputs sine wave electric signals of at least two different specific frequencies selected from the loss measurement frequencies,
When calculating the tap coefficient, the control circuit interpolates or extrapolates the amplitude value at each loss measurement frequency of the loss characteristic based on the amplitude value at the specific frequency output from the drive circuit. What is desired is a coherent CFP optical transmitter. The coherent CFP optical transmitter according to claim 1 or 2,
The coherent CFP optical transmitter, wherein the control circuit is mounted in the DSP-LSI. The input transmission data is digitally processed to generate a plurality of digital electrical signals indicating symbols used for optical modulation, and these digital electrical signals are converted into analog electrical signals by individual D / A converters and output. A driving circuit provided for each D / A converter, which is electrically connected to the DSP-LSI via a DSP-LSI and a pluggable connector, and receives analog electric signals from the D / A converter, respectively. Loss used in a coherent CFP optical transmitter including a hot-line insertion / extraction optical module that amplifies and digitally modulates continuous light based on the obtained optical modulation drive signal to generate and output a digital coherent optical signal A characteristic compensation method comprising:
For each internal transmission line from the D / A converter to which the analog electrical signal propagates to the drive circuit, a control circuit that calculates a tap coefficient for compensating for the loss characteristic of the internal transmission line,
The DSP-LSI sequentially switches and outputs a sinusoidal electric signal having a different frequency for loss measurement specified in advance, instead of the analog electric signal, from each D / A converter at the time of loss measurement,
At the time of the loss measurement, the optical module detects and outputs the amplitude value of the sine wave electric signal from the D / A converter corresponding to the drive circuit for each loss measurement frequency by the drive circuits. Steps,
The control circuit receives the amplitude value of the sine wave electric signal output from the driving circuit for each internal transmission line from the D / A converter to the driving circuit through which the analog electric signal propagates, and these amplitude values Calculating and outputting a tap coefficient having a frequency characteristic to compensate for the loss characteristic, based on the loss characteristic related to the internal transmission line consisting of:
For each internal transmission line, the DSP-LSI digitally converts a digital electric signal corresponding to an analog electric signal propagating through the internal transmission line based on the tap coefficient of the internal transmission line output from the control circuit. Compensating the frequency characteristics of each analog electric signal that is lost in the internal transmission line by performing a filtering process.

Images