JP2014154926A - Reception signal processing apparatus and reception signal processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that, in a reception signal processing apparatus using a constant modulus algorithm (CMA), reception sensitivity may change while depending on a polarization state of a carrier wave.SOLUTION: A reception signal processing apparatus comprises: a polarization separation section for receiving a transmission signal in which carrier waves are modulated, separating the carrier waves in polarizing directions which are orthogonal to each other, and outputting a first polarization signal and a second polarization signal which are signal components of the carrier waves; a power branching ratio calculation section for calculating an average power branching ratio from amplitudes of the first polarization signal and the second polarization signal; a filter section for applying equalization processing to the first polarization signal and the second polarization signal and outputting a first equalized signal and a second equalized signal; and a tap coefficient control section which uses the average power branching ratio to calculate a tap coefficient specifying operation of the filter section and sets the tap coefficient to the filter section.

Description

  The present invention relates to a received signal processing apparatus and a received signal processing method, and more particularly, to a received signal processing apparatus and a received signal processing method using a constant envelope algorithm (CMA).

  In recent years, there has been an increasing demand for transmitting a huge amount of image data that reflects the state of land from an artificial satellite or an aircraft to the ground in a very short time in a disaster. In order to realize this, optical space communication capable of realizing high-speed and large-capacity transmission as compared with conventional microwave communication has attracted attention. In optical space communication, since uplink and downlink signal light propagates in the same space, it is necessary to ensure sufficient isolation between the uplink and the downlink. For this reason, it is common to use, for example, a single polarized and circularly polarized signal light. That is, a single polarization optical signal in which a single polarization component of an optical carrier wave is modulated is used as the transmission signal.

  On the other hand, in the field of optical fiber communication technology capable of high-speed and large-capacity transmission, the digital coherent communication method has become mainstream. An example of a digital coherent receiving apparatus used in a digital coherent communication system is described in Patent Document 1.

  The related digital coherent receiver described in Patent Document 1 includes a polarization beam splitter, a 90-degree hybrid circuit, a local oscillator, a photoelectric converter, and an analog / digital (A / D) converter. The polarization beam splitter separates, for example, polarization multiplexed QPSK (Quadrature Phase Shift Keying) signal light into two orthogonally polarized signal lights. The 90-degree hybrid circuit mixes each separated optical signal with the local oscillation light from the local oscillator, and outputs optical signals corresponding to the in-phase component and the quadrature component, respectively. The output optical signal of the 90-degree hybrid circuit is converted into an electric signal by a photodiode as a photoelectric converter, and then converted from an analog to a digital electric signal by an A / D converter.

In a related digital coherent receiving apparatus, a butterfly adaptive FIR (Finite Impulse Response) filter as shown in FIG. 12 is used for digital signal processing that separates polarization multiplexed signals and performs retiming. The filter coefficient of the adaptive FIR filter is estimated using a constant envelope algorithm (CMA). The butterfly adaptive FIR filter using the CMA shown in FIG. 12 inputs x _in [k] and y _in [k], which are signals digitized after coherent detection in a 90-degree hybrid circuit. Then, with respect to the input digital signals x _in [k] and y _in [k], a filter coefficient is expressed as shown in the following equation (2) using an error function expressed by the following equation (1). Update.
ε _x [k] = (r ² − | x _out [k] | ² ) ^m
ε _y [k] = (r ² − | y _out [k] | ² ) ^m
(1)
Here, “k” is a sample index, “r” is a constant determined by amplitude, and “m” is an arbitrary constant. Further, “μ” in the following formula (2) is a convergence coefficient and is a constant that can be arbitrarily set.
h _xx [k + 1] = h _xx [k] + με _x [k] x _in [k] x ^* _out [k]
h _xy [k + 1] = h _xy [k] + με _x [k] x _in [k] y ^* _out [k]
h _yx [k + 1] = h _yx [k] + με _y [k] y _in [k] x ^* _out [k]
h _yy [k + 1] = h _yy [k] + με _y [k] y _in [k] y ^* _out [k]
(2)

  The related digital coherent receiver performs a series of demodulation processing such as carrier phase / frequency estimation processing, symbol determination processing, decoding processing, etc. for compensating for the effects of phase noise and frequency offset on the equalized signal, The transmitted information is taken out.

  By applying such a digital coherent optical reception method to optical space communication, it is possible to suppress the influence of Doppler shift, which is a problem in optical space communication, by digital signal processing, and realizes reception with high sensitivity. It is expected to be performed (see, for example, Patent Document 2).

JP 2012-044626 A (paragraphs “0005” to “0015”, FIGS. 1 and 2) International Publication No. 2012/132374 (paragraphs “0002” to “0019”, “0050” to “0069”)

  However, since the butterfly type FIR filter using CMA shown in FIG. 12 is configured to assume signal processing for polarization multiplexed QPSK optical signals, it receives a single polarization QPSK optical signal generally used in optical space communication. In doing so, there are the following problems.

FIG. 13 shows a constellation after a single polarization QPSK optical signal is received and compensated by a carrier phase / frequency estimation process in an associated digital coherent receiver. From the figure, it can be seen that the quality of the signal reproduced after reception is insufficient. The polarization state of the single-polarized QPSK optical signal at this time is a polarization state where the polarization angle θ = 0 ° and the ellipticity χ = 0 ° when displayed by the Poincare sphere. Equalization using CMA when there is a difference in amplitude between x _in [k] and y _in [k], which are signals digitized after coherent detection, as in such a polarization state There is a problem in that good signal quality cannot be obtained even if processing is performed.

The cause is considered as follows. Digital signal _x in _[k], also for _y in [k] the amplitude is very small S / N (Signal / Noise) ratio is low signal of the upper so as to maximize the amplitude by the above formula (1) The filter coefficient is updated as shown in Equation (2). Then, it is considered that the signal quality is deteriorated by synthesizing the signal having the originally high S / N ratio before the equalization and the signal after the equalization of the signal having the low S / N ratio.

  Thus, when a single polarization signal in which a single polarization component of a carrier wave is modulated is received by a reception signal processing device using a constant envelope algorithm (CMA), good signal quality cannot be obtained. There was a problem. That is, the reception signal processing apparatus using CMA has a problem that the reception sensitivity changes depending on the polarization state of the carrier wave.

  An object of the present invention is to provide a received signal that solves the problem that the reception sensitivity changes depending on the polarization state of the carrier wave in the received signal processing apparatus using the constant envelope algorithm (CMA), which is the problem described above. A processing device and a received signal processing method are provided.

  The reception signal processing apparatus of the present invention receives a transmission signal in which a carrier wave is modulated, separates the carrier wave in polarization directions orthogonal to each other, and separates the first polarization signal and the second polarization signal as respective signal components. A polarization separation unit that outputs a signal, a power branching ratio calculation unit that calculates an average power branching ratio from the amplitudes of the first polarization signal and the second polarization signal, a first polarization signal, and a second polarization signal The tap coefficient that defines the operation of the filter unit is calculated using the filter unit that performs the equalization process on the polarization signal and outputs the first equalized signal and the second equalized signal, and the average power branching ratio. A tap coefficient control unit that sets the tap coefficient in the filter unit.

  The received signal processing method according to the present invention acquires a transmission signal in which a carrier wave is modulated, separates the carrier wave into polarization directions orthogonal to each other, and separates the first polarization signal and the second polarization signal as respective signal components. Output a signal, calculate an average power branching ratio from the amplitudes of the first polarization signal and the second polarization signal, and based on a constant envelope algorithm using an error function including a function of the average power branching ratio The tap coefficient used for the equivalent processing is calculated, and the first polarization signal and the second polarization signal are equalized by using the tap coefficient to obtain the first equalization signal and the second equalization signal. Output.

  According to the received signal processing apparatus and the received signal processing method of the present invention, even when the constant envelope algorithm (CMA) is used, good reception sensitivity can be obtained without depending on the polarization state of the carrier wave.

It is a block diagram which shows the structure of the received signal processing apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the structure of the digital coherent optical receiver which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the signal light receiving part with which the digital coherent optical receiver which concerns on the 2nd Embodiment of this invention is provided. It is a block diagram which shows the structure of the polarization reproducing part with which the digital coherent optical receiver which concerns on the 2nd Embodiment of this invention is provided. It is a flowchart for demonstrating operation | movement of the polarization reproduction part with which the digital coherent optical receiver which concerns on the 2nd Embodiment of this invention is provided. It is a figure which shows the constellation of the signal which reproduced | regenerated the single polarization QPSK optical signal by the digital coherent optical receiver which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the polarization reproducing part with which the digital coherent optical receiver which concerns on the 3rd Embodiment of this invention is provided. It is a flowchart for demonstrating operation | movement of the polarization reproduction part with which the digital coherent optical receiver which concerns on the 3rd Embodiment of this invention is provided. It is a figure which shows the constellation of the signal which reproduced | regenerated the single polarization QPSK optical signal by the digital coherent optical receiver which concerns on the 3rd Embodiment of this invention. It is a block diagram which shows another structure of the polarization reproducing part with which the digital coherent optical receiver which concerns on the 3rd Embodiment of this invention is provided. It is a flowchart for demonstrating another operation | movement of the polarization reproducing part with which the digital coherent optical receiver which concerns on the 3rd Embodiment of this invention is provided. It is a block diagram which shows the structure of the butterfly type adaptive FIR filter with which a related digital coherent receiver is equipped. It is a figure which shows the constellation after receiving a single polarization | polarized-light QPSK optical signal in the related digital coherent receiver, and compensating by the carrier phase / frequency estimation process.

  Embodiments of the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of a received signal processing apparatus 100 according to the first embodiment of the present invention. The received signal processing apparatus 100 includes at least a polarization separation unit 110, a power branching ratio calculation unit 120, a filter unit 130, and a tap coefficient control unit 140.

  The polarization separation unit 110 receives a transmission signal in which a carrier wave is modulated, separates the carrier wave into polarization directions orthogonal to each other, and separates the first polarization signal X and the second polarization wave as respective signal components. The signal Y is output. The power branching ratio calculation unit 120 calculates the average power branching ratio α from the amplitudes of the first polarization signal X and the second polarization signal Y.

  The filter unit 130 performs an equalization process on the first polarization signal X and the second polarization signal Y, and outputs a first equalization signal XE and a second equalization signal YE. Here, the tap coefficient control unit 140 calculates a tap coefficient that defines the operation of the filter unit 130 using the average power branching ratio α, and sets the calculated tap coefficient in the filter unit 130.

  FIG. 1 shows a configuration in which the received signal processing apparatus 100 includes a signal selection unit 150. The signal selection unit 150 compares the absolute values of the first equalization signal XE and the second equalization signal YE, selects the one with the larger absolute value, and outputs it as the equalization signal E. Not limited to this, as shown in the third embodiment, a configuration including a phase delay calculation unit and a phase difference correction unit may be employed.

  Here, the tap coefficient control unit 140 updates the tap coefficient based on a constant envelope algorithm (CMA) using an error function including the function f (α) of the average power branching ratio α. can do. As the transmission signal, for example, a single polarization optical signal in which a single polarization component of an optical carrier wave is subjected to multilevel phase modulation can be used.

Specifically, the average power branching ratio α is, for example, an average power determined by an amplitude ratio of x _in [k] and y _in [k], which is a signal digitized after coherent detection, as shown in the following formula (3): A branching ratio α [k] can be used.

The filter unit 130 can be an adaptive FIR filter. The tap coefficient control unit 140 uses a CMA algorithm based on an error function represented by the following formula (4) using the function f (α [k]) of the average power branching ratio α [k] calculated here as a parameter. Can do. In formula (4), f (α [k]) is a function satisfying 0 ≦ f (α [k]) ≦ 1, specifically, for example, f (α [k]) = α [k], It can be a monotonically increasing function such as f (α [k]) = α [k] ² . “N” represents a finite number of symbols used for averaging, “r” represents a constant determined by amplitude, and “m” represents an arbitrary constant.
ε _x [k] = (f (α [k]) r ² − | x _out [k] | ² ) ^m
ε _y [k] = ((1−f (α [k])) r ² − | y _out [k] | ² ) ^m
(4)

  In the received signal processing method according to the present embodiment, first, a transmission signal in which a carrier wave is modulated is acquired, the carrier wave is separated into polarization directions orthogonal to each other, and a first polarization signal as each signal component is obtained. A second polarization signal is output. Subsequently, an average power branching ratio between the first polarization signal and the second polarization signal is calculated. Then, a tap coefficient used for the equivalent process is calculated using the average power branching ratio, and the first polarization signal and the second polarization signal are equalized using the tap coefficient to perform the first process. An equalized signal and a second equalized signal are output. At this time, the absolute values of the first equalized signal and the second equalized signal are compared, and the one having the larger absolute value can be selected and output.

  As described above, in the received signal processing apparatus and the received signal processing method of the present embodiment, the tap coefficient is calculated using the average power branching ratio between the first polarization signal and the second polarization signal, and the tap coefficient is calculated. An equalization process is performed using the coefficients. As a result, it is possible to avoid deterioration in reception sensitivity even in a polarization state where the amplitude difference between the first polarization signal and the second polarization signal is large. That is, according to the received signal processing apparatus 100 of the present embodiment, even when the constant envelope algorithm (CMA) is used, good reception sensitivity can be obtained without depending on the polarization state of the carrier wave. .

[Second Embodiment]
Next, a second embodiment of the present invention will be described. In this embodiment, a case where the received signal processing apparatus is used as a digital coherent optical receiver will be described.

  FIG. 2 is a block diagram showing a configuration of a digital coherent optical receiver 2000 according to the second embodiment of the present invention. The digital coherent optical receiver 2000 includes a polarization beam splitter 2100, a local oscillator 2200, and a signal light receiving unit 2300.

  The polarization beam splitter 2100 separates a single polarized light signal incident from the optical transmission path into a first polarized light signal and a second polarized light signal whose polarization direction is orthogonal to the first polarized light signal. For example, a single polarization QPSK optical signal as a single polarization optical signal is separated into an optical signal X and an optical signal Y whose polarizations are orthogonal to each other. The local oscillator 2200 outputs continuous light. As the local oscillator 2200, for example, a distributed feedback laser diode or the like can be used.

  The signal light receiving unit 2300 uses the continuous light generated by the local oscillator 2200 to convert the optical signal X and the optical signal Y into an electrical signal X and an electrical signal Y by performing coherent detection. And the demodulation process which reproduces | regenerates the single polarization QPSK modulation optical signal transmitted from this electric signal X and the electric signal Y is performed.

  FIG. 3 shows the configuration of the signal light receiving unit 2300. The signal light receiving unit 2300 includes 90-degree hybrid circuits 2311 and 2312, photoelectric converters 2321 to 2324, A / D converters 2331 to 2334, a polarization regeneration unit 2400, and a demodulation processing unit 2340. Here, the polarization beam splitter 2100, the local oscillator 2200, the 90-degree hybrid circuits 2311 and 2312, the photoelectric converters 2321 to 2324, and the A / D converters 2331 to 2334 constitute a polarization separation unit.

The 90-degree hybrid circuit 2311 causes the optical signal X as the first polarization optical signal to interfere with the oscillation light (local oscillation light) of the local oscillator, and separates the in-phase component and the quadrature component to output corresponding optical signals. . Then, the photoelectric transducer 2321 photoelectrically converts an optical signal corresponding to the phase component of the optical signal X, which is the output of the quadrature hybrid circuit 2311, outputs an in-phase signal X _I as a first polarization signal consisting of in-phase component To do. On the other hand, the photoelectric converter 2322 photoelectrically converts an optical signal corresponding to the quadrature component of the optical signal X, which is the output of the quadrature hybrid circuit 2311, and outputs a quadrature signal X _Q as a first polarization signal consisting of orthogonal components .

Similarly, the 90-degree hybrid circuit 2312 causes the optical signal Y as the second polarization optical signal to interfere with the local oscillation light, and separates the in-phase component and the quadrature component to output corresponding optical signals. Then, the photoelectric converter 2323 photoelectrically converts the optical signal corresponding to the in-phase component of the optical signal Y that is the output of the 90-degree hybrid circuit 2312, and outputs the in-phase signal Y _I as the second polarization signal composed of the in-phase component. To do. On the other hand, the photoelectric converter 2324 photoelectrically converts an optical signal corresponding to the orthogonal component of the optical signal Y that is the output of the 90-degree hybrid circuit 2312, and outputs an orthogonal signal _YQ as a second polarization signal composed of the orthogonal component. .

The A / D converter 2331 converts the in-phase signal X _I corresponding to the in-phase component of the optical signal X into a digital signal and outputs a digital signal X _I [k]. A / D converter 2332 outputs a digital signal _X Q [k] and a quadrature signal _{X Q} corresponding to the quadrature component of the optical signal X is converted into digital signals. Similarly, the A / D converter 2333 converts the in-phase signal Y _I corresponding to the in-phase component of the optical signal Y into a digital signal and outputs a digital signal Y _I [k]. The A / D converter 2334 converts the orthogonal signal _YQ corresponding to the orthogonal component of the optical signal Y into a digital signal and outputs the digital signal _YQ [k]. These digital signals X _I [k], X _Q [k], Y _I [k], and Y _Q [k] are input to the polarization recovery unit 2400, respectively. Here, “k” is a sample index.

The polarization recovery unit 2400 receives each digital signal X _I [k], X _Q [k], Y _I [k], and Y _Q [k], and based on the signal components included in each, the adaptive FIR filter The equalization process is performed. At this time, an error function introduced using the average power branching ratio α [k] of the signal branched by the polarization beam splitter 2100 as a parameter can be used. Then, a signal having a high S / N ratio is selected from each signal subjected to equalization processing, and is output to the demodulation processing unit 2340. In addition, it is good also as selecting a signal with a high S / N ratio, after compensating a phase noise and a frequency offset by performing a carrier phase / frequency estimation process about each signal which performed the equalization process.

In this way, the polarization regeneration unit 2400 regenerates the in-phase signal E _I [k] and the quadrature signal E _Q [k] corresponding to the transmission signal based on the detected average power branching ratio α [k]. The demodulation processing unit 2340 then demodulates the in-phase signal E _I [k] and the quadrature signal E _Q [k] and extracts the transmitted information.

  FIG. 4 shows the configuration of the polarization regeneration unit 2400. The polarization recovery unit 2400 includes a power branching ratio calculation unit 2410, FIR filters 2421 to 2424 constituting a filter unit, a tap coefficient control unit 2430, a phase recovery unit 2440, and a signal selection unit 2450. Here, the polarization separation unit, the power branching ratio calculation unit 2410, the FIR filters 2421 to 2424, the tap coefficient control unit 2430, and the signal selection unit 2450 constitute a reception signal processing device.

  FIG. 5 shows a flowchart for explaining the operation of the polarization recovery unit 2400. Hereinafter, the operation of the polarization recovery unit 2400 will be described with reference to FIGS. 4 and 5.

First, the power branching ratio calculation unit 2410 calculates an average represented by the following formula (5) from each digital signal X _I [k], X _Q [k], Y _I [k], and Y _Q [k]. The power branching ratio α [k] is obtained (step S11 in FIG. 5). Here, X _I [k], X _Q [k], Y _I [k], and Y _Q [k] are the real part and imaginary part of the X-polarized signal and the real part and imaginary part of the Y-polarized signal, respectively. “N” is the number of symbols used for averaging.

The tap coefficient control unit 2430 uses the average power branching ratio α [k] obtained by the above equation (5) to use the error functions ε _x [k] and ε _y [k] represented by the following equation (6). Is calculated (step S12). This error function corresponds to the case where f (α [k]) = α [k] and m = 1 in the above equation (4). In the equation, “r” is a constant determined by the amplitude.
ε _x [k] = α [k] r ² −X ^′ _I [k] ² −X ^′ _Q [k] ²
_{ε y [k] = (1} -α [k]) r 2 -Y 'I [k] 2 -Y' Q [k] 2
(6)

The tap coefficient control unit 2430 updates the tap coefficients of the FIR filters 2421 to 2424 according to the following equation (7) based on the error functions ε _x [k] and ε _y [k] (step S13). In Expression (7), “μ” is a convergence coefficient and can be arbitrarily set.
h _xx [k + 1] = h _xx [k] + με _x [k] (X _I [k] + jX _Q [k]) (XE ^' _I [k] -jXE ^' _Q [k])
h _xy [k + 1] = h _xy [k] + με _x [k] (Y _I [k] + jY _Q [k]) (XE ^' _I [k] -jXE ^' _Q [k])
h _yx [k + 1] = h _yx [k] + με _y [k] (X _I [k] + jX _Q [k]) (YE ^' _I [k] -jYE ^' _Q [k])
h _yy [k + 1] = h _yy [k] + με _y [k] (Y _I [k] + jY _Q [k]) (YE ^' _I [k] -jYE ^' _Q [k])
(7)

The FIR filters 2421 and 2422 perform equalization processing on the digital signal X _I [k] + jX _Q [k], and the digital signal X _I ′ [k] + jX _Q ′ [k], Y _I ″ [k] + jY _Q '' [K] is output respectively. Further, the FIR filters 2423 and 2424 perform equalization processing on the digital signal Y _I [k] + jY _Q [k], and the digital signal X _I ″ [k] + jX _Q ″ [k], Y _I ′ [k]. ] + JY _Q ′ [k] are output respectively. Then, these digital signals are synthesized as shown in the following equation (8), and XE _I ′ [k] + jXE _Q ′ [k], YE _I ′ [k] + jYE _Q ′ [k] are obtained as equalized signals. Each is output (step S14). Further, the tap coefficient is updated by feeding back these output signals to the tap coefficient control unit 2430. In formula (8), “*” represents convolution.
XE ^' _I [k] + jXE ^' _Q [k] = h _xx [k] * (X ^' _I [k] + jX ^' _Q [k]) + h _xy [k] * (Y ^' _I [k] + jY ^' _Q [k])
YE ^' _I [k] + jYE ^' _Q [k] = h _yx [k] * (X ^' _I [k] + jX ^' _Q [k]) + h _yy [k] * (Y ^' _I [k] + jY ^' _Q [k])
(8)

In the present embodiment, the phase reproducing unit 2440 is provided. The phase regenerator 2440 uses the equalized signal XE _I ′ [k generated by phase noise caused by the line width of the local oscillation light in the optical synchronous detection or the frequency mismatch (frequency offset) between the local oscillation light and the transmission side optical signal. _{] + jXE Q '[k]} , YE I' [k] + jYE Q ' phase rotation of [k] is corrected by the carrier phase / frequency estimation process (step S15). Then, the first equalization signal XE _I [k] + jXE _Q [k] and the second equalization signal YE _I [k] + jYE _Q [k] are output. For the carrier phase / frequency estimation processing, methods such as Feed Forward M-th Power Algorithm, Decision-Directed Phase-Locked Loop, and Pre-decision-based Angle Differential Estimator can be used.

The signal selection unit 2450 compares the absolute values of the first equalization signal XE _I [k] + jXE _Q [k] and the second equalization signal YE _I [k] + jYE _Q [k] (step S16). . Then, the larger one is selected and output to the demodulation processing unit 2340 as the output signal E _I [k] + jE _Q [k]. That is, when the absolute value XE of the first equalization signal is larger than the absolute value YE of the second equalization signal (step S16 / YES), the signal selection unit 2450 uses the absolute value XE of the first equalization signal. Output (step S17). If the absolute value XE of the first equalization signal is not larger than the absolute value YE of the second equalization signal (step S16 / NO), the signal selection unit 2450 outputs the absolute value YE of the second equalization signal. Is output (step S18).

  FIG. 6 shows a constellation of a signal obtained by reproducing a single polarization QPSK optical signal by the digital coherent optical receiver 2000 of the present embodiment. The polarization state of the single-polarized QPSK optical signal at this time is a polarization state where the polarization angle θ = 0 ° and the ellipticity χ = 0 ° when displayed by the Poincare sphere. From the result of FIG. 6, it can be seen that the digital coherent optical receiver 2000 of the present embodiment can obtain better signal quality than the case of using the related digital coherent receiver shown in FIG. Thus, it can be seen that stable reception sensitivity can be obtained even when there is a difference in amplitude between the X polarization component and the Y polarization component as in the case of a single polarization signal.

  By bypassing the processing in the signal selection unit 2450, even the polarization multiplexed signal in the polarization state in which the difference in amplitude does not occur depending on the polarization direction can be demodulated by the digital coherent optical receiver 2000 of the present embodiment. Is possible. That is, according to the digital coherent optical receiver 2000 of this embodiment, even when the constant envelope algorithm (CMA) is used, good reception sensitivity can be obtained without depending on the polarization state of the carrier wave. it can.

[Third Embodiment]
Next, a third embodiment of the present invention will be described. In this embodiment, a case where the received signal processing apparatus is used as a digital coherent optical receiver will be described.

  The digital coherent optical receiver 3000 according to this embodiment includes a polarization beam splitter 2100, a local oscillator 2200, and a signal light receiving unit 3300. The signal light receiving unit 3300 includes 90-degree hybrid circuits 2311 and 2312, photoelectric converters 2321 to 2324, A / D converters 2331 to 2334, a polarization regeneration unit 3400, and a demodulation processing unit 2340. The digital coherent optical receiver 3000 according to this embodiment is different from the digital coherent optical receiver 2000 according to the second embodiment in the configuration of the polarization regeneration unit 3400. Since other configurations are the same as those of the digital coherent optical receiver 2000 of the second embodiment, description thereof is omitted.

  FIG. 7 shows the configuration of the polarization regeneration unit 3400 according to this embodiment. The polarization recovery unit 3400 includes a power branching ratio calculation unit 3410, FIR filters 3421 and 3422 constituting a filter unit, a tap coefficient control unit 3430, a phase recovery unit 3440, a phase delay calculation unit 3450, and a phase difference correction unit 3460. . Here, the polarization separation unit, the power branching ratio calculation unit 3410, the FIR filters 3421 and 3422, the tap coefficient control unit 3430, the phase delay calculation unit 3450, and the phase difference correction unit 3460 constitute a reception signal processing device. ing.

  The polarization regeneration unit 3400 is different from the polarization regeneration unit 2400 according to the second embodiment in that it includes a phase delay calculation unit 3450 and a phase difference correction unit 3460, and the signal selection unit 2450 is unnecessary. Further, the FIR filters 3421 and 3422 constituting the filter unit are not butterfly type.

  The phase delay calculation unit 3450 calculates an average phase delay δ between the first equalization signal and the second equalization signal, which is a signal after equalization of each polarization signal. The phase difference correction unit 3460 corrects and combines the phase difference between the first equalized signal and the second equalized signal using the average phase delay δ and outputs the resultant signal.

  FIG. 8 is a flowchart for explaining the operation of the polarization recovery unit 3400. Hereinafter, the operation of the polarization recovery unit 3400 will be described with reference to FIGS. 7 and 8.

First, the power branching ratio calculation unit 3410 calculates the average represented by the above equation (5) from each digital signal X _I [k], X _Q [k], Y _I [k], and Y _Q [k]. The power branching ratio α [k] is obtained (step S21 in FIG. 8). Here, X _I [k], X _Q [k], Y _I [k], and Y _Q [k] are the real part and imaginary part of the X-polarized signal and the real part and imaginary part of the Y-polarized signal, respectively. Represents a part.

The tap coefficient control unit 3430 uses the average power branching ratio α [k] obtained by the above equation (5) to use the error functions ε _x [k] and ε _y [k] represented by the above equation (6). ] Is calculated (step S22). The operation so far is the same as the operation of the polarization recovery unit 2400 according to the second embodiment.

The tap coefficient control unit 3430 updates the tap coefficients of the FIR filters 3421 and 3422 according to the following equation (9) based on the error functions ε _x [k] and ε _y [k] (step S23). In Expression (9), “μ” is a convergence coefficient and is a constant that can be arbitrarily set.
h _xx [k + 1] = h _xx [k] + με _x [k] (X _I [k] + jX _Q [k]) (X ^' _I [k] -jX ^' _Q [k])
h _yy [k + 1] = h _yy [k] + με _y [k] (Y _I [k] + jY _Q [k]) (Y ^' _I [k] -jY ^' _Q [k])
(9)

The FIR filters 3421 and 3422 perform equalization processing represented by the following expression (10) on the digital signal X _I [k] + jX _Q [k] and the digital signal Y _I [k] + jY _Q [k], respectively. first equalized signal _{X I '[k] + jX} Q' [k], and the second equalized signal _{Y I '[k] + jY} Q' [k] and outputs, respectively (step S24). In the formula (10), “*” represents convolution (convolution).
X ^′ _I [k] + jX ^′ _Q [k] = (X _I [k] + jX _Q [k]) * h _xx [k]
_{^{Y 'I [k] + jY}} ' Q [k] = (Y I [k] + jY Q [k]) * h yy [k + 1]
(10)

The phase delay calculation unit 3450 compares the absolute values of the first equalization signal X _I '[k] + jX _Q ' [k] and the second equalization signal Y _I '[k] + jY _Q ' [k]. (Step S25). Then, the average phase delay δ [k] of the first equalized signal X _I '[k] + jX _Q ' [k] with respect to the second equalized signal Y _I '[k] + jY _Q ' [k] It calculates by the following formula | equation (11) and (12) according to the absolute value of a digitization signal. In both equations, “L” represents the number of symbols used for averaging.

  That is, when the absolute value of the first equalized signal X ′ is smaller than the absolute value of the second equalized signal Y ′ (step S25 / YES), the phase delay calculation unit 3450 represents the above equation (11). The first average phase delay to be calculated is calculated (step S26). Conversely, if the absolute value of the first equalized signal X ′ is not smaller than the absolute value of the second equalized signal Y ′ (step S25 / NO), the second average represented by the above equation (12) A phase delay is calculated (step S27).

The phase difference correction unit 3460 uses the average phase delay δ [k] obtained by the phase delay calculation unit 3450 to calculate the first equalized signal X _I ′ [k] + jX _Q ′ [ k] and the second equalized signal Y _I '[k] + jY _Q ' [k] are corrected to output a composite signal E ′ (step S28). Thereby, the S / N ratio of the received signal can be improved.
_{^{E 'I [k] + jE}} ' Q [k]
_{^{= X 'I [k] +}} jX' Q [k] + e jδ [k] (Y 'I [k] + jY' Q [k])
(13)

In the present embodiment, the phase reproducing unit 3440 is provided. The phase regenerator 3440 generates a combined signal E _S '[k] that is generated by phase noise caused by the line width of the local oscillation light in optical synchronous detection or the frequency mismatch (frequency offset) between the local oscillation light and the transmission side optical signal. = E _I ′ [k] + jE _Q ′ [k] phase correction is corrected by carrier phase / frequency estimation processing (step S29). Then, the corrected composite signal E _S [k] = E _I [k] + jE _Q [k] is output. For the carrier phase / frequency estimation processing, methods such as Feed Forward M-th Power Algorithm, Decision-Directed Phase-Locked Loop, and Pre-decision-based Angle Differential Estimator can be used.

  FIG. 9 shows a signal constellation in which a single-polarized QPSK optical signal is reproduced by the digital coherent optical receiver 3000 of this embodiment. The polarization state of the single-polarized QPSK optical signal at this time is a polarization state where the polarization angle θ = 0 ° and the ellipticity χ = 0 ° when displayed by the Poincare sphere. From the result of FIG. 9, it can be seen that the digital coherent optical receiver 3000 of the present embodiment can obtain better signal quality than the case of using the related digital coherent receiving apparatus shown in FIG.

  As described above, according to the digital coherent optical receiver 3000 of the present embodiment, even in a polarization state in which a difference in amplitude occurs between the X polarization component and the Y polarization component as in the case of a single polarization signal, Stable reception sensitivity can be obtained. Furthermore, the circuit scale can be reduced as compared with the butterfly FIR filter, and the effect that the convergence of the adaptive filter is more stable can be obtained.

  Further, the polarization recovery unit 3400 can further include a phase delay control unit 3470 as shown in FIG. The phase delay control unit 3470 obtains the average phase delay δ [k] from the phase delay calculation unit 3450, and converts the average phase delay controlled so that the time change of the average phase delay δ [k] does not exceed a predetermined value to the phase difference The data is output to the correction unit 3460. By adopting such a configuration, even when the average phase delay δ [k] output from the phase delay calculation unit 3450 becomes a random value, the S / N ratio is improved in the phase difference correction unit 3460. Is possible.

  The reason why the average phase delay δ [k] becomes a random value is, for example, that even if the average phase delay δ [k] is a finite value (for example, 0 ≦ δ <2π), the average phase exceeding the upper limit or the lower limit A wrap phenomenon in which the value of the delay δ [k] is folded is considered. Accordingly, the phase delay control unit 3470 determines whether or not the average phase delay δ [k] output from the phase delay calculation unit 3450 is random, and the time change of the average phase delay δ [k] exceeds a predetermined value. It can be set as the structure controlled so that it may not exist.

  FIG. 11 is a flowchart for explaining operations of the phase delay calculation unit 3450 and the phase delay control unit 3470. Hereinafter, the operation of the polarization recovery unit 3400 will be described with reference to FIGS. 10 and 11.

The phase delay calculation unit 3450 calculates the average phase delay δ [k] and outputs it to the phase delay control unit 3470 (step S31). The phase delay control unit 3470 calculates an average phase delay difference D = | δ [k] −δ [k−1] | which is a difference from the average phase delay δ [k−1] sampled one sample before (step S32). The phase delay control unit 3470 compares the magnitude of the average phase delay difference D with the magnitude of the predetermined threshold value _Dth (step S33). When the average phase delay difference D is larger than the predetermined threshold value _Dth (step S33 / YES), the average phase delay δ [k-1] sampled one sample before is output (step S34). If the average phase delay difference D is not larger than the predetermined threshold value _Dth (step S33 / NO), the sampled average phase delay δ [k] is output as it is (step S35).

Here, the value of the threshold value _Dth can be set arbitrarily. Further, the phase delay control unit 3470, may have the ability to reset upon reaching the predetermined number of symbols k _re. When reset, the calculation of the average phase delay difference D is restarted with δ [k _re −1] = δ [k _re ]. The predetermined number of symbols k _re can also be set arbitrarily.

  By adopting such a configuration, it is possible to improve the S / N ratio even when the average phase delay δ [k] is a random value.

  The present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the invention described in the claims, and it is also included within the scope of the present invention. Not too long.

DESCRIPTION OF SYMBOLS 100 Reception signal processing apparatus 110 Polarization separation part 120 Power branching ratio calculation part 130 Filter part 140 Tap coefficient control part 150 Signal selection part 2000, 3000 Digital coherent optical receiver 2100 Polarization beam splitter 2200 Local oscillator 2300, 3300 Signal light receiving part 2311, 2312 90-degree hybrid circuit 2321 to 2324 Photoelectric converter 2331 to 2334 A / D converter 2340 Demodulation processing unit 2400, 3400 Polarization regeneration unit 2410, 3410 Power branch ratio calculation unit 2421 to 2424, 3421, 3422 FIR filter 2430 , 3430 Tap coefficient control unit 2440, 3440 Phase reproduction unit 2450 Signal selection unit 3450 Phase delay calculation unit 3460 Phase difference correction unit 3470 Phase delay control unit

Claims (10)

A polarization separation unit that receives a transmission signal in which a carrier wave is modulated, separates the carrier wave into polarization directions orthogonal to each other, and outputs a first polarization signal and a second polarization signal as respective signal components When,
A power branching ratio calculating unit that calculates an average power branching ratio from the amplitudes of the first polarization signal and the second polarization signal;
A filter unit that performs an equalization process on the first polarization signal and the second polarization signal and outputs a first equalization signal and a second equalization signal;
And a tap coefficient control unit that calculates a tap coefficient that defines the operation of the filter unit using the average power branching ratio and sets the tap coefficient in the filter unit. The received signal processing apparatus according to claim 1, wherein the tap coefficient control unit updates the tap coefficient based on a constant envelope algorithm using an error function including a function of the average power branching ratio. The received signal processing apparatus according to claim 1, further comprising: a signal selection unit that compares absolute values of the first equalized signal and the second equalized signal, selects a larger absolute value, and outputs the selected signal. . A phase delay calculation unit and a phase difference correction unit;
The phase delay calculating unit calculates an average phase delay between the first equalized signal and the second equalized signal;
The phase difference correction unit corrects and synthesizes a phase difference between the first equalized signal and the second equalized signal using the average phase delay, and outputs the resultant. Received signal processing device. A phase delay control unit;
The phase delay control unit acquires the average phase delay from the phase delay calculation unit, and outputs an average phase delay that is controlled so that a time change of the average phase delay does not exceed a predetermined value to the phase difference correction unit. The received signal processing apparatus according to claim 4. The received signal processing apparatus according to any one of claims 1 to 5, wherein the transmission signal is a single-polarization optical signal in which a single polarization component of an optical carrier wave is subjected to multilevel phase modulation. The polarization separation unit includes a polarization beam splitter, a 90-degree hybrid circuit, a local oscillator, and a photoelectric converter.
The polarization beam splitter separates the single polarization optical signal into a first polarization optical signal and a second polarization optical signal whose polarization direction is orthogonal to the first polarization optical signal,
The 90-degree hybrid circuit causes the first polarized light signal and the second polarized light signal to interfere with the oscillation light of the local oscillator, respectively, and the first polarized light signal and the second polarized light signal are Separately output into in-phase component and quadrature component,
The photoelectric converter performs photoelectric conversion on the output of the 90-degree hybrid circuit and outputs the first polarization signal and the second polarization signal each having an in-phase component and a quadrature component, respectively. Received signal processing device. Obtaining a transmission signal in which a carrier wave is modulated, separating the carrier wave into polarization directions orthogonal to each other, and outputting a first polarization signal and a second polarization signal as respective signal components;
An average power branching ratio is calculated from the amplitudes of the first polarization signal and the second polarization signal,
Using an error function including a function of the average power branching ratio, calculating a tap coefficient used for equivalent processing based on a constant envelope algorithm,
A reception signal processing method for performing equalization processing on the first polarization signal and the second polarization signal using the tap coefficient to output a first equalization signal and a second equalization signal. The received signal processing method according to claim 8, wherein absolute values of the first equalized signal and the second equalized signal are compared, and a larger absolute value is selected and output. Calculating an average phase delay between the first equalized signal and the second equalized signal;
The received signal processing method according to claim 8, wherein the average phase delay is used to correct and synthesize a phase difference between the first equalized signal and the second equalized signal.

Images