JP6009953B2 - Optical receiver - Google Patents

Description

  The present invention relates to an optical receiver used in an optical communication system.

  For example, Non-Patent Document 1 discloses an optical communication system that uses a modulation scheme of 64QAM (Quadrature Amplitude Modulation).

W. R. Peng, et. al, "DAC-free generation and 1200-km transmission of 41.4-GBd PDM-64QAM using a single I / Q modulator", Proc. Opto-Electron. Comm. Conf. (OECC), Paper PDP1-3, 2012

  For example, in the optical communication system, a reflected wave from the optical transmission device to the optical reception device may be generated due to the imbalance of the modulator in the optical transmission device. The reflected light is generated from the original optical signal transmitted from the optical transmitter to the optical receiver, and is superimposed on the original optical signal with a predetermined time delay from the original optical signal. As an example of generation of a reflected wave, there is an impedance mismatch of a circuit for applying an electric signal corresponding to a data value to a Mach-Zehnder optical modulator. Specifically, the impedance mismatch of this circuit produces a reflected electrical signal that is a reflected wave of the electrical signal applied to this circuit, and as a result, the continuous light is also modulated by this reflected electrical signal. Occurs.

  The present invention provides an optical receiver that suppresses the influence by suppressing the reflected light even when an optical signal including the reflected light is received from the optical transmitter.

  According to one aspect of the present invention, an optical receiver is configured to receive a first complex value corresponding to a symbol of a received optical signal and indicating a signal position on a complex plane in the past from a symbol received last. First holding means for holding the number of symbols, second holding means for holding a second complex value that is a complex conjugate of each of the first complex values held by the first holding means, A coefficient determining means for determining a coefficient to be multiplied to each of the complex value of 1 and the second complex value; each coefficient determined by the coefficient determining means; and the first complex value and the second complex value. Output means for outputting the sum of the obtained values as a signal position on the complex plane after filtering, wherein the coefficient determining means is an error between the output value of the output means and the reference position of the symbol. The coefficient is updated based on To.

  Even if an optical signal including reflected light is received from the optical transmitter, the reflected light can be suppressed to suppress the influence thereof.

1 is a schematic configuration diagram of an optical receiver according to an embodiment. FIG. The figure which shows the structure of the filter part by one Embodiment. The figure which shows the structure of the filter part by one Embodiment.

  Hereinafter, exemplary embodiments of the present invention will be described with reference to the drawings. In the following drawings, components that are not necessary for the description of the embodiments are omitted from the drawings.

  FIG. 1 is a schematic configuration diagram of an optical receiver according to the present embodiment. The optical receiver 2 according to the present embodiment receives an optical signal modulated according to an arbitrary modulation scheme such as 64 quadrature amplitude modulation (QAM) from the optical transmitter 1. It is assumed that a plurality of reflected waves of the optical signal to be transmitted may be superimposed on this optical signal in a state delayed by a predetermined time due to imbalance in the optical transmitter 1.

  The optical receiver 2 according to the present embodiment includes an optical / electrical (O / E) conversion unit 21, an equalization unit 22, and a filter unit 23. The optical / electrical (O / E) converter 21 converts the received optical signal into an electrical signal, and outputs a signal indicating amplitude information of the in-phase component (I) and a signal indicating amplitude information of the quadrature component (Q). To do. That is, the O / E conversion unit 21 outputs a signal indicating a position on the complex (IQ) plane. In the following description, a signal indicating a position on the IQ plane is referred to as a complex signal.

  The equalizer 22 equalizes the complex signal received from the O / E converter 21 using various known algorithms. That is, the position on the IQ plane indicated by the complex signal received from the O / E converter 21 is converted based on the equalization information at that time, and a complex signal indicating the same or other position on the IQ plane is output. . The equivalent unit 22 may use, for example, a constant envelope reference algorithm (CMA: Constant-Modulus Algorithm) or a radius-directed algorithm (RDA: Radius-Directed Algorithm). Note that an equalization target of the equalization unit 22 is a change in a signal generated in the optical transmission line.

  The filter unit 23 performs digital filter processing on the complex signal received from the equalization unit 22 to remove the reflection component. Here, before describing the configuration of the filter unit 23 of the present embodiment, the configuration of the filter unit disclosed in Non-Patent Document 1 will be described.

  FIG. 2 is a diagram illustrating a schematic configuration of a filter unit disclosed in Non-Patent Document 1. As illustrated in FIG. The shift register 50 receives the equalized complex signal. As shown in FIG. 2, the shift register 50 has N registers, and each register holds a complex value of a complex signal corresponding to one symbol of the received optical signal. The coefficient processing unit 51 has a multiplication unit corresponding to each register of the shift register 50 (indicated by a triangle in the figure), and each multiplication unit calculates the value of the corresponding register and the coefficient set at that time. The multiplied value is output.

The adding unit 52 adds each value output from the coefficient processing unit 51 and outputs the result. The output of the adder 52 is an output after filtering in the filter unit, and the output after filtering is decoded in the decoding unit downstream of the filter unit by comparing with each signal position corresponding to the modulation method. On the other hand, the coefficient updating unit 53 obtains an error that is a difference between the filtered complex value output from the adding unit 52 and a reference position that is an ideal signal position according to the modulation method, and based on this error, a coefficient each coefficient of the processing unit 51 updates the value of FIG. 2 _a 1 ~a _N. For updating the coefficient, a known algorithm used in the equalizer, for example, a least mean square (LMS) method can be used.

For example, if the reflected wave is superimposed with a delay of N symbols and its amplitude is 0.1 times the original signal, the reflected wave is removed by setting the coefficient a _{N in} FIG. 2 to −0.1. be able to. In this way, if the reflected light has only the amplitude of the original signal changed and the phase thereof is the same as that of the original signal, the reflected light can be removed in the configuration of FIG. However, if the ratio of the amplitude of the reflected light to the original signal is different in the in-phase (I) direction and the quadrature (Q) direction, and therefore the phase of the reflected light changes with respect to the original signal, FIG. The reflected light cannot be removed with this configuration.

  Therefore, in this embodiment, the filter part 23 is comprised as shown in FIG. In the filter unit 23 of FIG. 3, the complex value from the equalization unit 22 is branched into two, and one is input to the shift register 50. The other is converted into a conjugate complex value by the converter 54 and input to the shift register 55. That is, when the input complex value is a + jb, the conversion unit 54 outputs a−jb. The shift register 50 and the shift register 55 have N registers. Note that the number of registers N is set to be larger than the number of symbols corresponding to the maximum value of the delay amount of the reflected light included in the signal received by the optical receiver. Each register of the shift register 50 holds a complex value of a complex signal corresponding to one symbol of the modulation signal, and each register of the shift register 55 holds a complex conjugate of the corresponding register of the shift register 50. .

  The coefficient processing unit 51 has a multiplication unit corresponding to each register of the shift register 50 (indicated by a triangle in the figure), and each multiplication unit calculates the value of the corresponding register and the coefficient set at that time. The multiplied value is output. Similarly, the coefficient processing unit 56 has a multiplication unit corresponding to each register of the shift register 55, and each multiplication unit multiplies the value of the corresponding register and the coefficient set at that time. Output.

The adder 52 adds the values output from the coefficient processors 51 and 56 and outputs the result. The output of the adder 52 is the output after filtering in the filter unit 23, and the output after filtering is compared with each signal position corresponding to the modulation method and decoded in the decoding unit downstream of the filter unit. . On the other hand, the coefficient updating unit 53 obtains an error that is a difference between the filtered complex value output from the adding unit 52 and a reference position that is an ideal signal position according to the modulation method, and based on this error, a coefficient each coefficient of the processing unit 51 and 56, and updates the FIG 3 _a 1 ~a _N and _b 1 ~b _N. For updating the coefficient, a known algorithm used in the equalizer, for example, a least mean square (LMS) method can be used.

For example, the reflected wave is superimposed with a delay of N symbols, and the amplitude thereof is 0.1 times in the I direction and 0.2 times in the Q direction. For example, the complex values stored in the register S _N in Figure 3 is assumed to be 8 + 4j, reflected light superimposed on the complex values stored in the register S ₁ becomes 0.8 + 0.8 J. In this case, the coefficient a _{N in} FIG. 3 is set to −0.15, and the coefficient b _N is set to +0.05. In that case, from the multiplier factor _{a N, -1.2-0.6j} is output from the multiplier coefficients _{b N, 0.4-0.2j} is output, therefore, the sum of the two values Can remove reflected light of -0.8-0.8j.

  With the above configuration, even if the reflected wave is different between the in-phase direction and the orthogonal direction, the influence of the reflected wave can be removed on the reception side. In the above embodiment, a shift register and a multiplier are used. However, the present invention is not limited to such a form. That is, the complex value corresponding to the received N symbols in the past and the conjugate complex value thereof are stored in a memory or the like, and further, the value corresponding to each coefficient of the multiplier is also stored in the memory or the like. An addition process may be performed to output a filtered complex value, and a coefficient may be updated.

Claims (4)

An optical receiver,
First holding means for holding a first complex value corresponding to a symbol of a received optical signal and indicating a signal position on a complex plane from the last received symbol by a predetermined number of symbols in the past;
Second holding means for holding a second complex value that is a complex conjugate of each of the first complex values held by the first holding means;
Coefficient determining means for determining a coefficient to be multiplied to each of the first complex value and the second complex value;
An output means for outputting a sum of values obtained by multiplying each coefficient determined by the coefficient determination means and the first complex value and the second complex value as a signal position on the complex plane after filtering;
With
The optical reception apparatus, wherein the coefficient determination unit updates the coefficient based on an error between an output value of the output unit and a reference position of the symbol.   2. The optical receiver according to claim 1, wherein the predetermined number of symbols is a number of symbols corresponding to a time longer than a delay time of a reflected wave included in an optical signal received by the optical receiver.   3. The optical receiving apparatus according to claim 1, wherein the first complex value held by the first holding unit is a value after equalization of the received optical signal.   4. The optical receiver according to claim 1, wherein the coefficient determination unit updates the coefficient by a least mean square algorithm. 5.

Images