JP7028295B1 - Transmission device, receiving side transmission device and transmission method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To compensate for an IQ-to-IQ skew by using zero forcing and to compensate for an IQ-to-IQ imbalance without including multivaluation in the calculation means. A transmission device according to the present invention is a transmission device including a transmission means for generating and transmitting an electric signal including an information signal and a receiving means for restoring an information signal from the detected electric signal. , A first training sequence generator that generates a first electrical signal pattern determined by transmission and reception, and a second training sequence generator that generates a second electrical signal pattern that is different from the first electrical signal pattern. A compensator that compensates for the imbalance between the in-phase component and the orthogonal component of the received signal based on the first electric signal pattern extracted at a predetermined timing and the second electric signal pattern by the receiving means. It is characterized by having. [Selection diagram] Fig. 1

Description

The present invention relates to a transmission device, a receiving side transmission device, and a transmission method, and can be applied to, for example, a transmission device that transmits information by subscriber optical network communication.

In recent years, communication demand has been rapidly increasing due to the sophistication of mobile terminals such as smartphones, the development of high-definition video applications, and the emergence of IoT (Internet of Things) technology. Communication demand is expected to continue increasing as services using 5th generation mobile communication system (5G) technology are planned to be commercialized in the future. On the other hand, as a transmission device for realizing a communication service, not only a backbone optical network but also an optical transmission device of a subscriber optical network is required to have a large capacity. Therefore, research is underway to apply the coherent optical transmission device, which has been conventionally applied to the backbone optical network, to the subscriber optical network.

Since coherent optical transmission uses both the amplitude and phase of signal light to transmit and receive information, it per unit time compared to the intensity-modulated direct detection method optical transmission used in existing subscriber optical networks. It carries a large amount of information and is suitable for large-capacity communication.

An optical transmission system called PON (Passive Optical Network) is applied to an existing subscriber optical network. In PON, the station and a plurality of subscribers perform one-to-many communication. In particular, in the uplink system, which is the communication from the subscriber to the station, since the time division multiple access is performed between the subscribers, burst communication is performed per subscriber. This burst communication is a feature of optical transmission applied to subscriber optical networks, not to backbone optical networks. From the above, there is a demand for large-capacity coherent optical transmission compatible with burst communication.

FIG. 1 shows the configuration of a coherent optical transmission system. The transmission means 11 generates an electric signal including notification information. Here, the electric signal is a time change of an electric amount such as a current and a voltage, and the way of the change represents information. The first light source 12 generates continuous light. The optical modulation means 13 modulates the continuous light output from the first light source 12 according to the electric signal output from the transmission means 11. At this time, the electric signal is converted into an optical signal so that the notification information is included in both the amplitude and / or the phase of the output light. The modulated light is transmitted to the transmission line 30. The second light source 22 generates continuous light. The optical coherent detection means 21 outputs the beat component of the signal light output from the transmission line 30 and the continuous light output from the second light source 22 as an electric signal. At this time, information about the amplitude and phase of the input signal light is converted into an electric signal. The receiving means 23 restores the notification information from the detected electric signal.

In the coherent optical transmission method, an electric waveform representing a report is converted into an optical waveform. At that time, the amplitude and phase of the modulated light change with time. The modulated light can be equivalently converted into the sum of two systems of light whose amplitudes change with time and which are orthogonal to each other. The light modulator inputs two systems of electric signals and outputs the modulated light. On the contrary, the optical demodulator separates and receives two systems of light orthogonal to each other from the input optical signal, and outputs two systems of electric signals. One of these two systems may be referred to as an in-phase (I; in-phase) component, and the other may be referred to as an orthogonal (Q; quadrature) component.

For example, in the modulation / demodulation format often used in existing wireless communication such as 4-phase shift keying (QPSK; quaternary phase shift keying) and quadrature amplitude shift keying (QAM), one type of notification is reported. Is represented by two systems of electrical waveforms. These two systems form a set of signals, and the relationship of the waveform amplitude between the two systems at each time point becomes one piece of information representing the notification.

Therefore, when transmitting and receiving a report through a coherent optical transmission system, an imbalance (amplitude imbalance, phase imbalance, skew) occurs between the I component and the Q component in the middle of the system, and the electrical signal on the receiving side If the imbalance exists before demodulation, the sender's report cannot be restored correctly. IQ imbalance occurs due to the difference in wiring length between the two electrical and optical systems, the inhomogeneity of the signal propagation medium, and the like. Eliminating this IQ-to-IQ imbalance is one of the important technical issues in optical transmission design.

Techniques for resolving such IQ-to-IQ imbalances have been reported in several literatures. In particular, with the recent development of digital signal processing devices, it is possible to compensate for high-speed signals exceeding 10 Gbad, and it is possible to produce a large number of the same processing devices that do not have the inhomogeneity and variation that occur in analog devices. It is becoming.

In Non-Patent Document 1, the amplitude imbalance, phase imbalance and skew generated in the receiving device are compensated by using adaptive signal processing. In Non-Patent Document 2, the amplitude imbalance and the phase imbalance generated in the transmitting side device are compensated by using adaptive signal processing. In Non-Patent Document 3, the amplitude and phase imbalance caused in the transmitting side device are compensated by zero forcing equalization (or one tap equalization).

M.S.Faruk and K.Kikuchi, “Compensation for In-Phase / Quadrature Imbalance in Coherent-Receiver Front End for Optical Quadrature Amplitude Modulation,” IEEE Photon.J., vol.5, no.2, Apr.2013. Q.Zhang, et al, “Modulation-format-transparent IQ imbalance estimation of dual-polarization optical transmitter based on maximum likelihood independent component analysis,” Opt.Exp.vol.27, no.13, pp.18055-18068,24 Jun.2019. R.Matsumoto, et al., “Fast, Low-Complexity Widely-Linear Compensation for IQ Imbalance in Burst-Mode 100-Gb / s / λ Coherent TDM-PON,” OFC 2018, paper M3B.2.

In Non-Patent Documents 1 and 2 described above, adaptive signal processing is used as a means for compensating for IQ-to-IQ imbalance. In adaptive signal processing, the error becomes "0" immediately after the start of processing, and instead of continuing to be "0", the error decreases over time and the error converges to "0" or a certain small value. .. The time spent for the convergence is called the convergence time. Alternatively, the time until the magnitude of the error reaches the magnitude required for the received signal to fall below the error rate required for the system may be referred to as the convergence time. The received information up to the convergence time has many errors, and if the error rate exceeds the error rate required for the system, the received information is invalid. If such an invalid time exists, the information transfer efficiency becomes small, and especially in burst communication, the transmitted / received information is limited to a finite time range, and the influence becomes serious.

Non-Patent Document 3 uses zero forcing as a means of compensating for IQ-to-IQ imbalance. Therefore, there is no problem of the above-mentioned invalid time. The method is to branch the received signal, estimate the amplitude unbalance ratio and phase unbalance amount contained in the received signal from one of them, and calculate the opposite amount of those estimated values to the other received signal. Compensate for interbalance.

However, one problem arises that one of the parameters to be estimated is topology. The phase is calculated using the inverse trigonometric function. The inverse trigonometric function is a multivalued function, and the value is usually calculated only for "0 to 2π radians" or "−π to π radians". In order to remove the influence of noise, it is often the case that measures are taken to improve the accuracy of the estimated phase by calculating and averaging a plurality of phases, and Non-Patent Document 3 also performs such averaging processing. If the calculated phase is limited to the range of 2π radians as described above, the phase after the averaging process may deviate from the true value.

For example, if the result of calculating the true value of 0 radians 5 times is "0, 0, 0, 0.01π, -0.01π radians" without limiting the range of values, if it is limited to 0 to 2π radians, The calculation result is "0, 0, 0, 0.01π, 1.99π radians". In the former case, when the calculation results are averaged, the phase becomes "0 radians" and the phase after the averaging process becomes equal to the true value, whereas in the latter case, when the calculation results are averaged, the phase after the averaging process becomes. It becomes "0.4π radian" and is different from the true value.

Further, Non-Patent Document 3 does not compensate for the skew between IQs.

The present invention uses zero forcing to compensate for IQ-to-IQ skew and to compensate for IQ-to-IQ imbalance without including polyvalence in the calculation means. Transmission device, receiver-side transmission device, transmission-side transmission device, and transmission. It seeks to provide a method.

In order to solve such a problem, the first aspect of the present invention is a transmission device including a transmitting means for generating and transmitting an electric signal including an information signal and a receiving means for restoring an information signal from the detected electric signal. , The first electric signal pattern is a single-sided wave signal in which the amplitude of the negative frequency component is 0 in the two-sided spectrum notation, and the second electric signal pattern is the single-sided wave signal in which the amplitude of the positive frequency component is 0. A second training sequence generator that is a signal and the transmission means generates a first electric signal pattern determined by transmission and reception, and a second electric signal pattern that is different from the first electric signal pattern. The receiving means compensates for the imbalance between the in-phase component and the orthogonal component of the received signal based on the extracted first electric signal pattern and the second electric signal pattern. It has a coefficient derivation unit that derives a coefficient and a compensation unit that compensates for the imbalance between the in-phase component and the orthogonal component of the received signal using the coefficient, and the transmitting means is a first electric signal pattern and a second electric signal pattern. The electric signal pattern is time-divided and multiplexed, and the coefficient derivation unit separates the positive frequency component and the negative frequency component of a predetermined frequency from the extracted spectrum of the first electric signal pattern. A second frequency component separating section that separates a positive frequency component and a negative frequency component of a predetermined frequency from the spectrum of the extracted second electric signal pattern, and a positive frequency component from the first frequency component separating section. A first frequency multiplexing section that obtains a first frequency component by frequency-multiplexing the negative frequency component from the second frequency component separating section, a negative frequency component from the first frequency component separating section, and a first. The second frequency multiplexing section that obtains the second frequency component by frequency multiplexing the positive frequency components from the frequency component separating section of 2, the value of the first frequency component, and the complex conjugate of the first frequency component. A coefficient calculation unit that derives the first coefficient and the second coefficient based on the value obtained by inverting the frequency, the value of the second frequency component, and the value obtained by inverting the frequency of the complex conjugate of the second frequency component. And, the compensator adds the value obtained by multiplying the value of the received signal by the first coefficient and the value obtained by multiplying the value of the complex conjugate of the received signal by the second coefficient, and the in-phase of the received signal. It is characterized by compensating for an imbalance between a component and an orthogonal component .

The second invention is a two- sided spectrum notation, in which a first electric signal pattern is a single-sided band signal having a negative frequency component amplitude of 0 and a single-sided wave signal having a positive frequency component amplitude of 0. A receiving-side transmitting device including a receiving means for recovering an information signal from an electric signal obtained by detecting a received received signal from a transmitting-side transmitting device that transmits a second electric signal pattern in a time-division multiplex. In the coefficient derivation unit, the receiving means derives a coefficient for compensating for the imbalance between the in-phase component and the orthogonal component of the received signal based on the extracted first electric signal pattern and the second electric signal pattern. , A compensator that compensates for the imbalance between the in-phase component and the orthogonal component of the received signal using the coefficient, and the coefficient derivation unit has a positive frequency of a predetermined frequency from the spectrum of the first electric signal pattern extracted. A first frequency component separation unit that separates components and negative frequency components, and a second frequency component separation that separates positive frequency components and negative frequency components of a predetermined frequency from the spectrum of the extracted second electrical signal pattern. A first frequency multiplexing section that obtains a first frequency component by frequency-multiplexing a positive frequency component from a first frequency component separating section and a negative frequency component from a second frequency component separating section. A second frequency multiplexing section that obtains a second frequency component by frequency-multiplexing a negative frequency component from the first frequency component separating section and a positive frequency component from the second frequency component separating section, and a first Based on the value of the frequency component, the inverted value of the complex conjugate frequency of the first frequency component, the value of the second frequency component, and the inverted value of the complex conjugate frequency of the second frequency component. It has a first coefficient and a coefficient calculation unit for deriving a second coefficient, and the compensation unit has a value obtained by multiplying the value of the received signal by the first coefficient and a value obtained by multiplying the value of the complex conjugate of the received signal by the second. It is characterized in that the imbalance between the in-phase component and the orthogonal component of the received signal is compensated by adding the value multiplied by the coefficient .

A third aspect of the present invention is a transmission method for transmitting information between a transmitting means for generating and transmitting an electric signal including an information signal and a receiving means for restoring an information signal from the detected electric signal . The electric signal pattern of 1 is a single-sided wave signal in which the amplitude of the negative frequency component is 0 in the two-sided spectrum notation, and the second electric signal pattern is a single-sided wave signal in which the amplitude of the positive frequency component is 0. Yes, the transmitting means generates a first electric signal pattern determined by transmission and reception and a second electric signal pattern different from the first electric signal pattern, and the first electric signal pattern and the second electric signal are generated. The signal pattern is time-divided and multiplexed, and the receiving means transmits the in -phase component and the orthogonal component of the received signal based on the first electric signal pattern and the second electric signal pattern extracted by the coefficient derivation unit. The coefficient for compensating the imbalance is derived, the compensator uses the coefficient to compensate for the imbalance between the in-phase component and the orthogonal component of the received signal, and the coefficient derivation unit is the first frequency component separation unit. , The positive frequency component and the negative frequency component of a predetermined frequency are separated from the spectrum of the extracted first electric signal pattern, and the second frequency component separation unit is predetermined from the spectrum of the extracted second electric signal pattern. The positive frequency component and the negative frequency component of the frequency are separated, and the first frequency multiplexing section separates the positive frequency component from the first frequency component separating section and the negative frequency component from the second frequency component separating section. The frequency is multiplexed to obtain the first frequency component, and the second frequency multiplexing section picks up the negative frequency component from the first frequency component separating section and the positive frequency component from the second frequency component separating section. The second frequency component is obtained by multiplexing, and the coefficient calculation unit determines the value of the first frequency component, the value obtained by inverting the complex conjugate frequency of the first frequency component, and the value of the second frequency component. , The first coefficient and the second coefficient are derived based on the inverted value of the complex conjugate of the second frequency component, and the compensator multiplies the value of the received signal by the first coefficient. And the value obtained by multiplying the value of the complex conjugate of the received signal by the second coefficient are added to compensate for the imbalance between the in-phase component and the orthogonal component of the received signal .

According to the present invention, zero forcing can be used to compensate for IQ-to-IQ skew and to compensate for IQ-to-IQ imbalance without including multivaluation in the calculation means.

It is an overall configuration diagram which shows the whole configuration of a coherent optical transmission system. It is a block diagram which shows the structure of the transmission means which concerns on embodiment. It is a block diagram which shows the structure of the receiving means which concerns on embodiment. It is a block diagram which shows the structure of the coefficient calculation means which concerns on embodiment. It is a block diagram which shows the structure of the compensation means which concerns on embodiment. In the embodiment, it is explanatory drawing explaining the 1st TS and the 2nd TS transmitted by the transmitting means, and 3rd TS and a 4th TS received by a receiving means.

(A) Main Embodiments In the following, embodiments of a transmission device, a receiving side transmission device, a transmitting side transmission device, and a transmission method according to the present invention will be described in detail with reference to the drawings.

In this embodiment, the case where the present invention is applied to an optical transmission system is illustrated. In particular, a transmission device using an optical coherent detection method in subscriber optical network communication will be exemplified.

(A-1) Configuration of Embodiment This embodiment will also be described with reference to an overall configuration diagram showing the overall configuration of the coherent optical transmission system illustrated in FIG. 1.

In FIG. 1, the optical transmission system 100 includes a transmission side transmission device 10 and a reception side transmission device 20. The transmission side transmission device 10 and the reception side transmission device 20 transmit and receive information via the transmission line 30.

The optical transmission system 100 exemplifies, for example, a case where the optical transmission system 100 is a subscriber optical communication network that employs an optical coherent detection method. Generally, one transmission device includes the transmission side transmission device 10 and the reception side transmission device 20 illustrated in FIG. 1, but in FIG. 1, the configuration of the transmission side and the reception side is distinguished. In FIG. 1, the number of the transmission side transmission device 10 and the number of the reception side transmission device 20 is one. That is, although the transmission system of one transmission device and one transmission value is illustrated, a transmission system of one transmission device and a plurality of transmission devices may be used. For example, in a subscriber optical communication network, one-to-many communication may be performed between one station-side device and a plurality of subscriber-side devices.

[Transmitting device 10 on the transmitting side]
The transmitting side transmission device 10 photomodulates according to the information transmitted to the receiving side transmission device 20, and transmits the modulated light toward the transmission line 30.

As illustrated in FIG. 1, the transmitting side transmission device 10 includes a transmitting means 11, a first light source 12, and an optical modulation means 13.

The transmission means 11 gives an electric signal including the notification information to the optical modulation means 13. Here, the electric signal is a time change of an electric amount such as a current and a voltage, and the way of the time change represents information. For example, in the subscriber optical communication network, the notification information exchanged between the transmitting means 11 of the transmitting side transmission device 10 and the receiving means 23 of the receiving side transmission device 20 is represented by a bit string.

Further, the transmitting means 11 generates a training sequence (TS: Training Sequence) for compensating for IQ-to-IQ imbalances such as amplitude imbalance, phase imbalance, and skew, and transmits the training sequence (TS) to the receiving means 23. That is, on the receiving means 23 side of the receiving side transmission device 20, the transmitting means 11 transmits the training sequence to the receiving means 23 in order to compensate for the imbalance between IQs.

The first light source 12 generates continuous light and gives it to the light modulation means 13.

The optical modulation means 13 modulates the continuous light output from the first light source 12 according to the electric signal output from the transmission means 11. At this time, the electric signal is converted into an optical signal so that the notification information is included in both the amplitude and / or the phase of the output light output from the optical modulation means 13. The modulated light is transmitted to the transmission line 30.

[Structure of transmission means 11]
FIG. 2 is a configuration diagram showing the configuration of the transmission means 11 according to the embodiment. In FIG. 2, the transmission means 11 includes an information symbol sequence generation means 111, a first selection means 112, a first training sequence (TS) generation means 113, a first enable signal generation means 114, and a second selection means 115. , A second training sequence (TS) generating means 116, and a second enable signal generating means 117.

The information symbol sequence generating means 111 generates an electric signal modulated according to the notification information and outputs it to the first selection means 112. The output from the information symbol sequence generating means 111 is defined as an information symbol sequence.

The first TS generating means 113 generates a specific electric signal pattern arranged by transmission / reception as a training sequence (TS) signal and outputs it to the first selection means 112. The training series (TS) signal output by the first TS generating means 113 is referred to as a “first TS”.

The second TS generating means 116 generates a specific electric signal pattern arranged by transmission / reception as a training sequence (TS) signal and outputs it to the second selection means 115. The training series (TS) signal output by the second TS generating means 116 is referred to as a “second TS”. Here, the signal patterns of the second TS and the first TS are different from each other. A detailed description of the first TS and the second TS will be described later.

The first enable signal generation means 114 generates a binary signal indicating the transmission timing of the first TS and outputs the binary signal to the first selection means 112. That is, the first enable signal generation means 114 controls the timing at which the first selection means 112 outputs the first TS to the second selection means 115. The signal output by the first enable signal generating means 114 is referred to as a "first enable signal".

For example, at the time when the first selection means 112 transmits the first TS, the first enable signal generation means 114 outputs High, and at other times, the first enable signal generation means 114 outputs Low. Output.

The second enable signal generation means 117 generates a binary signal indicating the transmission timing of the second TS and outputs the binary signal to the second selection means 115. That is, the second enable signal generation means 117 controls the timing at which the second selection means 115 transmits the second TS. The signal output by the second enable signal generating means 117 is referred to as a “second enable signal”.

For example, at the time when the second selection means 115 transmits the second TS, the second enable signal generation means 117 outputs High, and at other times, the second enable signal generation means 117 outputs Low. Output.

Here, the time when the first enable signal is High and the time when the second enable signal is High are different from each other.

The first selection means 112 inputs the information symbol sequence, the first TS, and the first enable signal, the first enable signal is the first TS at High time, and the first enable signal is Low. It is a means to select and output the information symbol series of the time at the time of.

The second selection means 115 inputs the output signal of the first selection means 112, the second TS, and the second enable signal, and the second enable signal sets the second TS at the time when the second enable signal is High. The enable signal of 2 is a means for selecting and outputting the output signal of the first selection means 112 at the time of Low.

In addition, in the practical transmission means 11, there is a means for further waveform shaping or the like of the converted electric signal in order to improve the transmission quality, but it is omitted here because it deviates from the gist of the present invention.

[Receiving side transmission device 20]
The receiving-side transmission device 20 receives the signal light transmitted from the transmitting-side transmission device 10 via the transmission path 30, and reverse-converts (optically demodulates) the received signal light to acquire information.

As shown in FIG. 1, the receiving side transmission device 20 includes an optical coherent detecting means 21, a second light source 22, and a receiving means 23.

The second light source 22 generates continuous light and gives it to the optical coherent detection means 21.

The optical coherent detection means 21 outputs the beat component of the signal light output from the transmission line 30 and the continuous light output from the second light source 22 as an electric signal. At this time, information about the amplitude and phase of the input signal light is converted into an electric signal.

The receiving means 23 restores the notification information from the electric signal output from the optical coherent detecting means 21. The receiving means 23 of this embodiment extracts the first TS and the second TS from the received signal received from the transmitting means 11, and IQ of the received signal based on the first TS and the second TS. Compensate for imbalances. A detailed description of the receiving means 23 will be described later.

[Receiving means 23]
FIG. 3 is a configuration diagram showing the configuration of the receiving means 23 according to the embodiment.

In FIG. 3, the receiving means 23 includes a third enable signal generating means 231, a first TS extracting means 232, a fourth enable signal generating means 233, a second TS extracting means 234, a Fourier transform means 235, and a compensating means. It has 237, an inverse Fourier transform means 238, and a symbol inverse transform means 239.

The third enable signal generating means 231 is a means for generating a binary signal indicating the arrival timing when the first TS generated by the transmitting means 11 arrives at the receiving means 23 via the optical transmission system. The signal output by the third enable signal generation means 231 is referred to as a "third enable signal". The third enable signal generation means 231 is set to High at the time when the first TS arrives, and is set to Low at other times.

The fourth enable signal generating means 233 is a means for generating a binary signal indicating the arrival timing when the second TS generated by the transmitting means 11 arrives at the receiving means 23 via the optical transmission system. The signal output by the fourth enable signal generation means 233 is referred to as a "fourth enable signal". The fourth enable signal generation means 233 sets High at the time when the second TS arrives, and Low at other times.

Here, the timing at which the first TS and the second TS arrive at the receiving means 23 is, for example, the timing obtained by the receiving means 23 extracting a timing signal indicating the transmission / reception timing with the transmitting means 11. Alternatively, for example, the timing may be known by the receiving means 23, that is, the timing when the receiving means 23 is predetermined with the transmitting means 11.

The first TS extracting means 232 inputs the third enable signal and the received signal, extracts the received signal only when the third enable signal is High, and outputs the received signal to the coefficient calculating means 236. The signal output by the first TS extraction means 232 is referred to as a "third TS".

The second TS extracting means 234 inputs the fourth enable signal and the received signal, extracts the received signal only when the fourth enable signal is High, and outputs the received signal to the coefficient calculating means 236. The signal output by the second TS extraction means 234 is referred to as a "fourth TS".

The coefficient calculation means 236 inputs the third TS and the fourth TS, calculates a numerical value used for the calculation for compensating the IQ-to-IQ imbalance, and gives it to the compensation means 237. That is, the coefficient calculating means 236 calculates the numerical value required for compensating the IQ imbalance based on the third TS and the fourth TS extracted from the received signal from the transmitting side (that is, the transmitting means 11). presume. A detailed description of the configuration and processing of the coefficient calculation means 236 will be described later.

The Fourier transform means 235 inputs a received signal, performs a Fourier transform, and outputs the received signal to the compensating means 237.

The compensating means 237 inputs an output value calculated by the coefficient calculating means 236 (for example, a value of the first coefficient, a value of the second coefficient) and an output signal of the Fourier transform means 235, and outputs the coefficient calculating means 236. Compensate for the IQ imbalance of the received signal according to the value.

The inverse Fourier transform means 238 inputs the output signal of the compensating means 237, performs an inverse Fourier transform, and outputs the signal.

The symbol inverse transform means 239 is a means for restoring the notification information from the output signal of the inverse Fourier transform means 238.

[Coefficient calculation means 236]
FIG. 4 is a configuration diagram showing the configuration of the coefficient calculation means 236 according to the embodiment.

In FIG. 4, the coefficient calculation means 236 includes a first Fourier transform means 301, a first frequency separation means 302, a second Fourier transform means 303, a second frequency separation means 304, and a first frequency multiplexing means 305. First complex conjugate means 306, first frequency inversion means 307, first multiplication means 308, second frequency multiplexing means 309, second complex conjugate means 310, second frequency inversion means 311, first. It has two multiplication means 312, a first addition means 313, a first division means 314, and a second division means 315.

The first Fourier transform means 301 Fourier transforms the third TS and outputs it to the first frequency separation means 302.

The second Fourier transform means 303 Fourier transforms the fourth TS and outputs it to the second frequency separation means 304.

The first frequency separating means 302 separates and outputs a positive frequency component and a negative frequency component with respect to the output signal from the first Fourier transform means 301. Here, the positive frequency component obtained by separation is output to the first frequency dividing means 305 by the first frequency separating means 302, and the negative frequency component is output to the first frequency dividing means 302 by the first frequency separating means 302. The case of outputting to the frequency multiplexing means 309 will be illustrated.

The second frequency separating means 304 separates and outputs a positive frequency component and a negative frequency component with respect to the output signal from the second Fourier transform means 303. Here, the positive frequency component obtained by separation is output to the second frequency dividing means 309 by the second frequency separating means 304, and the negative frequency component is output to the second frequency dividing means 304 by the second frequency separating means 304. The case of outputting to the frequency multiplexing means 305 will be illustrated.

The first frequency multiplexing means 305 has a frequency of a positive frequency component output signal output from the first frequency dividing means 302 and a negative frequency component output signal output from the second frequency separating means 304. Multiplex. Further, the first frequency multiplexing means 305 outputs the signal obtained by frequency multiplexing to the first multiplication means 308 and the first complex conjugate means 306.

The second frequency multiplexing means 309 has a frequency of a positive frequency component output signal output from the second frequency dividing means 304 and a negative frequency component output signal output from the first frequency separating means 302. Multiplex. Further, the second frequency multiplexing means 309 outputs the signal obtained by frequency multiplexing to the second multiplication means 312, the second complex conjugate means 310, and the second division means 315.

The first complex conjugate means 306 outputs the complex conjugate of the output signal from the first frequency multiplexing means 305 to the first frequency inversion means 307.

The first frequency inversion means 307 outputs a signal obtained by inverting the frequency of the output signal from the first complex conjugate means 306 to the first multiplication means 308 and the first division means 314.

The first multiplication means 308 inputs the output signal from the first frequency multiplexing means 305 and the output signal from the first frequency inversion means 307, multiplies the signal values of these signals, and obtains the multiplication result as the first. Is output to the addition means 313.

The second complex conjugate means 310 outputs the complex conjugate of the output signal from the second frequency multiplexing means 309 to the second frequency inversion means 311.

The second frequency inversion means 311 outputs a signal obtained by inverting the frequency of the output signal from the second complex conjugate means 310 to the second multiplication means 312.

The second multiplication means 312 inputs the output signal from the second frequency multiplexing means 309 and the output signal from the second frequency inversion means 311, multiplies the signal values of these signals, and obtains the multiplication result as the first. Is output to the addition means 313.

The first adding means 313 inputs an output signal from the first multiplying means 308 and an output signal from the second multiplying means 312, adds the signal values of the signals to these, and obtains the addition result as the first. Is output to the dividing means 314 and the second dividing means 315.

The first dividing means 314 divides the signal value of the output signal from the first frequency inversion means 307 by the signal value of the output signal from the first adding means 313, and outputs the result.

The second dividing means 315 divides the signal value of the output signal from the second frequency inversion means 311 by the signal value of the output signal from the first adding means 313, and outputs the result.

Here, the signal output by the first division means 314 is used as the first coefficient, and the signal output by the second division means 315 is used as the second coefficient. The first coefficient and the second coefficient are output to the compensating means 237.

[Compensation means 237]
FIG. 5 is a configuration diagram showing the configuration of the compensation means 237 according to the embodiment.

In FIG. 5, the compensating means 237 includes a third multiplication means 401, a third complex conjugate means 402, a third frequency inversion means 403, a fourth multiplication means 404, and a second addition means 405.

The third multiplication means 401 multiplies the signal value of the output signal from the Fourier transform means 235 with the first coefficient from the first division means 314, and transfers the multiplication result to the second addition means 405. Output.

The third complex conjugate means 402 outputs the complex conjugate of the output signal from the Fourier transform means 235 to the third frequency inversion means 403.

The third frequency inversion means 403 outputs a signal obtained by inverting the frequency of the output signal from the third complex conjugate means 402 to the fourth multiplication means 404.

The fourth multiplication means 404 multiplies the signal value of the output signal from the third frequency inversion means 403 by the second coefficient from the second division means 315, and the multiplication result is added to the second. Output to means 405.

The second addition means 405 adds the output signal of the third multiplication means 401 and the output signal of the fourth multiplication means 404, and outputs the result to the inverse Fourier transform means 238.

(A-2) Operation of the Embodiment Next, the processing in the transmission side transmission device 10 and the reception side transmission device 20 according to the embodiment will be described in detail.

Here, an example is a transmission system in which the transmission side transmission device 10 is IQ-modulated and the receiving side transmission device 20 performs homodyne detection.

Further, in the following, a TS signal (for example, a first TS signal and a second TS signal) is transmitted and received between the transmission means 11 of the transmission side transmission device 10 and the reception means 23 of the reception side transmission device 20. Then, the process of compensating for the imbalance between IQs on the receiving side (that is, the receiving means 23) will be mainly described.

In the transmission side transmission device 10, the I component and the Q component of the electric signal transmitted by the transmission means 11 (that is, the electric signal output by the transmission means 11) are s _I and s _Q , respectively. s _I and s _Q have the same power. On the other hand, in the receiving side transmission device 20, the I component and the _Q component of the modulated optical signal are referred to as _EI and EQ, respectively.

G _I and g _Q are defined as quantities representing the amplitude imbalance between IQs. That is, the input / output relationship when there is no phase imbalance between IQs is _EI = g _{Is I} and _EQ = g _Q s _Q. When g _I = g _Q , it means that there is no amplitude imbalance between IQs, and when g _I ≠ g _Q , it means that there is an amplitude imbalance between IQs.

Assuming that the phase imbalance between IQs is φ, the input / output relationship when there is an imbalance in both amplitude and phase between IQs can be expressed by equations (1) and (2).

When φ = 0, it means that there is no phase imbalance between IQs. The fact that g _I = 1 and g _Q = g corresponds to the description technique of Non-Patent Document 2.

Further, when the complex electric field amplitude E is used, the E _I and _EQ of the modulated optical signal can be expressed as Eqs. (3) and (4). At this time, g _I = 1 + g and g _Q = 1-g correspond to the description technique of Non-Patent Document 3. Here, i is an imaginary unit, and ^E- (the accent symbol "-" indicates bar. The same applies to the notation of the accent symbol bar) is the complex conjugate of E.

By incorporating the influence of skew between IQs into the above-mentioned equations (1) and (2) and using the notation of time change, the input / output relationship can be expressed by equations (5) and (6).

Here, t represents time and τ _s represents skew. τ _s = 0 means that no skew occurs, τ _s > 0 means that the Q component lags behind the I component, and τ _s <0 means that the I component lags behind the Q component.

The receiving side transmission device 20 receives an optical signal from the transmitting side transmission device 10. The I component and _Q component of the received optical signal are _EI and EQ, respectively.

The optical coherent detection means 21 outputs a beat component obtained by superimposing the received optical signal and the continuous light output from the second light source 22 as a homodyne detection output signal (electric signal). Here, the I component and the Q component of the homodyne detection output signal (that is, the signal input to the receiving means 23) are referred to as r _I and r _Q , respectively.

Let GI and GQ be the quantities representing the amplitude imbalance between IQs, θ be the _phase imbalance between _IQs , and τ _r be the quantity representing the skew between IQs. In this case, the input / output relationship can be expressed by Eqs. (7) and (8).

For example, the equations (2) and (3) of Non-Patent Document 1 show the influence of the phase imbalance between IQ, and δ XI _{, YY} = 0, δ _{XQ, YQ} = θ. Then, it corresponds to the above-mentioned equations (7) and (8). Equation (6) of Non _- Patent Document ₁ shows the influence of the amplitude _imbalance between _IQs . It corresponds to the formula (7) and the formula (8). The equation (9) of Non-Patent Document 1 shows the influence of skew, and if τ XI _{and Y-I} are 0 and τ _XQ and _YQ are τr, the above-mentioned equation (7) and Consistent with equation (8).

From equations (5) to (8), and the relationship between complex numbers and real and imaginary parts s = s _I + is _Q , r = r _I + ir _Q , the relationship between transmitted electrical signals and received electrical signals is expressed in equation (9). , Can be expressed by equations (10) to (12).

In the receiving means 23, the Fourier transform means 235 Fourier transforms the output signal from the optical coherent detection means 21. That is, the Fourier transform means 235 performs a two-sided Fourier transform on both sides of the equation (9). Here, assuming that the Fourier transform pair of frequency is f and s is S and the Fourier transform pair of γ is R, it can be expressed by the equation (13). Here, A (f) in the formula (13) can be represented by the formula (14), and B (f) can be represented by the formula (15).

Here, when the equations (13) to (15) are solved for S (f), the equation (16) is obtained.

When A (f) and B (f) are known from the equation (14), the symbol inverse transform means 239 is subjected to the inverse Fourier transform by the inverse Fourier transform means 238 after executing the operation of the equation (16). , The transmitted signal can be restored from the received signal. Therefore, it is a problem to estimate A (f) and B (f) in the receiving means 23.

Therefore, in the following, a method of estimating A (f) and B (f) in the receiving means 23 will be described. The process of estimating A (f) and B (f) is executed by the coefficient calculating means 236.

It is assumed that the time waveform s (t) of the transmission signal is represented by the equation (17). The value obtained by performing a two-sided Fourier transform on both sides of the equation (17) is substituted into the equation (13). Then, the two-sided spectrum R (f) of the received signal can be expressed by the equation (18).

Here, δ is Dirac's delta function. Therefore, when the received signal spectrum is observed, A (f ₀ ) is obtained from the positive frequency component, and B (−f ₀ ) is obtained from the negative frequency component.

Further, it is assumed that the time waveform s (t) of the transmission signal is represented by the equation (19). The value obtained by performing a two-sided Fourier transform on both sides of the equation (19) is substituted into the equation (13). Then, the two-sided spectrum R (f) of the received signal can be expressed by the equation (20).

Similarly, in this case, when the received signal spectrum is observed, B (f ₀ ) is obtained from the positive frequency component, and A (−f ₀ ) is obtained from the negative frequency component.

By sequentially transmitting and receiving the signals represented by the equations (17) and (19) between the transmitting means 11 and the receiving means 23 while changing the frequency f ₀ , the response A in the required frequency band ( f) and B (f) are obtained on the receiving means 23 side.

FIG. 6 is an explanatory diagram illustrating the relationship between the first TS and the second TS transmitted by the transmitting means 11 and the third TS and the fourth TS received by the receiving means 23 in the embodiment. ..

In FIGS. 6A to 6D, the horizontal axis represents frequency and the vertical axis represents amplitude (complex amplitude). In FIGS. 6A to 6D, the right side of the frequency “0” shows the amplitude of the positive frequency component, and the left side shows the amplitude of the negative frequency component.

FIG. 6 (A) shows the first TS, and FIG. 6 (B) shows the second TS. The first TS and the second TS are signals having a two-sided spectrum, respectively.

The first TS is a single side-band (SSB) signal in which the complex amplitude of the negative frequency component is “0” and has the complex amplitude of the positive frequency component. On the other hand, the second TS is a single-sided wave (SSB) signal in which the complex amplitude of the positive frequency component is "0" and has the complex amplitude of the negative frequency component.

That is, the transmitting means 11 sets the single-sided wave signal in which the complex amplitude of the negative frequency component is "0" as the "first TS (training symbol series)" in the two-sided spectrum notation, and the complex amplitude of the positive frequency component is ". A "second TS (training symbol sequence)" is transmitted as a single-sided wave signal of "0".

When these first TS and second TS are time-division-multiplexed, on the receiving side, the third TS and the fourth TS as shown in FIGS. 6 (C) and 6 (D) are displayed from different times. can get.

Single sideband (SSB) signals can be generated, for example, by applying the Hilbert transform principle.

A (f) is obtained by frequency-multiplexing A (f') and A (-f') (f'> 0) shown in FIGS. 6 (C) and 6 (D), and FIG. 6 (C) is obtained. , B (f') and B (−f') shown in FIG. 6 (D) are frequency-multiplexed to obtain B (f).

That is, the first frequency multiplexing means 305 frequency-multiplexes A (f') and A (-f') to calculate A (f), and the second frequency multiplexing means 309 is B (f'). B (f) is calculated by frequency-multiplexing B (−f ′).

Then, the first complex conjugate means 306 obtains the complex conjugate A ⁻ (f) of A (f), and the first frequency inversion means 307 inverts the frequency of A ⁻ (f) to A ⁻ (. -F) is calculated. Further, the first multiplication means 308 multiplies A (f) and A ⁻ (−f).

On the other hand, the second complex conjugate means 310 obtains the complex conjugate B ⁻ (f) of B (f), and the second frequency inversion means 311 inverts the frequency of B ⁻ (f) to B ⁻ (. -F) is calculated. Further, the second multiplication means 312 multiplies B (f) and B ⁻ (−f).

The first adding means 313 has an output signal A (f) · A ⁻ (−f) from the first multiplying means 308 and an output signal B (f) · B ⁻ (−) from the second multiplying means 312. f) and are added.

Then, the first dividing means 314 transfers the output signal A ^{− (−f) from the first frequency inversion means 307 to the output signals A (f) · A − (} −f) from the first adding means 313. Divide by + B (f) and B- ⁽ -f) to calculate the first coefficient. This first coefficient is the coefficient of R (f) in the equation (16).

On the other hand, in the second dividing means 315, the output signal B (f) from the second frequency multiplexing means 309 is combined with the output signals A (f) and A ⁻ (−f) + B (from the first adding means 313). f) · B ^- divide by (-f) to calculate the second coefficient. This second coefficient is the coefficient of R ⁻ (f) in the equation (16).

The first coefficient and the second coefficient are output to the compensating means 237. In the compensating means 237, the Fourier transform vs. R of the signal r (r _I , r _Q ) is performed by the Fourier transform means 235 using the first coefficient and the second coefficient from the coefficient calculation means 236 according to the equation (16). Compensate for IQ-to-IQ imbalances.

That is, the third multiplication means 401 multiplies the two-sided spectrum R (f) of the received signal by the first coefficient. On the other hand, the third complex conjugate means 402 obtains the complex conjugate R ⁻ (f) of the two-sided spectrum R (f) of the received signal, and the third frequency inversion means 403 obtains the frequency of the complex conjugate R ⁻ (f). Is inverted to obtain R ⁻ (−f). Then, the fourth multiplication means 404 multiplies R ⁻ (−f) by the second coefficient.

Then, the second adding means 405 adds the output signal of the third multiplying means 401 and the output signal from the fourth multiplying means 404, and outputs S (f) of the equation (16). As described above, in the compensating means 237, the imbalance between IQs is compensated by executing the calculation of the equation (16) from the received signal using the first coefficient and the second coefficient.

(A-3) Effect of Embodiment As described above, according to the embodiment, the following effects can be obtained.

Whereas the conventional method using adaptive signal processing has a feedback structure on the receiving side, in this embodiment, the IQ-to-IQ imbalance is estimated by transmitting and receiving training sequence (TS) signals and comparing them. .. Therefore, compensation without a feedback structure on the receiving side can be realized.

Therefore, (a) instability due to divergence of the feedback process does not occur, and (b) an invalid information signal spent until the error converges does not occur, and efficient information transfer becomes possible.

Further, the influence of the IQ-to-IQ imbalance is observed as a mixture of the signal spectrum and the image signal spectrum on the receiving side. For example, as shown in FIG. 6, by transmitting and receiving TS in the SSB signal format, it becomes possible to independently estimate the distortion of the signal spectrum and the distortion of the image signal spectrum. Further, since the estimation does not have multivalued properties unlike the conventional estimation of the phase imbalance between IQs, it is possible to improve the estimation accuracy by performing the averaging process.

(B) Other Embodiments The present invention can also be applied to the following modified embodiments.

(B-1) Quantitative factors related to IQ imbalance such as amplitude imbalance, phase imbalance, and skew are effects of individual device differences, manufacturing variations, secular variation, and the surrounding environment of the device. Therefore, those time changes are sufficiently gradual compared to the symbol rate of the notification information, and are on the order of several seconds, several minutes, or several days. Therefore, it is not necessary for the TS to be sent and received frequently between the transmitting means 11 and the receiving means 23, and for example, there is a non-reporting time when introducing the device, when replacing parts, or as compared with the backbone network. In the subscriber optical communication network, the report may be sent and received at a time when the report is not sent and received.

(B-2) Since the frequency of TS transmission / reception is sparse after the first coefficient and the second coefficient are determined by the present invention at the time of introducing the device, the first coefficient and the second coefficient are obtained by the receiving means 23. The coefficient of 2 does not need to be calculated at a rate comparable to the reporting symbol. Therefore, the processing of the coefficient calculation means 236 may be realized by software.

(B-3) In order to improve the estimation accuracy of the first coefficient and the second coefficient, a plurality of first TS and second TS are generated by the transmitting means 11, and the corresponding third TS is generated by the receiving means 23. , A means for extracting the fourth TS and averaging the third TS and the fourth TS, respectively, is also included in the present invention.

100 ... Optical transmission system, 10 ... Transmitter transmission device, 11 ... Transmission means, 12 ... First light source, 13 ... Optical modulation means, 20 ... Receiver side transmission device, 21 ... Optical coherent detection means, 22 ... Second Light source, 23 ... Receiving means, 30 ... Transmission line,
111 ... Information symbol sequence generation means, 112 ... First selection means, 113 ... First TS generation means, 114 ... First enable signal generation means, 115 ... Second selection means, 116 ... Second TS generation means Means, 117 ... Second enable signal generating means,
231 ... Third enable signal generating means, 232 ... First TS extracting means, 233 ... Fourth enable signal generating means, 234 ... Second TS extracting means, 235 ... Fourier transform means, 236 ... Coefficient calculating means, 237 ... Compensation means, 238 ... Inverse Fourier transform means, 239 ... Inverse symbol transform means,
301 ... 1st Fourier conversion means, 302 ... 1st frequency separation means, 303 ... 2nd Fourier conversion means, 304 ... 2nd frequency separation means, 305 ... 1st frequency multiplexing means, 306 ... 1st Complex conjugate means, 307 ... first frequency inversion means, 308 ... first multiplication means, 309 ... second frequency multiplexing means, 310 ... second complex conjugate means, 311 ... second frequency inversion means, 312 ... second multiplication means, 313 ... first addition means, 314 ... first division means, 315 ... second division means,
401 ... third multiplication means, 402 ... third complex conjugate means, 403 ... third frequency inversion means, 404 ... fourth multiplication means, 405 ... second addition means.

Claims (3)

In a transmission device including a transmitting means for generating and transmitting an electric signal including an information signal and a receiving means for restoring an information signal from the detected electric signal.
The first electric signal pattern is a single-sided wave signal in which the amplitude of the negative frequency component is 0 in the two-sided spectrum notation, and the second electric signal pattern is a single-sided wave signal in which the amplitude of the positive frequency component is 0. And
The transmission means
A first training series generator that generates the first electric signal pattern determined by transmission and reception, and
It has a second training sequence generator that generates the second electric signal pattern different from the first electric signal pattern.
The receiving means
A coefficient derivation unit for deriving a coefficient for compensating for an imbalance between the in-phase component and the orthogonal component of the received signal based on the extracted first electric signal pattern and the second electric signal pattern.
With the compensation unit that compensates for the imbalance between the in-phase component and the orthogonal component of the received signal using the coefficient.
Have,
The transmission means time-division-multiplexes the first electric signal pattern and the second electric signal pattern and transmits the second electric signal pattern.
The coefficient derivation unit
A first frequency component separator that separates a positive frequency component and a negative frequency component of a predetermined frequency from the extracted spectrum of the first electric signal pattern,
A second frequency component separator that separates a positive frequency component and a negative frequency component of a predetermined frequency from the extracted spectrum of the second electric signal pattern,
A first frequency multiplexing section that obtains a first frequency component by frequency-multiplexing the positive frequency component from the first frequency component separating section and the negative frequency component from the second frequency component separating section. ,
A second frequency multiplexing section that obtains a second frequency component by frequency-multiplexing the negative frequency component from the first frequency component separating section and the positive frequency component from the second frequency component separating section. ,
The value of the first frequency component, the value obtained by inverting the frequency of the complex conjugate of the first frequency component, the value of the second frequency component, and the frequency of the complex conjugate of the second frequency component are inverted. With the coefficient calculation unit that derives the first coefficient and the second coefficient based on the obtained values.
Have,
The compensator adds the value of the received signal multiplied by the first coefficient and the value of the complex conjugate of the received signal multiplied by the second coefficient to obtain the received signal. Compensate for imbalance between in-phase and orthogonal components
A transmission device characterized by that. In the two-sided spectrum notation, the first electric signal pattern is a single-sided band signal having a negative frequency component amplitude of 0, and the second electric signal pattern is a single-sided wave signal having a positive frequency component amplitude of 0. In the receiving side transmission device provided with the receiving means for recovering the information signal from the electric signal obtained by detecting the received received signal from the transmitting side transmitting device that transmits the time division multiplex .
The receiving means
A coefficient derivation unit for deriving a coefficient for compensating for an imbalance between the in-phase component and the orthogonal component of the received signal based on the extracted first electric signal pattern and the second electric signal pattern.
With the compensation unit that compensates for the imbalance between the in-phase component and the orthogonal component of the received signal using the coefficient.
Have,
The coefficient derivation unit
A first frequency component separator that separates a positive frequency component and a negative frequency component of a predetermined frequency from the extracted spectrum of the first electric signal pattern,
A second frequency component separator that separates a positive frequency component and a negative frequency component of a predetermined frequency from the extracted spectrum of the second electric signal pattern,
A first frequency multiplexing section that obtains a first frequency component by frequency-multiplexing the positive frequency component from the first frequency component separating section and the negative frequency component from the second frequency component separating section. ,
A second frequency multiplexing section that obtains a second frequency component by frequency-multiplexing the negative frequency component from the first frequency component separating section and the positive frequency component from the second frequency component separating section. ,
The value of the first frequency component, the value obtained by inverting the frequency of the complex conjugate of the first frequency component, the value of the second frequency component, and the frequency of the complex conjugate of the second frequency component are inverted. With the coefficient calculation unit that derives the first coefficient and the second coefficient based on the obtained values.
Have,
The compensator adds the value of the received signal multiplied by the first coefficient and the value of the complex conjugate of the received signal multiplied by the second coefficient to obtain the received signal. Compensate for imbalance between in-phase and orthogonal components
Receiving side transmission device characterized by that. In a transmission method for transmitting information between a transmitting means for generating and transmitting an electric signal including an information signal and a receiving means for restoring an information signal from the detected electric signal.
The first electric signal pattern is a single-sided wave signal in which the amplitude of the negative frequency component is 0 in the two-sided spectrum notation, and the second electric signal pattern is a single-sided wave signal in which the amplitude of the positive frequency component is 0. And
The transmission means generates the first electric signal pattern determined by transmission / reception and the second electric signal pattern different from the first electric signal pattern, and the first electric signal pattern and the said. The second electric signal pattern and the second electric signal pattern are time-division-multiplexed and transmitted,
The receiving means
The coefficient derivation unit derives a coefficient for compensating for the imbalance between the in-phase component and the orthogonal component of the received signal based on the extracted first electric signal pattern and the second electric signal pattern.
The compensator uses the coefficient to compensate for the imbalance between the in-phase component and the orthogonal component of the received signal.
The coefficient derivation unit is
The first frequency component separation unit separates the positive frequency component and the negative frequency component of a predetermined frequency from the extracted spectrum of the first electric signal pattern.
The second frequency component separation unit separates the positive frequency component and the negative frequency component of a predetermined frequency from the extracted spectrum of the second electric signal pattern.
The first frequency multiplexing section frequency-multiplexes the positive frequency component from the first frequency component separating section and the negative frequency component from the second frequency component separating section to obtain the first frequency component. Get,
The second frequency component multiplexing unit frequency-multiplexes the negative frequency component from the first frequency component separation unit and the positive frequency component from the second frequency component separation unit to obtain a second frequency component. Get,
The coefficient calculation unit has the value of the first frequency component, the value obtained by inverting the frequency of the complex conjugate of the first frequency component, the value of the second frequency component, and the complex of the second frequency component. The first coefficient and the second coefficient are derived based on the inverted value of the conjugate frequency, and the first coefficient and the second coefficient are derived.
The compensator adds the value of the received signal multiplied by the first coefficient and the value of the complex conjugate of the received signal multiplied by the second coefficient to obtain the received signal. Compensate for imbalance between in-phase and orthogonal components
A transmission method characterized by that.

Images