JP2017158028A - Signal transmission system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To achieve phase conjugation diversity transmission by simple transmitter constitution, and to improve SNR in a signal transmission system using a DAC for the transmission side.SOLUTION: In a signal transmission system constituted of a transmitter 1100 for transmitting a signal through a transmission line and a receiver 1200 for receiving the transmitted signal, the transmitter includes digital-analog conversion means 1121 to be operated in an effective sampling frequency fs and simultaneously sends a main signal generated in an area of fs/2 or less and an image signal appearing in an area of fs/2 or more and fs or less to the transmission line. The receiver includes: signal separation means 1211 for separating a received signal into a low frequency signal of fs/2 or less and a high frequency signal of fs/2 or more; main signal regeneration means 1221 for regenerating the main signal; image signal regeneration means 1222 for regenerating a phase conjugate signal of the image signal; and addition means 1231 for mutually adding output signals of the main signal regeneration means and the image signal regeneration means at a predetermined amplitude ratio.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a signal transmission system using a digital-analog converter.

  In recent signal transmission systems, in order to increase the capacity and functionality of the system, it has been studied to perform advanced multilevel modulation and waveform shaping using a digital signal processing circuit (Digital Signal Processor, DSP) on the transmission side. It is actively done. In introducing such a technique, a digital-to-analog converter (DAC) that converts a digital signal generated by a DSP into a high-speed analog signal to be sent to a transmission line is indispensable. is there.

  As is well known, an essential feature of a DAC is that an image signal (a folded signal, a harmonic signal) is generated at a frequency interval of the sampling rate (sampling frequency) of the DAC simultaneously with the main signal of the baseband in the output signal. including.

  As shown in Non-Patent Document 1 below, conventionally, as a transmission signal, this image signal is treated as an unnecessary signal, and in order to remove this signal, the transmission side has a half of the DAC sampling rate (Nyquist frequency). It has been common to install a low-pass filter (image removal filter) having a cutoff frequency in the vicinity.

  On the receiving side across the transmission path, the analog received signal is converted into a digital signal by an analog-digital converter (ADC) through a similar band limiting filter, and the received signal is processed.

R. Plassche, “Integrated Analog-to-Digital and Digital-to-Analog Converters,” Kluwer Academic Publishers, 1994, pp. 27-33. X. Liu et al., “Performance Improvement of Space-Division Multiplexed 128-Gb / s PDM-QPSK Signals by Constructive Superposition in a Single-Input-Multiple-Output Configuration,” in Proc. OFC / NFOEC2012, paper OTu1D.3 . T. Iida et al., “Optical diversity transmission and maximum-ratio combine in multi-core fiber to mitigate fiber non-linear distortion,” in Proc. OECC2012, paper 6B2-5. X. Liu et al., “Fiber-Nonlinearity-Tolerant Superchannel Transmission via Nonlinear Noise Squeezing and Generalized Phase-Conjugated Twin Waves,” J. Lightw. Technol. Vol. 32, pp. 766-775 (2014). H. Yamazaki, “160-Gbps Nyquist PAM4 Transmitter Using aDigital-Preprocessed Analog-Multiplexed DAC,” in Proc. ECOC2015, paper PD2.2.

  As a problem common to communication in general, there is an improvement in signal-to-noise ratio (SNR). In optical fiber communication, compensation for nonlinear phase noise generated in a transmission line is also a problem.

  As one method for solving these problems, a technique called diversity transmission has been proposed. In diversity transmission, modulated signals carrying the same information are transmitted using a plurality of transmission paths, and signal phases are aligned and added on the receiving side.

  When the noise added in each transmission line is uncorrelated with each other, the signal components are intensified in the same phase by addition, whereas the noise components are added in a random relative phase and are attenuated relative to the signal components. The For this reason, SNR can be improved compared with transmission on a single transmission path under a condition where the input signal intensity per transmission path is constant.

  Non-Patent Document 2 reports diversity optical transmission in which the same modulated signal is transmitted to a plurality of cores of a multi-core fiber. This document also describes phase offset detection / correction means for aligning signal phases between QPSK signals transmitted via different cores.

Non-Patent Document 3 analyzes the optimum value of the addition ratio of received signals in diversity transmission, and shows that the SNR after addition can be maximized by making the addition ratio equal to the SNR ratio of each transmission line. Has been. (Maximum-ratio combine, maximum ratio synthesis)
Furthermore, Non-Patent Document 4 proposes diversity transmission using a phase conjugate signal pair as means for realizing compensation for nonlinear phase distortion in an optical fiber transmission line. The phase conjugate signal pair is a signal pair composed of two signals whose phase rotation directions between signal points are opposite to each other, and in complex notation, they are in a complex conjugate relationship.

  For example, for an original signal of quaternary phase modulation (Quadrature Phase-Shift Keying, QPSK) in which data values 0, 1, 2, and 3 are associated with phases of 45 °, 135 °, 225 °, and 315 °, A QPSK signal in which 315 °, 225 °, 135 °, and 45 ° are associated as phases corresponding to the same data values 0, 1, 2, and 3 corresponds to a phase conjugate signal. Called.

  In phase conjugate diversity transmission, after transmitting the original signal and the phase conjugate signal through different transmission paths (multiple transmission paths), the receiving side reverses the phase rotation direction of one received signal (takes complex conjugate), and the other Is added to the received signal. Unlike noise signals that are uncorrelated with each other, the phase rotation direction of the nonlinear phase distortion received in the transmission path is the same between the phase conjugate signal pairs. A signal pair in which the phases are the same and the phase rotation direction due to nonlinear phase distortion is reversed can be obtained, and by adding them together, the phase rotation component due to nonlinear phase distortion can be canceled out.

  A plurality of transmission paths do not necessarily require a physically separate transmission medium. Even if a single transmission path is used, for example, two orthogonal polarization channels in an optical fiber, a plurality of frequency channels, a plurality of time slots It has been proposed to use a plurality of transmission paths.

  In phase conjugate diversity, noise signals that are uncorrelated with each other are added with a random relative phase and relatively attenuated according to the same principle as that of non-phase conjugate diversity described above. Under certain conditions, the SNR can be improved as compared with transmission on a single transmission line.

  However, in conventional diversity transmission (not phase conjugate) as shown in Non-Patent Documents 2 and 3 described above, a signal transmitted from a transmitter is branched and input to a plurality of transmission lines. As compared with the case of the transmission, there is a problem that it is necessary to prepare a high-power transmitter that outputs a signal strength that is twice the number of transmission paths.

  In the conventional phase conjugate diversity transmission as shown in Non-Patent Document 4, the phase conjugate signal pair is generated using an individual optical modulator (or an individual polarization channel of an integrated polarization multiplexing modulator). There is a problem that the number of modulators (or modulator circuit scale), the number of driver amplifiers, the number of wires in the driving electric circuit, etc. is doubled compared to transmission on a single transmission path, and the transmission apparatus becomes complicated. there were.

  The present invention has been made in view of such problems, and an object of the present invention is to realize phase conjugate diversity transmission with a simpler transmitter configuration compared to the prior art in a signal transmission system using a DAC on the transmission side. It is to provide a signal transmission system with improved SNR.

  In order to achieve such an object, in the present invention, phase conjugate diversity transmission is realized with a simpler transmitter configuration than in the past by using an image signal that has not been conventionally used as a redundant signal for the main signal. To do.

  Specifically, the configuration of the present invention is as follows.

(Structure 1 of the invention)
A signal transmission system comprising a transmitter for transmitting a signal to a receiver via a transmission line and the receiver for receiving a signal transmitted from the transmitter,
The transmitter comprises digital-to-analog conversion means operating at an effective sampling frequency f _s ;
It said digital - analog conversion means, together with the main signal appearing in a frequency f _s / 2 or less in the frequency domain, and sent to the same time the transmission path image signals appearing in the following regions frequency f _s / 2 or f _s,
The receiver is
A signal separating means for separating the received signal into a frequency f _s / 2 or less the low frequency signal and the frequency f _s / 2 or more high-frequency signal,
Main signal reproducing means for reproducing a main signal from the low frequency signal;
Image signal reproducing means for reproducing the phase conjugate signal of the image signal from the high frequency signal, and addition means for adding the output signals of the main signal reproducing means and the image signal reproducing means to each other with a predetermined amplitude ratio,
A signal transmission system characterized by that.

(Configuration 2)
The adding means has an output signal of the main signal reproducing means and the image signal reproducing means when the estimated value of the ratio of the signal band noise ratio (SNR) of the main signal to the SNR of the image signal is 1: R. The amplitude ratio is adjusted so that the amplitude ratio with respect to the output signal is larger than 1: 0 and smaller than 1: 2R / (1-R), and addition processing is performed. Signal transmission system.

(Structure 3 of the invention)
The invention is characterized in that the adding means adjusts the amplitude ratio so that the amplitude ratio of the output signal of the main signal reproducing means and the output signal of the image signal reproducing means is 1: R, and performs addition processing. The signal transmission system according to Configuration 2.

(Configuration 4)
The adding means uses an intensity ratio between the main signal and the image signal output from the signal separating means as an estimated value of a ratio between the SNR of the main signal and the SNR of the image signal. The signal transmission system according to any one of claims 1 to 3.

(Structure 5 of the invention)
The main signal is a multi-carrier signal using a plurality of frequency subchannels,
The signal transmission system according to any one of configurations 1 to 4, wherein the adding means adjusts an amplitude ratio for each frequency subchannel and performs an addition process.

(Structure 6 of the invention)
In the receiver,
The main signal reproducing means includes
First analog-to-digital conversion means for sampling the low frequency signal at a sampling frequency f _s and converting it into a digital signal;
Comprising first transmission path response compensation means for processing a digital signal from the first analog-digital conversion means,
The image signal reproducing means includes
Second analog-to-digital conversion means for sampling the high frequency signal at a sampling frequency f _s and converting it to a digital signal;
Comprising a second transmission line response compensation means for processing a digital signal from the second analog-digital conversion means;
The receiver further includes determination means for determining the output of the adding means and converting it to a digital data string.
The signal transmission system according to any one of configurations 1 to 5 of the invention.

(Configuration 7)
The receiver is
Prior to the signal separation means, the received signal is converted into a digital signal at a time with a sampling frequency sufficiently larger than twice the frequency of the image signal, while the signal information of the main signal and the image signal is preserved, and is sent to the signal separation means. Comprising analog-to-digital conversion means for sending,
The main signal reproducing means includes
Comprising first transmission line response compensation means for signal processing in the digital domain a low frequency signal including the main signal sent from the signal separation means,
The image signal reproducing means includes
Comprising a second transmission line response compensation means for performing signal processing on the high frequency signal including the image signal sent from the signal separation means in the digital domain,
The receiver further includes determination means for determining the output of the adding means and converting it to a digital data string.
The signal transmission system according to any one of configurations 1 to 5 of the invention.

(Configuration 8)
The digital-analog conversion means of the transmitter is:
First and second sub-digital-analog converting means;
Pre-digital signal processing means for processing an input digital signal and outputting it to the first and second sub-digital-analog conversion means, respectively;
Operates at a clock frequency f _c, the first and second sub-digital - to output to the transmission path the two analog signals output from the analog converter synthesized to become a analog multiplexer,
The pre-digital signal processing means is
The input digital signal is divided into a low frequency signal and a high frequency signal at a frequency f _c / 2,
After emphasizing the divided high-frequency signal, the signal is folded at a frequency f _c / 2, added with the divided low-frequency signal and zero relative phase, and output to the first sub-digital-to-analog converting means,
After emphasizing the divided high-frequency signal, the signal is folded at a frequency f _c / 2, added at the divided low-frequency signal with a relative phase π, and output to the second sub-digital-analog converting means.
The signal transmission system according to any one of configurations 1 to 7 of the invention.

  According to the present invention, in a signal transmission system using a DAC on the transmission side, it is possible to provide a signal transmission system that realizes phase conjugate diversity transmission with a simpler transmitter configuration than the conventional one.

It is a figure which shows the structure of the signal transmission system which concerns on 1st embodiment of this invention. It is a figure which shows the spectrum and constellation of the signal of each part of the signal transmission system which concerns on 1st embodiment of this invention. It is a figure which shows the spectrum and constellation of the signal of each part of the receiver of 1st embodiment of this invention. It is a figure which shows the structure of the signal transmission system which concerns on 2nd embodiment of this invention. It is a figure which shows the spectrum and constellation of the transmission signal in the signal transmission system which concerns on 3rd embodiment of this invention. It is a figure which shows the spectrum of each signal at the time of using a single carrier signal as a main signal as a comparative example. It is a figure which shows the structure of the transmitter in the signal transmission system which concerns on 4th embodiment of this invention. It is a figure which shows the spectrum of the transmission signal in the signal transmission system which concerns on 4th embodiment of this invention.

  Embodiments for carrying out the present invention will be described in detail with reference to the drawings as appropriate.

  The transmission path in the following description includes modulation means for placing a baseband signal on a carrier wave, demodulation means for returning a signal sent on the carrier wave to a baseband signal, and an amplifier used in the transmission path and for a transmission / reception terminal. Shall be.

  Specifically, in the case of optical fiber communication, a driver amplifier, an electro-optical converter (direct modulation laser or optical modulator), an optical amplifier, an optical multiplexer / demultiplexer, an optical switch, and an optical-electric converter (photodiode or coherent). Receiver), a transimpedance amplifier, and the like. For wireless communication, it is a wireless transmission path including an amplifier, a modulator / demodulator, and a space between the transmitting / receiving antenna and the antenna.

  The basic idea of the present invention is that the main signal output from the DAC and the image signal become a phase conjugate signal pair that is frequency-shifted from each other. Thus, phase conjugate diversity transmission is realized with a simpler transmitter configuration than the conventional one, while eliminating the need for an image removal filter.

(First embodiment)
FIG. 1 schematically shows a configuration of a signal transmission system 1000 according to the first embodiment of the present invention.

  The input signal 1001 to the transmitter 1100 in FIG. 1 is converted into a modulation signal by the digital signal processing means 1111 and then converted to an analog signal by the digital-analog conversion means 1121 operating at the sampling rate fs. Both image signals are sent to the transmission line.

  The transmission signal to which noise is added in the transmission path is input to the receiver 1200 as a reception signal, and is converted into a low frequency signal having a frequency fs / 2 or less and a high frequency signal having a frequency fs / 2 or more by the signal separation unit 1211. After the band division, the low frequency signal is sent to the main signal reproducing means 1221, and the high frequency signal is sent to the image signal reproducing means 1222.

  In the signal transmission system according to the present invention, the frequency band of the low frequency signal having the frequency fs / 2 or less at which the main signal is transmitted and the band of the high frequency signal having the frequency fs / 2 or more at which the image signal is transmitted are two frequencies. Diversity transmission is realized as a channel.

  The main signal regeneration unit 1221 of the receiver 1200 of FIG. 1 includes an analog-digital conversion unit 1221a and a transmission path response compensation unit 1221b that operate at a sampling rate fs.

  Similarly, the image signal reproduction unit 1222 includes an analog-digital conversion unit 1222a and a transmission path response compensation unit 1222b that operate at the sampling rate fs.

  Output signals from the main signal reproduction unit 1221 and the image signal reproduction unit 1222 are added to each other at a predetermined amplitude ratio in the addition unit 1231. The added signal is determined by the determination unit 1241 and output as an output signal 1002.

(Frequency spectrum and constellation of signals in each part of the system)
FIG. 2 is a diagram schematically showing a frequency spectrum and a constellation (constellation, signal point arrangement) of signals of each part of the system of FIG. 1 in the first embodiment of the present invention. The operation of this embodiment will be described below with reference to FIGS.

  First, the input signal 1001 (digital data string) is converted into a modulated signal to be transmitted by the digital signal processing unit 1111 of the transmitter 1100.

FIG. 2A shows the spectrum and constellation of the main signal of the input signal 1001 in this embodiment. This diagram is a conceptual diagram showing the spectrum of a signal considered as a process of the arithmetic processing by the digital signal processing unit 1111 and is not a diagram showing the spectrum of a signal as an actual electromagnetic wave. In the present embodiment, a quadrature phase-shift keying (QPSK) signal having a frequency f ₁ as a subcarrier is transmitted as a main signal. f ₁ is set so that f ₁ <f _s / 2 with respect to the sampling rate f _s of the digital-analog converter 1121.

  Note that “0” to “3” attached to each signal point of the constellation in each figure in FIG. 2 indicate data values associated with the signal points of the QPSK signal. It was written to make the relationship easier to understand. Here, in the main signal, the arrangement of the data values “0” to “3” is counterclockwise on the constellation.

  This signal point arrangement is merely an example, and an arbitrary arrangement can be used in practice. For example, an arrangement such as “0”, “1”, “3”, “2” may be used counterclockwise. In addition to the above signal mapping, the digital signal processing unit 1111 also performs processing such as error correction coding, pulse shaping, and channel pre-equalization as necessary.

  FIG. 2B shows the spectrum and constellation of the digital signal generated by the digital signal processing unit 1111 of the transmitter 1100 and input to the digital-analog conversion unit 1121. The signal in this figure is also considered as a process of arithmetic processing by the digital signal processing unit 1111 and is not a signal spectrum as an actual electromagnetic wave.

This signal is virtually discrete signal sampled at a sampling rate f _s of the main signal, the main signal in the spectrum image signal (see FIG. 2 (b) left) (see FIG. 2 (b) the right) appears. Here, for the sake of simplicity, only image signals appearing in the region below the frequency f _s are considered. This image signal is a signal obtained by folding the main signal at the frequency f _s / 2, that is, a QPSK signal having the frequency f _s −f ₁ as a subcarrier.

  Further, the QPSK modulation component of the image signal has a phase conjugate relationship with the QPSK modulation component of the main signal. That is, as shown in FIG. 2B, if the arrangement of the data values “0” to “3” associated with the signal points on the constellation is counterclockwise in the main signal, it is clockwise in the image signal. .

  FIG. 2C shows a spectrum and a constellation of a transmission signal output from the digital-analog conversion unit 1121 and sent to the transmission path. This transmission signal has a form obtained by multiplying the spectrum of the discrete signal shown in FIG. 2B by the analog frequency response characteristic of the digital-analog conversion means 1121.

For example, assuming that the digital-analog converter 1121 has an ideal 0th-order hold characteristic, the frequency response characteristic has a sinc function (sin (x) / x) shape having a null point at an integral multiple of f _s. As shown in FIG. 2C, in the transmission signal, the image signal is output with lower power than the main signal. In practice, the frequency response characteristic is generally such that the higher frequency side is attenuated than the ideal zeroth-order hold characteristic (sinc response).

  As described above, conventionally, the image signal is generally removed using an analog low-pass filter or the like, and only the main signal is transmitted. However, in the present invention, the output signal of the digital-analog conversion means 1121 is sent from the transmitter 1100 as it is to the transmission line without passing through the filter. Thus, the transmission signal is transmitted as a phase conjugate signal pair in which the main signal and the image signal are frequency-multiplexed, and phase conjugate diversity transmission can be realized.

  That is, unlike the transmitters in the conventional diversity transmissions shown in Non-Patent Documents 2 to 4, diversity transmission is realized in the present invention by utilizing the image signals that have been discarded in the past without discarding them.

  For this reason, in this invention, it is possible to simplify a transmitter structure compared with a prior art example. Further, as compared with the transmitter using the conventional DAC shown in Non-Patent Document 1, an image removal filter is not necessary, and the number of parts and excess loss can be reduced.

  FIG. 2D shows the spectrum and constellation of the received signal that reaches the receiver 1200 via the transmission path. The main signal and the image signal of the received signal are given different intensity changes, phase rotations, delays, and the like according to the frequency response characteristics of the transmission path, and noise is added.

  In many cases, the noise added in the transmission line can be regarded as white noise having no frequency dependency. Further, the noise added to the main signal and the noise added to the image signal are uncorrelated with each other.

  Therefore, the receiver 1200 separates the main signal and the image signal, individually compensates the transmission path response (intensity conversion, phase rotation, delay, etc.), reproduces each QPSK modulation component, and takes one phase conjugate. In addition, by adding together with the phases aligned, the SNR can be improved compared to the case where only the main signal is transmitted according to the principle of diversity transmission shown in Non-Patent Documents 2 to 4.

  Further, when nonlinear phase distortion occurs in a transmission line (such as an optical fiber), the compensation effect of nonlinear phase distortion can be obtained by the principle of phase conjugate diversity transmission described in Non-Patent Document 4.

  However, as described above, since the image signal is generally transmitted with weaker intensity than the main signal, if the noise added by the transmission path is white noise, the SNR of the image signal included in the signal reaching the receiver 1200 is the main signal. The value is smaller (bad) than the SNR.

  The receiver 1200 is basically configured as shown in FIG. 6 or the like, it can be configured based on the same idea as conventional phase conjugate diversity transmission using frequency (wavelength) multiplexing. That is, after the main signal and the image signal of the received signal are frequency-separated, the signals are individually reproduced, and the phases are aligned and added. It is desirable to perform signal reproduction processing and addition processing in the digital domain.

  Various methods are conceivable as a specific configuration and processing procedure of the receiver 1200, and an example thereof will be described below with reference to FIGS.

(Signal spectrum and constellation of each part of the receiver)
FIG. 3 is a diagram showing a spectrum and a constellation of signals in each part of the receiver 1200 in the first embodiment of the present invention.

FIG. 3A at the top of FIG. 3 shows the spectrum and constellation of the received signal received by the receiver 1200, which is the same as that shown in FIG. First, the received signal is converted into a low-frequency signal having a frequency f _s / 2 or less as shown in FIG. 3B on the left side of FIG. 3 and a signal on the right side of FIG. 3 (e), and is separated into a high frequency signal having a frequency f _s / 2 or higher. Specifically, the signal separation means 1211 can be easily realized by using, for example, a low-pass filter and a high-pass filter having a cutoff frequency of f _s / 2.

  As shown in FIGS. 3B and 3E, the low frequency signal includes a main signal and noise, and the high frequency signal includes an image signal and noise.

  Next, the low frequency signal and the high frequency signal are sent to the main signal reproduction means 1221 and the image signal reproduction means 1222 of the receiver 1200, respectively.

In the main signal reproducing means 1221, first, a low frequency signal (FIG. 3B) including the main signal sent from the signal separating means 1211 is sampled at the sampling rate f _s in the analog-digital conversion means 1221 a and converted into a digital signal. Is then sent to the transmission path response compensation means 1221b and processed in the digital domain.

FIG. 3C shows a spectrum and a constellation of a discrete signal obtained by sampling a low frequency signal including the main signal at a sampling rate f _s . A component folded around the Nyquist frequency f _s / 2 appears on the frequency axis, but in digital signal processing, since the signal is handled in a region below the Nyquist frequency, the digital signal sent to the transmission line response compensation means 1221b is It can be considered that the signal information of the low frequency signal is stored as it is.

  The transmission path response compensation unit 1221b of the main signal reproduction unit 1221 reproduces the QPSK modulation component of the main signal by a general algorithm used in demodulation of a normal QPSK signal. Specifically, for example, after converting the main signal into a baseband signal, noise outside the signal band is removed by a low-pass filter, and symbol timing reproduction is performed by an adaptive equalization filter driven by a Constant Modulus Algorithm (CMA) or the like. Linear distortion compensation is performed, and phase rotation is compensated using a fourth power method or the like to reproduce a QPSK signal.

FIG. 3D shows an output signal from the transmission line response compensating means 1221b of the main signal reproducing means 1221. However, in order to make it easy to obtain an intuitive understanding, the signal spectrum is drawn around the subcarrier frequency f ₁ , but in practice, as described above, it is common to perform QPSK signal reproduction in the baseband.

Similarly, in the image signal reproducing means 1222, first, a high frequency signal (FIG. 3 (e)) including the image signal sent from the signal separating means 1211 is sampled at the sampling rate f _s in the analog-digital converting means 1222a and digitally outputted. After being converted into a signal, it is sent to the transmission path response compensation means 1222b of the image signal reproduction means 1222 and processed in the digital domain.

Usually, an anti-alias filter having a cutoff frequency below the frequency f _s / 2 is provided in front of the digital-analog conversion means in order to prevent aliasing in which the high frequency component turns back at the Nyquist frequency f _s / 2 and interferes with the low frequency component. In general, an anti-aliasing filter is not used in this example, and a high-frequency signal including an image signal is converted as it is into a digital signal by the analog-digital conversion means 1222a.

However, the more high frequency noise frequency f _s, it is desirable to remove using a low-pass filter in the region of the cutoff frequency f _s. (This low-pass filter is unnecessary when high-frequency noise of f _s or more is naturally removed by the analog response characteristic of the analog-to-digital converter 1222a.)
FIG. 3F shows the spectrum and constellation of a discrete signal obtained by sampling a high frequency signal including the image signal at a sampling rate f _s . In the low frequency region below the Nyquist frequency f _s / 2, a signal obtained by folding the high frequency signal including the image signal at the Nyquist frequency appears, and the folded high frequency signal is processed in the digital region.

The image signal included in the folded high frequency signal is a QPSK signal having the same frequency f ₁ as the main signal as a subcarrier, and the QPSK modulation component is a complex conjugate of the received image signal, that is, the main signal and The QPSK signals have the same signal point arrangement (the arrangement of data values “0” to “3” is counterclockwise).

  In the transmission line response compensation means 1222b of the image signal reproduction means 1222, similarly to the transmission line response compensation means 1221b on the main signal side, the phase of the received image signal is obtained by a general algorithm used in demodulation of a normal QPSK signal. A QPSK modulation component to be conjugate is reproduced. As shown in FIG. 3G, the signal having the QPSK modulation component in the same phase rotation direction as the output of the transmission line response compensation unit 1221b on the main signal side shown in FIG. Is output.

In FIG. 3G, the signal spectrum is drawn around the subcarrier frequency f ₁ in order to easily obtain an intuitive understanding. In practice, however, QPSK signal reproduction is generally performed in baseband. Is.

  In general, there are 90 ° uncertainties in the phases of the two signals reproduced by the transmission path response compensation means 1221b and 1222b. That is, it is basically impossible to control where the signal point corresponding to the data “0” is reproduced in the four quadrants.

  As a means to solve this and to always reproduce the “0” point at a certain position (for example, the first quadrant), for example, a training signal known to the receiver side is inserted into the payload signal at a certain time interval. In other words, these methods can be used as they are in this example.

  In addition, a method for correctly demodulating data even if there is a 90 ° phase uncertainty by using a differential code is widely used. In this example, however, the transmission line response compensation means 1221b and 1222b are used. Since it is necessary to align the phases of the two output signals (matching the quadrant in which the “0” point is reproduced), when a differential code is used, an additional device is required for a general QPSK demodulation method. .

  In such a case, for example, as shown in Equation (1) of Non-Patent Document 2, a method of detecting and compensating for the phase offset by averaging the phase difference between the two signals over a certain time may be used. Alternatively, for example, a total of four copies of one of the two output signals from the transmission line response compensation means 1221b and 1222b are generated and given a phase rotation of 0, 90, 180, and 270 degrees, respectively, and then at a fixed time with the other. A method is also conceivable in which cross-correlation is taken and the one having the largest cross-correlation is adopted as the addition pair.

  The two output signals from the main signal reproducing means 1221 and the image signal reproducing means 1222 shown in FIGS. 3D and 3G are added by the adding means 1231, and the diversity shown in Non-Patent Documents 2 to 4 is added. Due to the principle of transmission, as shown in FIG. 3 (h), a QPSK signal with improved SNR is obtained. Further, when nonlinear phase distortion occurs in a transmission line (such as an optical fiber), the compensation effect of nonlinear phase distortion can be obtained by the principle of phase conjugate diversity transmission described in Non-Patent Document 4.

In FIG. 3H, the signal spectrum is drawn around the subcarrier frequency f ₁ in order to facilitate an intuitive understanding. In practice, however, QPSK signal reproduction is generally performed in baseband. It is.

  Finally, the output of the adding means 1231 in FIG. 1 (FIG. 3 (h)) is determined by the determining means 1241 and converted into a digital data string. Here, decoding of an error correction code or the like is performed as necessary.

(Setting of addition ratio)
As described with reference to FIG. 2 (c), there is a difference in transmission power between the main signal and the image signal, and attenuation in each transmission path is also different. (Non-Patent Document 3: Maximum ratio synthesis). Below, the setting of the addition ratio in the addition means 1231 will be described.

First, assuming that output signal waveforms from the main signal reproducing means 1221 and the image signal reproducing means 1222 in FIG. 1 are v _m (t) and v _i (t), respectively.

It can be expressed as.

However, s (t) is a QPSK modulation component of the main signal generated in the digital signal processing unit 1111 on the transmitter 1100 side. N _m (t) and n _i (t) are noises obtained by shifting the noise to baseband on the frequency axis in the vicinity of frequencies f ₁ and f _s −f ₁ at which the main signal and the image signal are transmitted, respectively. Are uncorrelated with each other.

The adding means 1231 adds v _m (t) and v _i (t) with an amplitude ratio of 1: α. If the output signal of the adding means 1231 is y (t),

It is.

Herein, SNR so is defined as the ratio of the average intensity of the signal component and noise, the main signal reproduction section 1221 respectively and SNR _m the SNR of the image signal reproduced by the main signal and image signal reproducing means 1222 reproduced by SNR _i , SNR of the output signal of the adding means 1231 is SNR _sum , and the average calculation is represented by E [].

It is.

However, in developing SNR _sum , since n _m (t) and n _i (t) are uncorrelated with each other, E [n _m (t) n _i (t)] = 0 is used. In general, since the SNR of the main signal is larger (good) than that of the image signal, the discussion will proceed below assuming that SNR _i <SNR _m .

When the ratio of the SNR of the reproduced image signal and the reproduced main signal, that is, SNR _i / SNR _m is R, R <1 from the above-mentioned preconditions, and SNR _sum is

It can be expressed as.

Here, if the SNR ratio between the output signal of the adding means 1231 and the main signal reproduced by the main signal reproducing means 1221 is SNR _gain ,

It becomes.

Therefore, if α = 0 or α = 2R / (1-R), SNR _gain = 1, that is, the SNR of the output from the adding means 1231 and the reproduced main signal is equal, but if α = R. Since SNR _gain = 1 + R> 1, it can be seen that the output from the adding means 1231 has a higher SNR than the reproduced main signal.

In order to confirm the setting range of α such that SNR _gain > 1, taking the first derivative with respect to α of SNR _gain ,

Therefore, it can be seen that SNR _gain increases monotonously when 0 <α <R, decreases monotonically when R <α, and takes a maximum value 1 + R when α = R.

Therefore, it is understood that the optimum value of α is R, and SNR _gain = 1 + R is obtained at this time, and SNR _gain > 1 is obtained when 0 <α <2R / (1-R).

In the above, the discussion proceeded as SNR _i <SNR _m , but when SNR _i > SNR _m , that is, when the SNR of the image signal is larger, SNR _m / SNR _i is defined as R, If SNR _sum / SNR _i is defined as SNR _gain and the addition ratio of the reproduced main signal and the reproduced image signal in the adding means 1231 is α: 1, SNR _gain > 1 from the same argument. It is easily derived that α can be set.

  In accordance with the principle described above, by appropriately setting the addition ratio in the adding means 1231, the output signal from the adding means 1231 becomes a signal with improved SNR compared to the main signal or the image signal alone.

  In order to appropriately set α in an actual signal transmission system, it is necessary to know the value of R. When the frequency response characteristics and noise characteristics of the parts constituting the digital-analog conversion means 1121, the transmission path, and the receiver 1200 are given in advance, the SNR of the main signal and the image signal is calculated from these values, and R Can be obtained.

  In addition, even when the transmission path characteristics and the like are not completely known, when the noise is white, that is, the noise intensity can be regarded as frequency-independent, and the SNR is estimated to be sufficiently large (for example, 10 dB or more), R is The intensity ratio of the received image signal and the main signal (both including noise) is approximately equal. Therefore, the intensity ratio between the main signal and the image signal included in the receiver 1200 can be simply used as an approximate value of R.

  More specifically, for example, in the transmission path response compensation means 1221b and 1222b, after dropping the main signal and the folded image signal into the baseband to remove out-of-band noise, the average intensity of each signal is calculated, and the ratio is calculated. What is necessary is just to use as an approximate value of R.

  However, if the transmission line characteristics are not given in advance and the noise intensity is frequency-dependent or the SNR is estimated to be small (for example, less than 10 dB), the signal transmission system 1000 may have the SNR of the main signal and the image signal. It is necessary to have a means to obtain the value of. This can be realized, for example, by transmitting a signal known to the receiver 1200 as a training signal from the transmitter 1100 at regular intervals and analyzing the signal at the receiver 1200.

  In this example, for the sake of simplicity, an example using QPSK modulation is shown. However, in practice, any modulation scheme can be used. For example, a modulation method such as m-level phase-shift keying (mPSK) or m-level quadrature amplitude modulation (mQAM) may be used, and the arrangement of data values for signal points In addition, not only the arrangement shown in FIG. 3 but also any arrangement can be used.

  In the description of this example, the image signal that appears in the region of the frequency fs or higher is ignored. However, if the analog frequency response band of the digital-analog conversion means 1121 is sufficiently wide and the power of the image signal appearing in the region of the frequency fs or higher cannot be ignored, the subsequent stage of the digital-analog conversion unit 1121 of the transmitter or the receiver It is desirable to cut the image signal appearing in the region of the frequency fs or higher by using a low-pass filter in the previous stage of the signal separation means 1211.

  Alternatively, an image signal appearing in a region of frequency fs or higher is also separated from the main signal and other image signals, folded back so as to overlap with the main signal, and individually reproduced using signal reproduction means, and then added to the main signal. This can also be used to improve SNR.

(Second embodiment)
FIG. 4 schematically shows the configuration of a signal transmission system 4000 according to the second embodiment of the present invention.

  The difference between the configuration of the signal transmission system 4000 of the second embodiment and the signal transmission system 1000 of the first embodiment shown in FIG. 1 is that an analog-digital conversion means 4251 in the receiver 4200 of FIG. Is separated from the main signal reproducing means 4221 and the image signal reproducing means 4222, and is integrated and installed at the front stage of the signal separating means 4211, and other configurations are basically shown in FIG. This is equivalent to the signal transmission system 1000 shown. Further, the transmitted signal is also equivalent to the transmitted signal in the first embodiment shown in FIG.

  In this example, by using a high-speed analog-digital conversion means 4251 as compared with the first embodiment, the received signal including the main signal and the image signal is converted into a digital signal at once, and then processing including signal separation is performed. All in the digital domain.

Specifically, the sampling rate f _s ′ of the analog-to-digital conversion means 4251 is made sufficiently larger than twice the subcarrier frequency f _s −f ₁ of the image signal, and the Nyquist frequency is f _s ′ / 2> f _s −. If f ₁ is used, the received signal including the main signal and the image signal as shown in FIG. 3A can be converted into a digital signal in a lump while the signal information is preserved.

However, at this time, in order to remove excessive high-frequency noise, it is desirable to use a low-pass filter having a cutoff frequency f _s ′ / 2 or less as an anti-aliasing filter before sampling. If the analog-digital conversion means 4251 satisfying such conditions is used, all signals corresponding to the reception-side processing shown in FIG. 3 can be performed in the digital domain.

  The reception side processing in this example is merely a part of the processing in the first embodiment shown in FIG. 3 (the signal separation means and the phase conjugate processing of the image signal) is shifted from the analog domain to the digital domain. In this case, the same effect as in the first embodiment can be obtained.

In the configuration of this example, since the signal separation unit 4211 is realized in the digital domain, for example, the cutoff frequency setting of the low-pass and high-pass filters used in the signal separation unit 4211 is made variable, and the sampling rate of the transmission side digital-analog conversion unit 4121 is changed. It is possible to provide flexibility such that the receiver side can cope with the change of f _s only by rewriting the software.

  However, the technical difficulty is high in that it is necessary to prepare a receiving side analog-digital conversion unit 4251 that operates at a higher sampling rate than the transmission side digital-analog conversion unit 4121.

(Third embodiment)
FIG. 5 schematically shows the spectrum and constellation of the transmission signal in the signal transmission system according to the third embodiment of the present invention. As the configuration of the signal transmission system in this example, either the configuration of the signal transmission system 1000 shown in FIG. 1 or the configuration of the signal transmission system 4000 shown in FIG. 4 may be used.

In the third embodiment, unlike the transmission signal in the first embodiment shown in FIG. 2, a multicarrier signal using subcarrier frequencies f _{1 to} f ₄ as shown in FIG. 5A as the main signal. Is used. However, it is assumed that f ₁ <f ₂ <f ₃ <f ₄ <f _s / 2, and a sub channel corresponding to the sub carrier frequency f _m (m is an integer of 1 to 4) is a sub channel m.

At this time, the image signal included in the signal sampled at the frequency f _s is also a signal obtained by folding the main signal at the frequency f _s / 2, so that the subchannel included in the main signal is shown in FIG. A phase conjugate signal of subcarrier frequency f _s −fm is included in the image signal for _m signals.

  In this example, as shown in FIG. 5C, the analog frequency response of the digital-analog conversion means has a characteristic that it is greatly attenuated at a frequency less than fs, and the image signals corresponding to the subchannels 1 and 2 are Consider a case where the material is substantially cut.

  This example is simply an extension of the first embodiment to multi-carrier, and if the same reception processing as in the first embodiment is performed on each subcarrier signal, the same principle and An effect can be obtained. However, since the image signals corresponding to the subchannels 1 and 2 are substantially cut by the digital-analog converting means, the effects of SNR improvement and nonlinear phase distortion compensation can be obtained only for the subchannels 3 and 4.

  On the receiving side, after separating the main signal and the image signal, it is necessary to further separate each subchannel signal and individually perform transmission path response compensation. However, a frequency filter in the digital domain is simply used to separate the subchannel signals. That's fine.

  In addition, when an orthogonal frequency division (OFDM) signal or a discrete multitone (DMT) signal is used as a multicarrier signal, as is well known, a discrete Fourier transform (DF) or DF (Discrete Fourier Transform) is used. Alternatively, subcarrier separation can be performed using Fast Fourier Transform (FFT), which is a fast algorithm. Since the SNRs of the main signal and the image signal are different for each subchannel, it is necessary to adjust and set the addition ratio in the adding means for each subchannel.

  By using multicarrier signals as shown in the third embodiment, phase conjugate diversity can be achieved even when the signal spectrum bandwidth of the main signal is widened compared to the case of using the single carrier signal shown in the first embodiment. There is a merit that a transmission effect can be partially obtained, and a large-capacity signal transmission can be easily performed. This will be described below.

  In general, the frequency response of the digital-analog conversion means exhibits such a characteristic that the intensity drops to almost zero at a frequency equal to or lower than the frequency fs as shown in FIGS. 2 (c) and 5 (c). For this reason, the high frequency side component of the image signal corresponding to the low frequency side component (subchannels 1 and 2 in this example) of the main signal is greatly attenuated when it is output from the digital-analog conversion means, The receiving side cannot demodulate.

  As shown in FIG. 5, when a multi-carrier signal is used, phase conjugate diversity transmission is performed on some subchannels (3, 4) of the main signal using an image signal, and other subchannels (1 , 2), normal transmission (transmission of only the main signal) can be performed. Accordingly, as described above, SNR improvement and nonlinear phase distortion compensation effect by phase conjugate diversity transmission can be obtained for some subchannels (3, 4).

  However, as a comparative example, a single carrier signal as shown in FIG. 6A, that is, a single carrier signal having the same spectral bandwidth as the multicarrier signal (including subchannels 1 to 4) in FIG. 5 is used as a main signal. If used, the high frequency side of the image signal (the component corresponding to the low frequency side of the main signal) is greatly attenuated as shown in FIGS. 6C to 6D, and the image signal can be reproduced on the receiving side. become unable. For this reason, the effect of phase conjugate diversity transmission using an image signal is not obtained at all.

  As described above, when the signal spectrum bandwidth of the main signal is widened and the capacity is increased, it is difficult to reproduce the image signal and the phase conjugate diversity transmission cannot be established if the single carrier signal is widened to some extent, but the multi-carrier signal is not established. Since signal regeneration is performed for each frequency-divided subcarrier, phase conjugate diversity transmission can be performed for some subcarriers, and there is an advantage that SNR improvement and nonlinear phase distortion compensation effects can be obtained. .

  Although the number of subchannels is 4 in this example, the number of subchannels is actually arbitrary, and a multicarrier signal using a number of subchannels (4096 channels, etc.) used in general OFDM. Can also be used.

(Fourth embodiment)
FIG. 7 shows a configuration of a transmitter 7100 in the signal transmission system according to the fourth embodiment of the present invention.

  In the fourth embodiment, the digital-analog conversion unit 7121 employs a configuration using a pre-digital signal processing unit 7121a, sub DACs 7121b and 7121c, and an analog multiplexer 7121d as shown in Non-Patent Document 5. Yes. As a configuration on the receiver side, either the configuration of the receiver 1200 illustrated in FIG. 1 or the configuration of the receiver 4200 illustrated in FIG. 4 may be used.

In this digital-analog conversion means 7121, if the operation clock frequency of the analog multiplexer is f _c , the effective sampling frequency is f _s = 2f _c .

As described in Non-Patent Document 5, the input digital signal (main signal) is first divided into a low-frequency signal and a high-frequency signal at the frequency f _c / 2 in the pre-digital signal processing means 7121a.

In the pre-digital signal processing means 7121a, the divided high-frequency signal is emphasized, folded at the frequency f _c / 2, and added to the divided low-frequency signal. At this time, a signal added with a relative phase of zero is sent to one sub DAC 7121b, and a signal added with a relative phase π is sent to the other sub DAC 7121c, and converted into analog signals.

Each sub DAC outputs an analog signal having a bandwidth of about f _c / 2. These two analog signals, that is synthesized by the analog multiplexer 7121d operating at a clock frequency f _c, the analog signal corresponding to the original input digital signal (main signal) is output to the frequency region 0 to F _c .

As a result, the analog-to-digital conversion unit 7121 as a whole can perform digital-to-analog conversion on signals that are wider than the output analog band of each of the sub-DACs 7121b and 7121c. However, at this time, as specified in Non-Patent Document 5, an image signal obtained by folding the high-frequency signal of the original input digital signal at the frequency f _c = f _s / 2 is simultaneously output from the analog multiplexer 7121d. .

That is, the digital in this example - analog conversion means 7121, the effective sampling frequency f operate in _s, of the main signal _{f s / 4 (= f c} / 2) or more frequency high-frequency components f _s / 2 ( = F _c ), the image signal folded back is output simultaneously.

  In Non-Patent Document 5, this image signal should be removed as an unnecessary component, but in this example, phase conjugate diversity transmission is performed using this image signal, as in the embodiments described so far.

  In this example, a multicarrier signal is used as the main signal. FIG. 8 schematically shows a spectrum of a transmission signal (output signal of the digital-analog conversion means 7121) used in this example. A multicarrier signal having 8 subchannels is used as the main signal.

However, in FIG. 8, subchannels 1 to 4 have a frequency f _c / 2 (= f _s / 4) or less, and sub channels 5 to 8 have a frequency f _c / 2 (= f _s / 4) or more and f _c (= f _s). / 2) Arrange in the following frequency region. If the method of Non-Patent Document 5 is applied, only the image signals of the subchannels 5 to 8 appear in the frequency region of the frequency f _c (= f _s / 2) or more and 3 f _c / 2 (= 3 f _s / 4) or less.

  For this reason, phase conjugate diversity transmission can be performed only for the subchannels 5 to 8 by the same method as in the third embodiment, and the effects of SNR improvement and nonlinear phase distortion compensation can be obtained. For subchannels 1 to 4, normal transmission using only the main signal is performed.

Also in this example, the selection of the number of subchannels is arbitrary, and a multicarrier signal using a number of subchannels as used in general OFDM can also be used. The sub in the present example is half of all the sub-channels are arranged in the region of the frequency f _c / 2~f _c, which not necessarily have to be a half, it is arranged in the region of the frequency f _c / 2~f _c The number of channels may be more or less than half of the total number of subchannels.

  As described above, according to the present invention, in a signal transmission system using a DAC on the transmission side, it is possible to provide a signal transmission system that realizes phase conjugate diversity transmission with a simpler transmitter configuration than the conventional one.

1000, 4000 Signal transmission system 1100, 4100, 7100 Transmitter 1001, 4001, 7001 Input signal 1111, 4111, 7111 Digital signal processing means 1121, 4121, 7121 Digital-analog conversion means 7121a Pre-digital signal processing means 7121b, 7121c Sub DAC
7121d Analog multiplexer 1200, 4200 Receiver 1211, 4211 Signal separating means 1221, 4221 Main signal reproducing means 1222, 4222 Image signal reproducing means 1221a, 1222a, 4251 Analog-digital converting means 1221b, 1222b, 4221b, 4222b Transmission path response compensating means 1231, 4231 Adding means 1241, 4241 Determination means 1002, 4002 Output signal

Claims (8)

A signal transmission system comprising a transmitter for transmitting a signal to a receiver via a transmission line and the receiver for receiving a signal transmitted from the transmitter,
The transmitter comprises digital-to-analog conversion means operating at an effective sampling frequency f _s ;
It said digital - analog conversion means, together with the main signal appearing in a frequency f _s / 2 or less in the frequency domain, and sent to the same time the transmission path image signals appearing in the following regions frequency f _s / 2 or f _s,
The receiver is
A signal separating means for separating the received signal into a frequency f _s / 2 or less the low frequency signal and the frequency f _s / 2 or more high-frequency signal,
Main signal reproducing means for reproducing a main signal from the low frequency signal;
Image signal reproducing means for reproducing the phase conjugate signal of the image signal from the high frequency signal, and addition means for adding the output signals of the main signal reproducing means and the image signal reproducing means to each other with a predetermined amplitude ratio,
A signal transmission system characterized by that. The adding means has an output signal of the main signal reproducing means and the image signal reproducing means when the estimated value of the ratio of the signal band noise ratio (SNR) of the main signal to the SNR of the image signal is 1: R. 2. The signal according to claim 1, wherein the amplitude ratio is adjusted such that the amplitude ratio with respect to the output signal is greater than 1: 0 and smaller than 1: 2R / (1−R), and addition processing is performed. Transmission system. The addition means adjusts an amplitude ratio so that an amplitude ratio of an output signal of the main signal reproduction means and an output signal of the image signal reproduction means is 1: R, and performs addition processing. 2. The signal transmission system according to 2.   The adding means uses an intensity ratio between the main signal and the image signal output from the signal separating means as an estimated value of a ratio between the SNR of the main signal and the SNR of the image signal. The signal transmission system according to any one of claims 1 to 3. The main signal is a multi-carrier signal using a plurality of frequency subchannels,
5. The signal transmission system according to claim 1, wherein the adding unit adjusts an amplitude ratio for each frequency subchannel and performs addition processing. 6. In the receiver,
The main signal reproducing means includes
First analog-to-digital conversion means for sampling the low frequency signal at a sampling frequency f _s and converting it into a digital signal;
Comprising first transmission path response compensation means for processing a digital signal from the first analog-digital conversion means,
The image signal reproducing means includes
Second analog-to-digital conversion means for sampling the high frequency signal at a sampling frequency f _s and converting it to a digital signal;
Comprising a second transmission line response compensation means for processing a digital signal from the second analog-digital conversion means;
The receiver further includes determination means for determining the output of the adding means and converting it to a digital data string.
The signal transmission system according to any one of claims 1 to 5, wherein The receiver is
Prior to the signal separation means, the received signal is converted into a digital signal at a time with a sampling frequency sufficiently larger than twice the frequency of the image signal, while the signal information of the main signal and the image signal is preserved, and is sent to the signal separation means. Comprising analog-to-digital conversion means for sending,
The main signal reproducing means includes
Comprising first transmission line response compensation means for signal processing in the digital domain a low frequency signal including the main signal sent from the signal separation means,
The image signal reproducing means includes
Comprising a second transmission line response compensation means for performing signal processing on the high frequency signal including the image signal sent from the signal separation means in the digital domain,
The receiver further includes determination means for determining the output of the adding means and converting it to a digital data string.
The signal transmission system according to any one of claims 1 to 5, wherein The digital-analog conversion means of the transmitter is:
First and second sub-digital-analog converting means;
Pre-digital signal processing means for processing an input digital signal and outputting it to the first and second sub-digital-analog conversion means, respectively;
Operates at a clock frequency f _c, the first and second sub-digital - to output to the transmission path the two analog signals output from the analog converter synthesized to become a analog multiplexer,
The pre-digital signal processing means is
The input digital signal is divided into a low frequency signal and a high frequency signal at a frequency f _c / 2,
After emphasizing the divided high-frequency signal, the signal is folded at a frequency f _c / 2, added with the divided low-frequency signal and zero relative phase, and output to the first sub-digital-to-analog converting means,
After emphasizing the divided high-frequency signal, the signal is folded at a frequency f _c / 2, added at the divided low-frequency signal with a relative phase π, and output to the second sub-digital-analog converting means.
The signal transmission system according to claim 1, wherein the signal transmission system is a signal transmission system.

Images