Nft discrete spectrum optical communication system transmitter and receiver based on constellation symmetry

Abstract

The application relates to a constellation-symmetry-based NFT discrete spectrum optical communication system transmitter and receiver, wherein the transmitter and the receiver use the rotation and mirror symmetry of M-QAM corresponding constellation to simplify the types of discrete spectrum NFT signals according to the phase shift and conjugate properties of the phase distribution of the NFT signals, so that the discrete spectrum NFT signals can be stored in the form of LUT, and the complexity of the transmitter and the receiver is greatly reduced. Compared with the prior art, the application has the advantages of reducing the complexity of an arbitrary waveform generator and the storage space requirement.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical communication, and in particular to a transmitter and receiver of NFT discrete spectrum optical communication system based on constellation symmetry. BACKGROUND

[0002] NFT is a new emerging optical fiber nonlinear cancellation technology. Unlike traditional compensation algorithms, NFT utilizes Kerr nonlinearity in optical fiber for long-distance transmission of optical signals. At the same time, similar to Fourier transform (FT), NFT transforms time-domain signals to nonlinear frequency domain, so that the compensation of nonlinearity only needs to be multiplied by exp(4iλ 2 L); and due to this similarity, mature frequency division multiplexing technology and various modulation methods in linear frequency domain can also be applied to continuous and discrete nonlinear spectrum of NFT.

[0003] When only discrete spectrum is utilized, due to the rotation and symmetry of the constellation of the modulation format, the generated NFT signal is not only limited, but can also be divided into fewer categories according to the amplitude. This property can be utilized in properly designed transmitters and receivers to reduce the complexity of general-purpose DSP and AWG. Therefore, it is necessary to review examples of applying various modulation formats on nonlinear discrete spectrum. As in document one "GUI T, LU C, LAU A P T, et al. High-order modulation on a single discrete eigenvalue for optical communications based on nonlinear Fourier transform [J / OL]. Optics Express, 2017, 25(17): 20286. DOI: 10.1364 / OE.25.020286.", researchers modulated 2-ring 16-APSK, 16-QAM and 16-PSK signals with optimized radius on 1-soliton based on NFT and made performance comparison.

[0004] More cases, researchers will consider polarization multiplexing and the use of multiple eigenvalues to further increase the spectral utilization rate. As in document two“GAIARIN S, PEREGO A M, DASILVA E P, et al. Dual-polarization nonlinear Fourier transform-based optical communication system [J / OL]. Optica, 2018, 5(3): 263. DOI: 10.1364 / OPTICA.5.000263.”, researchers modulate the QPSK signal to the polarization multiplexing two NFT discrete nonlinear spectrum ({0.3j, 0.6j}), the article optimizes the amplitude and phase of QPSK to reduce the peak, and the constellation diagram is still rotated and mirror symmetric.

[0005] With the introduction of higher-order modulation and more eigenvalues, combined with polarization multiplexing, the types of signals that the transmitter needs to prepare are increasing, which also puts higher requirements on the circuit complexity and storage capacity of the AWG, thereby increasing the hardware cost. SUMMARY

[0006] The purpose of the present application is to provide a constellation symmetry-based NFT discrete spectrum optical communication system transmitter and receiver, which utilizes the symmetry characteristics of M-QAM constellation to classify and reduce the number of actual waveform signals generated by AWG, thereby reducing the hardware complexity of the transmitter.

[0007] The purpose of the present application can be achieved by the following technical solutions:

[0008] A constellation symmetry-based NFT discrete spectrum optical communication system transmitter and receiver, the transmitter comprising:

[0009] An M-QAM modulator modulates 0 / 1 sequences into M-QAM (Quadrature Amplitude Modulation) symbols, such as QPSK modulation format, 16QAM, and according to the eigenvalues of the NFT (Nonlinear Fourier Transform) signal, the corresponding symbol combination is composed, wherein the constellation diagram modulated by the M-QAM modulator satisfies the rotation and symmetry relationship;

[0010] A constellation symmetry classifier classifies the symbol combination and gives the category, rotation and symmetry information;

[0011] An INFT-DSP executes the Inverse NFT algorithm according to the symbol combination and the eigenvalue, generates NFT typical waveform data, and outputs to the LUT1 for storage;

[0012] LUT1, generating LUT1 waveform data of the signal to be transmitted according to the category, rotation and symmetry information and the stored NFT typical waveform data;

[0013] AWG, generating AWG waveform signal according to the LUT1 waveform data, and the phase shift and conjugate information of the constellation symmetry classifier;

[0014] Polarization multiplexing I / Q modulator, driven by the AWG waveform signal, modulating the fiber laser to output the NFT signal;

[0015] The receiver comprises:

[0016] Coherent receiver, coherently receiving the real and imaginary part signals of the NFT signal output;

[0017] Abs register, calculating and storing the absolute value of the received signal based on the real and imaginary part signals;

[0018] Angle register, calculating and storing the phase distribution of the received signal based on the real and imaginary part signals;

[0019] Waveform classifier, classifying according to the absolute value of the received signal, outputting the signal category and the typical signal of each category;

[0020] Phase comparator, comparing the phase distribution of the received signal and the typical signal to determine the relative phase shift and conjugate information;

[0021] NFT-DSP, calculating the eigenvalue and discrete spectrum of the typical signal according to the received NFT signal and link information, and performing transmission phase shift compensation on the discrete spectrum, and outputting the result of the NFT to LUT2 for storage;

[0022] LUT2, giving the discrete spectrum of the received signal according to the signal category information, the relative phase shift and conjugate information and the stored result of the NFT;

[0023] Decoder, decoding the discrete spectrum to output 0 / 1 signal.

[0024] The constellation symmetry classifier classifies the symbol combination as a whole, and the classification result is irrelevant, relevant, according to the irrelevant, giving the number of categories of symbol combinations and a typical symbol combination in each category and storing it in LUT1, according to the relevant, giving the rotation angle of each category of symbol combination relative to the typical symbol combination of the category and the symmetry information, wherein the symbol combination is the discrete spectrum value of the NFT signal.

[0025] The INFT-DSP generates NFT typical waveform signals and stores them into LUT1 according to preset discrete nonlinear eigenvalues and corresponding typical symbol combinations, wherein the preset multiple eigenvalues are on the upper complex plane and satisfy the symmetry about the y-axis.

[0026] The LUT1 selects and outputs typical signals according to category information, and performs phase shift and conjugate operations on the typical signals according to rotation angle and symmetry information, to generate NFT waveform data of the to-be-transmitted signals, i.e., LUT1 waveform data.

[0027] The AWG generates x / y polarization real and imaginary part electrical signals, i.e., AWG waveform signals, according to the NFT waveform data of the to-be-transmitted signals.

[0028] The polarization multiplexing I / Q modulator is driven by the x / y polarization real and imaginary part electrical signals, and outputs NFT signals according to I / Q modulation.

[0029] The coherent receiver uses a homodyne local oscillator and four balanced photodetectors to give real and imaginary part signals of the received NFT signals.

[0030] The Abs temporary register calculates the absolute value of the received signal according to the square sum of the real and imaginary part signals under the square root.

[0031] The Angle temporary register calculates the phase distribution according to the arctangent function atan(.) of the ratio of the real and imaginary parts of the received signal.

[0032] The waveform classifier classifies the signals according to the absolute values and gives the classification to which the output signals belong, and the waveform classifier also sends the amplitude and phase information of the typical signals of each category to the NFT-DSP for processing.

[0033] The phase comparator compares the phase distributions of two NFT waveforms according to the phase distributions of the received signal and the typical signal of the category to which the received signal belongs, and gives the relative phase shift size.

[0034] The NFT-DSP multiplies the discrete spectrum by exp(4iλ 2 L) to compensate for the phase shift caused by transmission, wherein L = l / Z0 is the normalized length, l is the actual transmission distance, Z0 is the nonlinear length, i is the imaginary unit, and λ is the discrete nonlinear frequency.

[0035] The LUT2 directly gives the discrete spectrum of the received signal according to the category information, the relative phase shift information, and the stored NFT-DSP result.

[0036] The decoder performs corresponding M-QAM decoding on the discrete spectrum.

[0037] Compared with the prior art, the application has the following beneficial effects:

[0038] The transmitter and receiver of the fiber communication system based on the constellation symmetry of the Nonlinear Fourier Transform (NFT) discrete spectrum of the application, according to the phase shift and conjugate properties of the phase distribution of the NFT signal, the rotation and mirror symmetry of the M-QAM (Quadrature Amplitude Modulation) corresponding constellation are used to simplify the discrete spectrum NFT signal category, and the classifier is used to classify the numerous symbol combinations in advance, so that it can be stored in the form of LUT (Lookup Table), so that the signals that really need to be generated and stored by the AWG are greatly reduced, and the complexity of the transmitter and receiver is greatly reduced, while ensuring to further improve the utilization rate of the NFT discrete spectrum, and reducing the hardware cost. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 The structure diagram of the transmitter and receiver of the fiber communication system of the application is shown in the figure.

[0040] Figure 2 The x / y polarization waveform amplitude graph classified in an embodiment is shown in the figure.

[0041] Figure 3 The waveform graph of the symbol combination satisfying the rotation and mirror symmetry of the sending end in an embodiment is shown in the figure.

[0042] Figure 4 The waveform graph of the symbol combination satisfying the rotation and mirror symmetry of the receiving end in an embodiment is shown in the figure. DETAILED DESCRIPTION

[0043] The application will be described in detail below in combination with the drawings and specific embodiments. The embodiments are implemented on the basis of the technical solutions of the application, and detailed implementation modes and specific operation processes are given, but the protection scope of the application is not limited to the following embodiments.

[0044] The embodiment provides a transmitter and receiver of an NFT discrete spectrum optical communication system based on constellation symmetry, as shown in the figure. Figure 1

[0045] 1) Transmitter

[0046] The transmitter includes the following components:

[0047] ​11) M-QAM modulator, modulating 0 / 1 sequence into M-QAM (Quadrature Amplitude Modulation) symbols, such as QPSK modulation format, 16QAM, and forming corresponding symbol combinations according to eigenvalues of the nonlinear Fourier transform (NFT) signal, the constellation diagram of which satisfies the rotation and symmetry relationship.

[0048] 12) Constellation symmetry classifier, classifying the symbol combinations (i.e. discrete frequency spectrum values of the NFT signal) as a whole, the classification result being irrelevant, relevant, according to the irrelevant, giving the number of symbol combination categories and a typical symbol combination in each category and storing them in LUT1, according to the relevant, giving the rotation angle of each symbol combination in the category relative to the typical symbol combination of the category and the symmetry information.

[0049] 13) INFT-DSP, generating NFT typical waveform signals according to the preset discrete nonlinear eigenvalues and corresponding typical symbol combinations, and storing them in LUT1; wherein, multiple eigenvalues are set in the upper complex plane and should satisfy the symmetry about the y-axis.

[0050] 14) LUT1, selecting and outputting the typical signals according to the category information, and performing phase shift and conjugate operation on the typical signals according to the rotation angle and symmetry information, to generate NFT waveform data of the to-be-transmitted signal, i.e. LUT1 (Lookup Table 1) waveform data.

[0051] 15) AWG, generating x / y polarization real and imaginary part electrical signals, i.e. AWG waveform signals, according to the LUT1 waveform data and the phase shift and conjugate information of the constellation symmetry classifier.

[0052] 16) Polarization multiplexing I / Q modulator, driven by the AWG waveform signals (i.e. x / y polarization real and imaginary part electrical signals), performing I / Q modulation on the fiber laser to output the NFT signal.

[0053] 2) Receiver

[0054] The receiver comprises the following components:

[0055] 21) Coherent receiver, receiving the NFT signal and outputting real and imaginary part signals using a homodyne local oscillator and four balanced photodetectors.

[0056] 22) Abs temporary storage, calculating and storing the absolute value of the received signal based on the square sum of the real and imaginary parts of the real and imaginary parts.

[0057] 23) Angle register, calculate and store the phase distribution of the received signal based on the arctangent function atan(.) of the ratio of the real and imaginary parts of the real and imaginary part signals.

[0058] 24) Waveform classifier, classify according to the absolute value of the received signal, output the signal category and the typical signal of each category; at the same time, the amplitude and phase information of the typical signal of each category is also sent to the NFT-DSP for processing.

[0059] 25) Phase comparator, compare the phase distribution of the two NFT waveforms according to the phase distribution of the received signal and the typical signal of the category, and give the relative phase shift size and conjugate information.

[0060] 26) NFT-DSP, calculate the eigenvalue and discrete spectrum of the typical signal according to the received NFT signal and link information, and multiply exp(4iλ 2 L) to compensate for the phase shift caused by the signal transmission distance, and output the NFT result to LUT2 for storage, where L = l / Z0 is the normalized length, l is the actual transmission distance, Z0 is the nonlinear length, i is the imaginary unit, and λ is the discrete nonlinear frequency.

[0061] 27) LUT2, directly give the discrete spectrum of the received signal according to the signal category information, the relative phase shift and the conjugate information, and the stored NFT result.

[0062] 28) Decoder, perform corresponding M-QAM decoding on the discrete spectrum, and output 0 / 1 signal.

[0063] The core of the application is to classify the M-QAM symbol combination of the discrete spectrum in NFT communication in advance, and the classification information guides the generation of the final signal. How to apply these classification information should be diverse.

[0064] After the pre-classification, the n-fold discrete spectrum classification result on M-QAM is shown in Table 1.

[0065] Table 1. Classification result

[0066]

[0067] Where N d ' and N d are the categories of polarization multiplexing including conjugate signals and not including conjugate signals, respectively. If single polarization N' and N are the categories including conjugate signals and not including conjugate signals, respectively, the following formula is satisfied

[0068] N d =N 2 , N d '=N d -(N d- (2N' - N) 2 ) / 2 = N 2 + 2N' 2 - 2NN' (1)

[0069] From Table 1, it can be seen that the category of NFT signal is reduced to about 3% of the original (6% if the conjugate is not considered), which greatly reduces the complexity and storage requirements of the AWG.

[0070] The transmitter and receiver given below are a foreseeable implementation.

[0071] The specific operation steps of the transmitter are as follows:

[0072] S1: According to the polarization multiplexing condition, generate two random information sequences {b ij}, i = 1, 2, j = 1,....

[0073] S2: Perform M-QAM modulation mapping into symbol combinations {S ij}, i = 1, 2, j = 1,.... ij = {s ijk}, k = 1,..., n, n is the number of eigenvalues, the constellation classifier obtains categories {{S ij} m}, m = 1,..., M, M is the number of categories, and rotation angles {{θ ij} m}, symmetry information {{t ij} m}, where t ij takes 1 if symmetric, otherwise 0.

[0074] S3: Take a typical symbol combination such as {S i1m} in each category and send it into the INFT-DSP to generate corresponding waveform data {q i1m}, the real and imaginary parts of which are {U i1m} and {V i1m}.

[0075] S4: In LUT1, q ijm = exp(I*(-1+2*t ijm )θ ijm )(U i1m + exp(I*π(1+2*t ijm ) / 2)V i1m ), where I is the imaginary unit, and is transmitted to the AWG to generate a corresponding electrical signal to drive the polarization multiplexing I / Q modulator.

[0076] S5: Output the NFT signal according to the polarization multiplexing I / Q modulation.

[0077] The specific operations of the receiver are as follows:

[0078] A1: Coherently receive the real and imaginary part information of each NFT signal {{U ij},{V ij}};

[0079] A2: Calculate in Abs and Angle registers, A ij =(U ij 2 +V ij 2 ) 1 / 2 , ψ ij =atan(U ij / V ij );

[0080] A3: In the waveform classifier, according to A ij Divided into M categories;

[0081] A4: According to the typical signal of each type i1m} corresponding to {U i1m ,V i1m}, get the eigenvalue and discrete spectrum in NFT-DSP, and compensate the discrete spectrum exp(4iλ 2 L) get information {S i1m}, {S i1m}Store in LUT2;

[0082] A5: Compare ψ ijm With {{A i1m},{Ac i1m}}phase distribution, the relative phase shift {φ ijm};

[0083] A6: In LUT2, S ijm =S i1m *exp(-I*φ ijm );

[0084] A7: After demodulation, the random information sequence {b i ' j}.

[0085] This embodiment can generate a variety of NFT signals. For the NFT signal with a discrete spectrum of {0.3j, 0.6j} using QPSK modulation and x / y polarization multiplexing, the types before classification are 4 4 =256. Figure 2 The classification shows that there are 16 types, 10 of which have unique amplitudes. Figure 3The waveforms corresponding to Q0=[1i;1-i](top right), Q1=[i-1;-i-1](bottom right), and Q2=[-1i;-1-i](bottom left) are shown. It can be seen that the rotation relationship between Q0 and Q1 is reflected between their waveforms and can be obtained by phase shift, and similarly, the mirror symmetry relationship between Q0 and Q2 is reflected between their waveforms and can be obtained by conjugation (or further rotation). Therefore, by preparing only one typical signal through the AWG, all amplitude distributions Figure 3 of the NFT signals on the top left can be obtained. Figure 4 The waveforms of the receiving end are shown. First, the phase shift relationship between q0 and q1 after transmission is maintained; second, it can be seen that the waveform q3 (the fourth group of diagrams below) obtained by multiplying Q0 by exp(4iλ 2 L) in the INFT is in a conjugate relationship with the transmitted waveform q2 corresponding to Q2. This property, which takes the conjugate into account at the receiving end, can also be used for further design of the receiving machine.

[0086] The present application is based on the symmetry characteristics of the constellation diagram, and classifies the symbol combinations in advance, thereby reducing the circuit complexity and storage requirements of the AWG, and has great practical significance and innovative value.

[0087] The above detailed description of the preferred embodiments of the present application. It should be understood that those skilled in the art can make many modifications and changes without creative labor according to the concept of the present application. Therefore, any technical solution obtained by logical analysis, reasoning, or limited experiment on the basis of the prior art according to the concept of the present application shall be within the scope of protection determined by the claims.

Claims

1. A constellation-symmetry-based NFT discrete-spectrum optical communication system transmitter and receiver, characterized in that, The transmitter comprises: an M-QAM modulator for modulating 0 / 1 sequences into M-QAM symbols and forming corresponding symbol combinations according to eigenvalues of the NFT signal, wherein the constellation diagram of the M-QAM modulator satisfies a rotation and symmetry relationship; a constellation diagram symmetry classifier for classifying the symbol combinations and giving category, rotation and symmetry information; an INFT-DSP for performing an Inverse NFT algorithm according to the symbol combinations and the eigenvalues, generating NFT waveform data and outputting the NFT waveform data to a LUT1 for storage; the LUT1 for generating LUT1 waveform data of the to-be-transmitted signal according to the category, rotation and symmetry information and the stored NFT waveform data; an AWG for generating an AWG waveform signal according to the LUT1 waveform data and phase shift and conjugate information of the constellation diagram symmetry classifier; a polarization multiplexing I / Q modulator driven by the AWG waveform signal for modulating an optical fiber laser to output the NFT signal; The receiver comprises: a coherent receiver for coherently receiving real and imaginary part signals of the NFT signal output; an Abs temporary storage for calculating and storing absolute values of the received signal based on the real and imaginary part signals; an Angle temporary storage for calculating and storing phase distributions of the received signal based on the real and imaginary part signals; a waveform classifier for classifying the received signal according to the absolute values of the received signal and outputting signal categories and signals of each category; a phase comparator for comparing the received signal and the phase distribution of the signal to determine relative phase shift and conjugate information; an NFT-DSP for calculating eigenvalues and discrete spectra of the signal according to the received NFT signal and link information, performing transmission phase shift compensation on the discrete spectra and outputting NFT results to a LUT2 for storage; the LUT2 for giving discrete spectra of the received signal according to signal category information, relative phase shift and conjugate information and the stored NFT results; a decoder for decoding the discrete spectra and outputting 0 / 1 signals; The constellation diagram symmetry classifier classifies the symbol combinations as a whole, and the classification result is irrelevant or relevant. According to the irrelevant, the number of categories of the symbol combinations and one symbol combination in each category are given and stored in the LUT1. According to the relevant, the rotation angle and symmetry information of each symbol combination relative to the symbol combination in the category are given. The symbol combinations are discrete spectral values of the NFT signal. The LUT1 selects output signals according to the category information, performs phase shift and conjugate operations on the signals according to the rotation angle and symmetry information, and generates NFT waveform data of the to-be-transmitted signal, i.e., LUT1 waveform data.

2. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, characterized in that, The INFT-DSP generates NFT waveform signals by using various Inverse NFT algorithms according to preset discrete nonlinear eigenvalues and corresponding symbol combinations, and stores the NFT waveform signals in the LUT1. The preset multiple eigenvalues are on the upper complex plane and satisfy symmetry about the y-axis.

3. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The AWG generates real and imaginary part electrical signals of x / y polarization, i.e., AWG waveform signals, according to the NFT waveform data of the to-be-transmitted signal.

4. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The coherent receiver uses a homodyne local oscillator and four balanced photodetectors to give real and imaginary part signals of the received NFT signal.

5. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The Abs register calculates the absolute value of the received signal according to the square root of the sum of squares of the real and imaginary parts of the signal.

6. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The Angle register calculates the phase distribution according to the arctangent function atan(.) of the ratio of the real and imaginary parts of the received signal.

7. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The Waveform Classifier sends the amplitude and phase information of the signals of each category to the NFT-DSP for processing after classifying the received signals according to the absolute values of the signals.

8. A constellation-symmetry-based NFT discrete spectral optical communication system transmitter and receiver according to claim 1, wherein, The NFT-DSP multiplies the discrete spectrum by exp(4 iλ 2 L to compensate for its phase shift due to transmission, where, L = l / Z 0 is the normalized length, l is the actual transmission distance, Z 0 is the non-linear length, i is the imaginary unit, λ is the discrete non-linear frequency.

Images