Balancing method for compensating frequency offset and channel linear damage without data assistance

Abstract

The invention discloses an equalization method for compensating frequency offset and channel linear damage without data assistance, and the method comprises the following steps: receiving an optical transmission signal and a local oscillator signal through a coherent receiver module, and carrying out the frequency mixing of the optical transmission signal and the local oscillator signal, so as to form a frequency mixing signal; converting the frequency mixing signal into an analog signal, sampling the analog signal, and converting the analog signal into a digital signal; and a digital signal processing module is used for processing the digital signal, and compensating signal frequency offset damage and linear damage such as dispersion and device bandwidth limitation. According to the equalization method for compensating frequency offset and channel linear damage without data assistance, any training sequence is not needed, spectrum resources are saved, the advantages of multiple rates and multiple modulation formats are met, and linear damage such as signal frequency offset damage, dispersion and device bandwidth limitation can be compensated at the same time.

Description

Technical Field

[0001] The present invention relates to the field of optical communications, and in particular to a data-free equalization method for compensating for frequency deviation and channel linear impairment. Background Art

[0002] With the rapid development of emerging services such as high-definition video, the internet, and cloud computing, coherent optical communications has become a key technology for the next generation of high-speed optical communication backbone networks. Coherent optical communication systems utilize high-order modulation formats to improve spectral efficiency and communication capacity, and use digital signal processing techniques to compensate for signal impairments during transmission.

[0003] Because higher-order modulation formats are more sensitive to frequency offset and phase noise, frequency offset estimation algorithms are required to compensate for signal impairments. Furthermore, the performance of high-baud-rate, high-order modulation format systems is easily affected by IQ amplitude imbalance, phase mismatch, IQ delay, and device bandwidth limitations within coherent optical transceivers. Therefore, how to compensate for these impairments using adaptive equalizers (AEQ) remains a hot topic of research. Current digital signal processing solutions for coherent optical communication systems primarily compensate for the signal's frequency offset and then use an equalizer to compensate for channel impairments. Summary of the Invention

[0004] In view of the above-mentioned shortcomings of the prior art, the technical problem to be solved by the present invention is that existing frequency offset estimation algorithms are computationally complex, have low estimation accuracy, are unsuitable for high-order modulation formats, and have poor compatibility. The present invention provides a data-free equalization method for compensating for frequency offset and channel linear impairments. This method does not require any training sequences, conserves spectrum resources, and meets the advantages of multi-rate and multi-modulation formats. It can simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations.

[0005] To achieve the above objectives, the present invention provides a data-free equalization method for compensating for frequency offset and channel linear impairment, comprising the following steps: receiving an optical transmission signal and a local oscillator signal by a coherent receiver module, and mixing the optical transmission signal and the local oscillator signal to form a mixed signal;

[0006] converting the mixed signal into an analog signal, sampling the analog signal and converting the analog signal into a digital signal;

[0007] The digital signal is processed using a digital signal processing module, while compensating for linear damage such as signal frequency deviation damage and dispersion, and device bandwidth limitation.

[0008] Furthermore, the digital signal is processed using a digital signal processing module, specifically comprising the following steps:

[0009] The matched filter module receives the digital signal output by the coherent receiver module and restores the digital signal to a baseband signal;

[0010] This baseband signal enters the clock synchronization module to eliminate clock misalignment caused by analog-to-digital conversion sampling clock mismatch;

[0011] The orthogonality and normalization module is connected to the clock synchronization module to accelerate convergence and compensate for IQ imbalance;

[0012] The blind adaptive equalization module is connected to the orthogonalization and normalization module to simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations.

[0013] The phase recovery module is connected to the blind adaptive equalization module to compensate for the random phase noise introduced by the laser linewidth.

[0014] Furthermore, the blind adaptive equalization algorithm includes the following steps:

[0015] The data after matched filtering, clock synchronization, orthogonalization and normalization is used as the input signal x(n), where n is a positive integer;

[0016] The equalizer weight coefficients are denoted as weight vector ω and weight coefficient ω f ;

[0017] Input x(n) into the blind adaptive equalizer and obtain the output data y(n) of the equalizer;

[0018] The output y(n) of the equalization module is input to the phase estimation module, which calculates the phase noise of y(n) and feeds it back to the equalizer;

[0019] The phase noise of y(n) is calculated get Combine y(n) with Multiplication represents the result of compensating the phase noise of y(n);

[0020] The symbol decision module receives the phase noise compensation result of y(n) as input and outputs

[0021] The phase noise of y(n) is calculated get Will and Multiply to get represents an ideal symbol sequence corrupted by phase noise;

[0022] The ideal symbol sequence obtained by calculation is interfered by phase noise Calculate the theoretical error value e(n) of blind adaptive equalization;

[0023] Obtain the step size factors used by blind adaptive equalization and record them as μ1 and μ2; obtain the sampling interval of the system and record it as T s ;

[0024] According to the theoretical error value e(n) of the blind adaptive equalizer, the step factors μ1 and μ2, and the sampling interval T of the system s , determine the filter weight coefficient ω′ used for the next blind adaptive equalization i and μ′ f ;

[0025] The obtained filter weight coefficient ω′ i and ω′ f As the filter weight coefficient ω i and ω f , repeat the above steps until the blind adaptive equalization process is completed.

[0026] Furthermore, the equalizer weight coefficients are recorded as weight vector ω and weight coefficient ω f Specifically, the equalizer weight coefficient is recorded as weight vector ω and weight coefficient ω f

[0027]

[0028] Where ω represents the compensation weight coefficient vector for linear impairments such as dispersion and bandwidth limitation, i is the linear damage compensation weighting coefficient of the i-th channel, i=1,2,…,N; ω f Represents the compensation coefficient for frequency offset error.

[0029] Furthermore, x(n) is input into the blind adaptive equalizer, and the output data y(n) of the equalizer is obtained as

[0030]

[0031] Where x(n) is the data after matched filtering, clock synchronization, orthogonalization, and normalization, which serves as the input signal of the blind adaptive equalizer; n is a positive integer; ω represents the compensation weighting coefficient vector for linear impairments such as dispersion and bandwidth limitation, and ω i is the linear damage compensation weighting coefficient of the i-th channel, i=1,2,…,N; ω f Represents the compensation coefficient for frequency offset error.

[0032] Furthermore, the ideal symbol sequence interfered by phase noise is obtained by calculation The theoretical error value e(n) of blind adaptive equalization is calculated; specifically, the following steps are included:

[0033]

[0034] Furthermore, according to the obtained theoretical error value e(n) of the blind adaptive equalizer, the step factors μ1 and μ2, and the sampling interval T of the system s , determine the filter weight coefficient ω′ used for the next blind adaptive equalization i and ω′ f , specifically using the following formula:

[0035]

[0036]

[0037] Technical Effects

[0038] The present invention provides a data-free equalization method for compensating for frequency offset and channel linear impairments. This method does not require a traditional frequency offset estimation algorithm, but only requires a blind adaptive equalizer module to simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations. The present invention provides a data-free equalization method for compensating for frequency offset and channel linear impairments. The method is applicable to multiple rates and modulation formats. It does not require an additional frequency offset estimation module, but only requires a blind adaptive equalizer to simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations. It also does not require any training sequences, thus conserving spectrum resources.

[0039] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic diagram of an optical transmission system according to a preferred embodiment of the present invention;

[0041] Figure 2 is a schematic diagram of a coherent light detection system according to a preferred embodiment of the present invention;

[0042] Figure 3 is a schematic diagram of a digital signal processing module according to a preferred embodiment of the present invention;

[0043] Figure 4 1 is a schematic diagram of a blind adaptive equalization method according to a preferred embodiment of the present invention;

[0044] Figure 5 is a schematic diagram of a balancing module according to a preferred embodiment of the present invention;

[0045] Figure 6is a constellation diagram of input data of a blind adaptive equalizer of an optical transmission system according to a preferred embodiment of the present invention;

[0046] Figure 7 is a constellation diagram of output data of a blind adaptive equalizer of an optical transmission system according to a preferred embodiment of the present invention;

[0047] Figure 8 It is a constellation diagram of output data of a phase recovery module of an optical transmission system according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0049] In the following description, specific details, such as certain internal procedures and techniques, are provided for purposes of illustration and not limitation to facilitate a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

[0050] like Figure 1 As shown, an equalization method for compensating frequency deviation and channel linear damage without data assistance according to an embodiment of the present invention is applicable to a coherent optical transmission system, and includes an analog signal generation module, an electro-optical conversion module, a single-mode optical fiber, a coherent receiver module, and a digital signal processing module. The analog signal generation module is used to convert the original digital signal into an analog signal, and is connected to the electro-optical conversion module via a radio frequency line. The electro-optical conversion module is used to modulate the analog signal onto an optical carrier to obtain a modulated optical signal. The optical signal is connected to the coherent receiver module via a single-mode optical fiber. The coherent receiver module converts the modulated optical signal into a digital signal again, and then performs signal compensation on the digital signal through the digital signal processing module. As shown Figure 2 As shown, the signal generated by the arbitrary waveform generator (AWG) can be a variety of high-order modulation formats, which are modulated onto an optical carrier through an IQ modulator. The signal light is transmitted in a single-mode optical fiber. The polarization state of the signal light is adjusted by manually adjusting the polarization controller so that it is consistent with the polarization state of the local oscillator light in the coherent receiver module. The coherent receiver module is used to receive the optical transmission signal and the local oscillator signal, and obtain a coupled signal based on the optical transmission signal and the local oscillator signal, and convert the coupled signal into a digital signal. The frequency value of the local oscillator signal differs from the frequency value of the optical transmission signal by a preset offset frequency value, and the preset offset frequency value is half the symbol rate value of the optical transmission signal.

[0051] like Figure 3 As shown in the figure, the received signal passes through the matched filtering module to obtain the baseband signal. Due to various factors such as the optical device manufacturing process, it is difficult for the frequency between the local oscillator (LO) and the transmitting laser (LD) to be completely aligned, introducing cumulative phase noise into the signal. The matched filtering module is connected to the clock synchronization module and the orthogonality and normalization module, followed by the blind adaptive equalization module, which simultaneously compensates for linear impairments such as signal frequency offset, dispersion, and device bandwidth limitations. The phase recovery module compensates for carrier phase noise in the equalized output signal.

[0052] A data-free equalization method for compensating for frequency offset and channel linear impairment in an embodiment of the present invention includes an equalizer module, a phase estimation module, and a symbol decision module, wherein the equalizer module updates the tap coefficient based on the least mean square algorithm; the phase estimation module calculates the phase noise of the signal and feeds it back to the equalizer module to calculate the error signal; the symbol decision module, whose input is the output of the equalizer module multiplied by the output of the phase estimator, and whose output represents an ideal symbol sequence interfered by the phase noise.

[0053] An embodiment of the present invention provides a data-free equalization method for compensating for frequency offset and channel linear impairment, comprising the following steps: receiving an optical transmission signal and a local oscillator signal by a coherent receiver module, and mixing the optical transmission signal and the local oscillator signal to form a mixed signal;

[0054] converting the mixed signal into an analog signal, sampling the analog signal and converting the analog signal into a digital signal;

[0055] The digital signal is processed using a digital signal processing module, while compensating for linear damage such as signal frequency deviation damage and dispersion, and device bandwidth limitation.

[0056] The digital signal is processed using a digital signal processing module, specifically comprising the following steps:

[0057] The matched filter module receives the digital signal output by the coherent receiver module and restores the digital signal to a baseband signal;

[0058] This baseband signal enters the clock synchronization module to eliminate clock misalignment caused by analog-to-digital conversion sampling clock mismatch;

[0059] The orthogonality and normalization module is connected to the clock synchronization module to accelerate convergence and compensate for IQ imbalance;

[0060] The blind adaptive equalization module is connected to the orthogonalization and normalization module to simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations.

[0061] The phase recovery module is connected to the blind adaptive equalization module to compensate for the random phase noise introduced by the laser linewidth.

[0062] like Figure 4 As shown, a data-free equalization method for compensating for frequency offset and channel linear impairments uses a blind adaptive equalization algorithm and specifically includes the following steps:

[0063] The input data x(n) of the adaptive equalization algorithm is an n-channel data block consisting of a signal sequence that carries information. After the tap coefficients of the blind adaptive equalizer converge, the signal sequence is equalized. x(n) is a signal that has been matched filtered, clock synchronized, orthogonalized, and normalized, and also contains frequency offset impairments, phase noise, dispersion, and other linear impairments. The equalizer output is:

[0064]

[0065] ω and ω f is the coefficient of the blind adaptive filter, where ω represents the compensation weighting coefficient for linear impairments such as dispersion and bandwidth limitation, and is initialized to [0,…,0,1+1i,0,…,0]; f Represents the compensation coefficient for frequency offset error, initialized to 0; T s is the sampling interval, which is equal to the sampling interval of the analog-to-digital conversion module of the coherent receiver module and must satisfy the Nyquist sampling theorem; N represents the order of the blind adaptive equalizer, and the order of the adaptive equalizer should be based on the principle of minimizing the error after equalization.

[0066] The output y(n) of the equalization module enters the phase estimation module. The phase estimation module calculates the phase noise of y(n) And feed it back to the equalizer to calculate the error signal e(n). Multiplication shows the result of compensating the phase noise of y(n). The symbol decision module receives As input and output and Multiply to get represents the ideal symbol sequence corrupted by phase noise. The error signal is then calculated as

[0067]

[0068] like Figure 5 The example of the present invention uses the LMS algorithm as an example to train the tap coefficient ω. f and ω, the cost function of the algorithm is:

[0069] J(n)=E[|e(n)|2 ]

[0070] E[·] represents the statistical expectation. The goal of the algorithm is to minimize the radial error using the stochastic gradient descent (SGD) method. The tap coefficient vector for the nth iteration is:

[0071]

[0072]

[0073] μ1 and μ2 are step coefficients, * represents the conjugate operation, and Im{·} represents the imaginary part operation.

[0074] Note that other optimization algorithms may also be used to update the tap coefficients. The present invention uses the LMS algorithm as an example to explain the present invention, but is not intended to limit the present invention.

[0075] The above equalization structure can effectively eliminate signal linearity impairments such as inter-symbol crosstalk caused by chromatic dispersion and device bandwidth limitations, while compensating for phase imbalance caused by the laser's additional frequency drift.

[0076] The following takes a 25 GBaud QPSK signal with an 11th-order equalization filter coefficient as an example to illustrate the technical effects of the embodiments of the present invention.

[0077] like Figure 1 The optical transmission system of this embodiment is shown, consisting of an analog signal generation module, an electro-optical conversion module, a single-mode optical fiber, a coherent receiver module, and a digital signal processing module. The analog signal generation module is used to generate an analog signal in a QPSK modulation format and is connected to the electro-optical conversion module via an RF line. The electro-optical conversion module is used to modulate the analog signal onto an optical carrier to obtain a modulated optical signal. This optical signal is then connected to the coherent receiver module via a single-mode optical fiber. The coherent receiver module obtains the optical transmission signal and a local oscillator signal. The frequency of the local oscillator signal differs from the frequency of the optical transmission signal by half the symbol rate. A coupled signal is generated based on the optical transmission signal and the local oscillator signal. The coupled signal is then converted into an analog signal. The analog signal is then sampled at a sampling frequency of 150 GS / s to obtain a digital signal. The digital signal is then signal-compensated by the digital signal processing module.

[0078] like Figure 2As shown, a coherent receiver module and optical transmission system of an embodiment of the present invention. An arbitrary waveform generator (AWG) generates a 25GBaud QPSK signal, which is modulated onto an optical carrier through an IQ modulator. The signal light is transmitted in a single-mode optical fiber. The polarization state of the signal light is adjusted by manually adjusting the polarization controller so that it is consistent with the polarization state of the local oscillator light in the coherent detection. The coherent detection module is used to receive the optical transmission signal and the local oscillator signal, and obtain a coupled signal based on the optical transmission signal and the local oscillator signal, and convert the coupled signal into a digital signal. The optical power of the local oscillator signal is 10dBm, and the optical power of the received optical transmission signal is -17dBm. The frequency value of the local oscillator signal differs from the frequency value of the optical transmission signal by a preset offset frequency value, and the preset offset frequency value is half of the symbol rate value of the optical transmission signal.

[0079] The digital signal is input into Figure 3 In the digital signal processing flow shown in the figure, the received signal passes through the matched filter module to obtain the baseband signal. Due to the influence of various factors such as the optical device manufacturing process, it is difficult for the frequency between the local oscillator light source LO and the transmitting laser LD to maintain complete consistency, which introduces cumulative phase noise into the signal. The matched filter module is connected to the clock synchronization module and the orthogonal and normalization module, and then follows Figure 4 The blind adaptive equalization module shown compensates for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations.

[0080] like Figure 4 Figure 1 illustrates a blind adaptive equalization algorithm according to an embodiment of the present invention. The equalizer input data x(n) is an n-channel data block. x(n) is a signal that has undergone matched filtering, clock synchronization, orthogonalization, and normalization. It also contains 78 MHz frequency offset impairments, phase noise, dispersion, and other linear impairments. Figure 6 The constellation diagram of the equalizer input data x(n) is shown. Because the equalizer input data x(n) also contains linear impairments such as frequency offset impairments, dispersion, and device bandwidth limitations, the constellation points rotate and diverge. Figure 7 The constellation diagram of the output data of the blind adaptive equalization algorithm is shown. Compared with the constellation diagram of the data x(n) before equalization, the output data of the blind adaptive equalization algorithm has significantly converged, and the 78MHz frequency offset is also compensated. Without the need for an additional frequency offset estimation module, a single blind adaptive equalizer can simultaneously compensate for linear impairments such as signal frequency offset, dispersion, and device bandwidth limitations. However, the output signal of the blind adaptive equalizer still contains some phase noise, which is caused by the laser linewidth and requires a phase recovery module to compensate for the carrier phase noise.

[0081] Finally, the phase recovery module performs carrier phase noise compensation on the equalized output signal. Figure 8The constellation diagram of the output signal of the phase recovery module is shown. The constellation points of the QPSK signal are clearly visible, and the bit error rate after judgment is 0.

[0082] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A data-free equalization method for compensating for frequency offset and channel linear impairment, characterized in that: The following steps are involved: The coherent receiver module receives the optical transmission signal and the local oscillator signal, and mixes the optical transmission signal and the local oscillator signal to form a mixed signal; converting the mixed signal into an analog signal, sampling the analog signal and converting the analog signal into a digital signal; The digital signal is processed using a digital signal processing module, while compensating for linear damage such as signal frequency deviation damage and dispersion, and device bandwidth limitation.

2. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 1, wherein: The digital signal is processed using a digital signal processing module, specifically comprising the following steps: The matched filter module receives the digital signal output by the coherent receiver module and restores the digital signal to a baseband signal; This baseband signal enters the clock synchronization module to eliminate clock misalignment caused by analog-to-digital conversion sampling clock mismatch; The orthogonality and normalization module is connected to the clock synchronization module to accelerate convergence and compensate for IQ imbalance; The blind adaptive equalization module is connected to the orthogonalization and normalization module to simultaneously compensate for linear impairments such as signal frequency offset impairments, dispersion, and device bandwidth limitations. The phase recovery module is connected to the blind adaptive equalization module to compensate for the random phase noise introduced by the laser linewidth.

3. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 2, wherein: The blind adaptive equalization algorithm includes the following steps: The data after matched filtering, clock synchronization, orthogonalization and normalization is used as the input signal x(n), where n is a positive integer; The equalizer weight coefficients are denoted as weight vector ω and weight coefficient ω f ; Input x(n) into the blind adaptive equalizer and obtain the output data y(y) of the equalizer; The output y(n) of the equalization module is input to the phase estimation module, which calculates the phase noise of y(n) and feeds it back to the equalizer; The phase noise of y(n) obtained by calculation get Combine y(n) with Multiplication represents the result of compensating the phase noise of y(n); The symbol decision module receives the phase noise compensation result of y(n) as input and outputs The phase noise of y(n) is calculated get Will and Multiply to get represents an ideal symbol sequence corrupted by phase noise; The ideal symbol sequence obtained by calculation is interfered by phase noise Calculate the theoretical error value e(n) of blind adaptive equalization; Obtain the step size factors used by blind adaptive equalization and record them as μ1 and μ2; obtain the sampling interval of the system and record it as T s ; According to the obtained theoretical error value e(n) of the blind adaptive equalizer, the step factors μ1 and μ2, and the sampling interval T of the system s , determine the filter weight coefficient ω′ used for the next blind adaptive equalization i and ω′ f ; The obtained filter weight coefficient ω′ i and ω′ f As the filter weight coefficient ω i and ω f , repeat the above steps until the blind adaptive equalization process is completed.

4. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 3, wherein: The equalizer weight coefficients are denoted as weight vector ω and weight coefficient ω f Specifically, the equalizer weight coefficient is recorded as weight vector ω and weight coefficient v f Where ω represents the compensation weight coefficient vector for linear impairments such as dispersion and bandwidth limitation, i is the linear damage compensation weighting coefficient of the i-th channel, i=1,2,…,N; ω f Represents the compensation coefficient for frequency offset error.

5. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 4, wherein: Input x(n) into the blind adaptive equalizer, and get the output data y(n) of the equalizer as Where x(n) is the data after matched filtering, clock synchronization, orthogonalization, and normalization, which serves as the input signal of the blind adaptive equalizer; n is a positive integer; ω represents the compensation weighting coefficient vector for linear impairments such as dispersion and bandwidth limitation, and ω i is the linear damage compensation weighting coefficient of the i-th channel, i=1,2,…,N; ω f Represents the compensation coefficient for frequency offset error.

6. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 5, wherein: The ideal symbol sequence obtained by calculation is interfered by phase noise The theoretical error value e(n) of blind adaptive equalization is calculated; specifically, the following steps are included:

7. The data-free equalization method for compensating for frequency offset and channel linear impairment according to claim 6, wherein: According to the theoretical error value e(n) of the blind adaptive equalizer, the step factors μ1 and μ2, and the sampling interval T of the system s , determine the filter weight coefficient ω′ used for the next blind adaptive equalization i and ω′ f , specifically using the following formula:

Images