A carrier recovery method for coherent optical communication based on minimum gradient algorithm
By using the minimum gradient algorithm and M-order absolute value operation in the carrier recovery method, the problem of accuracy dependence and high algorithm complexity of frequency offset and phase noise compensation in the prior art is solved, and the real-time and high carrier recovery are achieved Accuracy synchronization compensation.
Patent Information
- Application Number
- CN202210862012.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-20
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2042-07-20
AI Technical Summary
The existing carrier recovery methods have problems such as accuracy dependence and high algorithm complexity in frequency offset compensation and phase noise compensation, making it difficult to achieve synchronous compensation of real-time and high estimation accuracy.
Using a coherent optical communication carrier recovery method based on the minimum gradient algorithm, a suitable decision function is designed to achieve synchronous compensation of frequency offset and phase noise by introducing optimized adaptive estimation and M-order absolute value operations.
Synchronous compensation of frequency offset and phase noise is realized, which improves the real-time nature of frequency offset estimation and the accuracy of carrier phase estimation, and reduces the complexity of algorithm implementation.
Smart Images

Figure CN115412176B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a coherent optical communication carrier recovery method based on a minimum gradient algorithm, and belongs to the technical field of coherent optical communication. Background Art
[0002] In application scenarios where the frequency has fast time-varying characteristics, such as high-speed satellite-borne coherent optical communications, the presence of fast time-varying Doppler frequency shift, slow drift of the laser center frequency, and laser linewidth will cause the received optical signal frequency to have a large frequency offset and phase noise relative to the local oscillator laser. In order to correctly recover the transmitted data, it is first necessary to adjust the center frequency of the local oscillator laser through a rough estimate of the frequency offset to bring the frequency offset to a smaller range, and then accurately estimate and compensate for the residual frequency offset and phase noise of the coherent sampling output signal in the digital processing module (i.e., carrier recovery). The existing carrier recovery methods are mainly divided into two categories: the first category is to realize frequency offset compensation and phase noise compensation in sequence through different estimation and compensation algorithms. This type of method has two inherent problems. The compensation performance of phase noise often depends on the accuracy of frequency offset compensation. The estimation error in the frequency offset estimation process will continue to affect the signal as an additional phase error and reduce the subsequent phase noise compensation performance; in order to overcome the problems existing in the first type of carrier recovery processing, the second type of carrier recovery processing introduces a Kalman filter algorithm with excellent estimation accuracy and convergence performance. With the minimum mean square error as the goal, the frequency offset compensation and phase noise can be synchronously compensated. However, this type of method has high algorithm complexity and faces the problem of how to adaptively update the process noise covariance so that the carrier recovery processing has high estimation accuracy and fast tracking capability. Summary of the invention
[0003] The technical solution of the present invention is: to overcome the shortcomings of the prior art and provide a coherent optical communication carrier recovery method based on the minimum gradient algorithm, which can realize synchronous compensation of frequency offset compensation and phase noise, has good real-time performance and simple algorithm implementation. The present invention first introduces the optimized adaptive estimation based on the minimum gradient in the carrier recovery process to improve the real-time performance of the frequency offset estimation; secondly, for BPSK and QPSK modulation, a suitable decision function is designed to ensure the stability and accuracy of the carrier phase iterative estimation process; finally, the M-order absolute value operation is introduced to replace the M-order power operation, which greatly reduces the complexity of the algorithm implementation.
[0004] The purpose of the present invention is achieved through the following technical solutions:
[0005] The present invention discloses a coherent optical communication carrier recovery method based on a minimum gradient algorithm, comprising the following steps:
[0006] (1) Perform normalized orthogonal processing on parallel sampling signals;
[0007] (2) calculating the carrier phase estimated value of the parallel sampling signal after normalized orthogonal processing;
[0008] (3) Pre-compensating the parallel sampled signal using the carrier phase pre-estimated value to obtain a pre-compensated complex signal;
[0009] (4) performing an M-order absolute value operation on the pre-compensated complex signal;
[0010] (5) Design a decision function to update the carrier phase state using the minimum gradient algorithm based on the M-order absolute value calculation result of the pre-compensated complex signal;
[0011] (6) using the updated carrier phase state value, calculating and obtaining updated parallel signal carrier phase information;
[0012] (7) The sampled signal is compensated using the carrier phase information to obtain a carrier recovery processing output signal.
[0013] In the above-mentioned coherent optical communication carrier recovery method, the parallel sampling signal is subjected to normalized orthogonal processing. Specifically, the external signal is sampled at the kth clock moment, and the N parallel sampling signals are subjected to normalized orthogonal processing to obtain a normalized signal R k , expressed as:
[0014] R k = {r k,1 ,r k,2 ,...,r k,N}
[0015] Among them, r k,N It represents the Nth signal after normalization at the kth moment, where N is the number of parallel paths.
[0016] In the above-mentioned coherent optical communication carrier recovery method, the carrier phase pre-estimated value of the parallel sampling signal after normalized orthogonal processing is calculated, and the specific method is: according to the state [φ updated at the k-1 clock moment end,k-1 w k- ]1 T , the carrier phase estimated value of N parallel sampling signals at time k is obtained, and the calculation formula is:
[0017]
[0018] in, is the estimated value of the carrier phase of the nth signal at the kth time, n=1,2,…,N,φ end,k-1 is the carrier phase information of the Nth sampling signal at time k-1, w k-1 is the estimated value of the frequency offset at time k-1.
[0019] In the above-mentioned coherent optical communication carrier recovery method, the carrier phase pre-estimated value is used to pre-compensate the parallel sampling signal to obtain a pre-compensated complex signal, and the calculation formula is:
[0020]
[0021] Among them, Rc k is the complex signal after pre-compensation, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, is the carrier phase estimated value, {r k,1 ,r k,2 ,...,r k,N} is a parallel sampling signal.
[0022] In the above-mentioned coherent optical communication carrier recovery method, the M-order absolute value operation is performed on the pre-compensated complex signal, and the calculation formula is:
[0023]
[0024]
[0025]
[0026]
[0027] Among them, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, Respectively represent the complex signal rc k,n The real part information and imaginary part information of Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of Respectively represent the complex signal rc k,n 4 The real and imaginary information of , and |·| is the absolute value processing.
[0028] In the above-mentioned coherent optical communication carrier recovery method, the design decision function is specifically designed as follows:
[0029] Design the decision function for BPSK modulated signal, the formula is:
[0030]
[0031] Among them, J BPSK is the decision function under BPSK modulation, rc k,N is the carrier phase pre-compensation complex signal of the Nth signal at time k;
[0032] Design the decision function for QPSK modulated signal, the formula is:
[0033]
[0034] Among them, J QPSK is the decision function under QPSK modulation.
[0035] In the above-mentioned coherent optical communication carrier recovery method, the M-order absolute value operation result of the pre-compensated complex signal is used to update the carrier phase state using the minimum gradient algorithm. The specific method is:
[0036] For BPSK optical communication system, the minimum gradient method is used to calculate the update state corresponding to BPSK modulation w k , the formula is:
[0037]
[0038]
[0039] in, Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of is the derivative operation, μ is the update step size;
[0040] For QPSK optical communication system, the minimum gradient method is used to calculate the update state corresponding to QPSK modulation w k , the formula is:
[0041]
[0042] in, Respectively represent the complex signal rc k,n 4 The real part information and imaginary part information.
[0043] In the above-mentioned coherent optical communication carrier recovery method, the updated carrier phase state value is used to calculate the iteratively updated parallel signal carrier phase information {φ k,1 ,φ k,2 ,...,φ k,n}, the specific method is:
[0044]
[0045]
[0046] in, w k is the update state, φend,k It is the carrier phase information of the Nth sampling signal at time k.
[0047] In the above-mentioned coherent optical communication carrier recovery method, the carrier phase information {φ k,1 ,φ k,2 ,...,φ k,n}Compensate the parallel sampling signal to obtain the carrier recovery processing output signal. The specific method is:
[0048]
[0049] Among them, {r k,1 ,r k,2 ,...,r k,N} is the sampling signal, {φ k,1 ,φ k,2 ,...,φ k,n} is the carrier phase information, {s k,1 ,s k,2 ,...,s k,N} is the output signal of N-channel carrier recovery processing at time K.
[0050] The beneficial effects of the present invention and the prior art are:
[0051] (1) The present invention can simultaneously realize the compensation of frequency deviation and phase noise, introduce the optimal adaptive estimation based on minimum gradient, and improve the real-time performance of frequency offset estimation.
[0052] (2) Compared with the traditional FFT-based frequency difference estimation and VV phase noise estimation, the estimation accuracy of the present invention does not depend on the length of the algorithm processing data, and can simultaneously achieve high accuracy and high real-time performance of carrier phase estimation.
[0053] (3) The present invention designs a suitable decision function for BPSK and QPSK modulation systems. Compared with the carrier recovery algorithm based on Kalman filtering, the algorithm can achieve rapid convergence without the need for decision feedback operation.
[0054] (4) The present invention introduces an M-order absolute value operation to replace the M-order power operation, and provides a derived and simplified expression of the decision function gradient, which greatly simplifies the complexity of the algorithm implementation. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] Figure 1 It is a block diagram of the implementation principle of the method of the present invention;
[0056] Figure 2 This is a schematic diagram of carrier phase calculation of the present invention;
[0057] Figure 3 It is the frequency offset estimation convergence diagram of the present invention;
[0058] Figure 4 is the phase offset estimation result of the present invention;
[0059] Figure 5 This is the constellation diagram after BPSK carrier phase recovery of the present invention;
[0060] Figure 6 This is the constellation diagram after QPSK carrier phase recovery of the present invention. DETAILED DESCRIPTION
[0061] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0062] In order to make the technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings.
[0063] The present invention discloses a coherent optical communication carrier recovery method based on a minimum gradient algorithm, comprising the following steps:
[0064] Step 1: Perform normalized orthogonal processing on the parallel sampling signals; the specific method is: sample the external signal at the kth clock moment, and perform normalized orthogonal processing on the N parallel sampling signals to obtain the normalized signal R k , expressed as:
[0065] R k = {r k,1 ,r k,2 ,...,r k,N}
[0066] Among them, r k,N It represents the Nth signal after normalization at the kth moment, where N is the number of parallel paths.
[0067] Step 2: Calculate the estimated value of the carrier phase of the parallel sampling signal after normalized orthogonal processing; the specific method is: according to the state [φ updated at the k-1 clock moment end,k-1 w k-1 ] T , the carrier phase estimated value of N parallel sampling signals at time k is obtained, and the calculation formula is:
[0068]
[0069] in, is the estimated value of the carrier phase of the nth signal at the kth time, n=1,2,…,N,φ end,k-1 is the carrier phase information of the Nth sampling signal at time k-1, w k-1 is the estimated value of the frequency offset at time k-1.
[0070] Step 3: Use the carrier phase pre-estimated value to pre-compensate the parallel sampling signal to obtain a pre-compensated complex signal; the calculation formula is:
[0071]
[0072] Among them, Rc k is the complex signal after pre-compensation, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, is the carrier phase estimated value, {r k,1 ,r k,2 ,...,r k,N} is a parallel sampling signal.
[0073] Step 4: Perform an M-order absolute value operation on the pre-compensated complex signal; the calculation formula is:
[0074]
[0075]
[0076]
[0077]
[0078] Among them, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, Respectively represent the complex signal rc k,n The real part information and imaginary part information of Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of Respectively represent the complex signal rc k,n 4 The real and imaginary information of , and |·| is the absolute value processing.
[0079] Step 5: Design a decision function, and use a minimum gradient algorithm to update the carrier phase state based on the M-order absolute value calculation result of the pre-compensated complex signal;
[0080] Design the decision function. The specific method is:
[0081] Design the decision function for BPSK modulated signal, the formula is:
[0082]
[0083] Among them, J BPSK is the decision function under BPSK modulation, rc k,N is the carrier phase pre-compensation complex signal of the Nth signal at time k;
[0084] Design the decision function for QPSK modulated signal, the formula is:
[0085]
[0086] Among them, J QPSK is the decision function under QPSK modulation.
[0087] According to the M-order absolute value operation result of the complex signal after pre-compensation, the minimum gradient algorithm is used to update the carrier phase state. The specific method is:
[0088] For BPSK optical communication system, the minimum gradient method is used to calculate the update state corresponding to BPSK modulation w k , the formula is:
[0089]
[0090]
[0091] in, Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of is the derivative operation, μ is the update step size;
[0092] For QPSK optical communication system, the minimum gradient method is used to calculate the update state corresponding to QPSK modulation w k , the formula is:
[0093]
[0094]
[0095] in, Respectively represent the complex signal rc k,n 4 The real part information and imaginary part information.
[0096] Step 6: Using the updated carrier phase state value, calculate the iteratively updated parallel signal carrier phase information {φ k,1 ,φ k,2 ,...,φ k,n}, the specific method is:
[0097]
[0098]
[0099] in, wk is the update state, φ end,k It is the carrier phase information of the Nth sampling signal at time k.
[0100] Step 7: Using the carrier phase information {φ k,1 ,φ k,2 ,...,φ k,n}Compensate the parallel sampling signal to obtain the carrier recovery processing output signal. The specific method is:
[0101]
[0102] Among them, {r k,1 ,r k,2 ,...,r k,N} is the sampling signal, {φ k,1 ,φ k,2 ,...,φ k,n} is the carrier phase information, {s k,1 ,s k,2 ,...,s k,N} is the output signal of N-channel carrier recovery processing at time K.
[0103] Example 1
[0104] like Figure 1 The figure is a block diagram of the implementation principle of the present invention. In this embodiment, the modulation mode is BPSK / QPSK, the number of parallel paths is N, and the I / Q signals after normalized orthogonal processing at the kth clock moment are respectively:
[0105] dataI k ={di k,1 ,di k,2 ,...,di k,N}
[0106] data k ={dq k,1 ,dq k,2 ,...,dq k,N}
[0107] That is: r k =di k +j·dq k .
[0108] (1) The state updated according to k-1 clock moments [φ end,k-1 w k-1 ] T , we can get the carrier phase estimation of N parallel sampling at time k:
[0109]
[0110] (2) Using the carrier phase estimation result, the N parallel sampling signals at time k are compensated. The calculation method is as follows:
[0111]
[0112]
[0113] (3) Introduce M-order absolute value operation to replace M-order power operation to realize the processing of the square and fourth powers of (I+jQ). The calculation method is as follows:
[0114] yi2 k,n =23 / 16·(|di_est k,n |-|dq_est k,n |),n=1,2,...,N
[0115] yq2 k,n =(|di_est k,n +dq_est k,n |-|di_est k,n -dq_est k,n |),n=1,2,...,N
[0116] yi4 k,n =(|yi2 k,n |-|yq2 k,n |),n=1,2,...,N
[0117] yq4 k,n =(|yi2 k,n +yq2 k,n |-yi2 k,n -yq2 k,n |),n=1,2,...,N
[0118] Among them, yi2 k,n ,yq2 k,n They are The real and imaginary information of yi4 k,n ,yq4 k,n They are The real and imaginary information, y k,n =(di_est k,n +jdi_est k,n ) is the result of compensating the phase estimation of the nth data carrier at time k using the information at time k-1.
[0119] (4) Design the decision function for QPSK / BPSK modulation:
[0120]
[0121]
[0122] Under BPSK modulation, calculate the update state w k , we can get the update formula:
[0123]
[0124]
[0125] Under QPSK modulation, calculate the update status w k , we can get the update formula:
[0126]
[0127]
[0128] Among them, yQ2 k =[yq2 k,1 ,yq2 k,2 ...,yq2 k,N ],yI2 k =[yi2 k,1 ,yi2 k,2 ...,yi2 k,N ],yQ4 k =[yq4 k,1 ,yq4 k,2 ...,yq4 k,N ],yI4 k =[yi4 k,1 ,yi4 k,2 ...,yi4 k,N ]. mean(·) is the averaging operation, μ is the update step size, and .* is the vector dot product.
[0129] (5) Using updated w k , the carrier phase calculation of N parallel sampling after iterative update at time k can be obtained. The calculation method is as follows:
[0130]
[0131]
[0132] (6) Based on this, the N parallel sampling signals at time k are compensated, and the calculation method is as follows:
[0133] di_out k,n =di k,n cos(φ k,n )+dqk,n sin(φ k,n ),n=1,2,...,N
[0134] dq_out k,n =dq k,n cos(φ k,n )-di k,n sin(φ k,n ),n=1,2,...,N
[0135] Right now:
[0136] s k,n =di_out k,n +j·dq_out k,n ,n=1,2,...,N
[0137] At this point, the carrier recovery process of the k-th time sampling data is completed. The carrier phase calculation of N parallel sampling is as follows: Figure 2 As shown in the figure, the frequency offset estimation convergence diagram in the carrier recovery processing method is as follows Figure 3 As shown in FIG. 1 , it can be seen that the frequency difference estimation of the method of the present invention has the characteristics of fast convergence and high precision. Figure 4 As shown, it can be seen that the method of the present invention can realize real-time estimation of phase noise under large laser line width. Figure 5 , Figure 6 The output constellation diagrams of carrier recovery processing under BPSK and QPSK modulation are shown respectively. It can be seen that the method of the present invention can realize accurate estimation and compensation of carrier phase.
[0138] The contents not described in detail in the specification of the present invention belong to the common knowledge of the professionals in this field.
Claims
1. A method for coherent optical communication carrier recovery based on a minimum gradient algorithm, characterized in that: include: Performing normalized orthogonal processing on parallel sampling signals; Calculate the carrier phase pre-estimate value of the parallel sampling signal after normalized orthogonal processing; Using the carrier phase pre-estimated value, the parallel sampling signal is pre-compensated to obtain a pre-compensated complex signal; Perform M-order absolute value operation on the pre-compensated complex signal, and the calculation formula is: Among them, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, Respectively represent the complex signal rc k,n The real part information and imaginary part information of Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of Respectively represent the complex signal rc k,n 4 The real and imaginary information of |·| is the absolute value processing; Design a decision function, and use the minimum gradient algorithm to update the carrier phase state based on the M-order absolute value calculation result of the pre-compensated complex signal; Design the decision function for BPSK modulated signal, the formula is: Among them, J BPSK is the decision function under BPSK modulation, rc k,N is the carrier phase pre-compensation complex signal of the Nth signal at time k; Design the decision function for QPSK modulated signal, the formula is: Among them, J QPSK is the decision function under QPSK modulation; Using the updated carrier phase state value, the updated parallel signal carrier phase information is calculated; The sampled signal is compensated using the carrier phase information to obtain a carrier recovery processing output signal.
2. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 1, characterized in that: The method of performing normalized orthogonal processing on the parallel sampling signals is as follows: sampling the external signal at the kth clock moment, and performing normalized orthogonal processing on the N parallel sampling signals to obtain a normalized signal R k , expressed as: R k ={r k,1 ,r k,2 ,...,r k,N} Among them, r k,n It represents the Nth signal after normalization at the kth moment, n=1,2,…,N, where N is the number of parallel paths.
3. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 1, characterized in that: The method for calculating the carrier phase pre-estimated value of the parallel sampling signal after normalized orthogonal processing is as follows: according to the state [φ updated at the k-1 clock moment end,k-1 w k-1 ] T , the carrier phase estimated value of N parallel sampling signals at time k is obtained, and the calculation formula is: in, is the estimated value of the carrier phase of the nth signal at the kth time, n=1,2,…,N,φ end,k-1 is the carrier phase information of the Nth sampling signal at time k-1, w k-1 is the estimated value of the frequency offset at time k-1.
4. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 1, characterized in that: The carrier phase pre-estimated value is used to pre-compensate the parallel sampling signal to obtain a pre-compensated complex signal. The calculation formula is: Among them, Rc k is the complex signal after pre-compensation, rc k,N is the carrier phase pre-compensated complex signal of the Nth signal at time k, is the carrier phase estimated value, {r k,1 ,r k,2 ,...,r k,N } is a parallel sampling signal.
5. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 3, characterized in that: The M-order absolute value calculation result of the pre-compensated complex signal is used to update the carrier phase state using the minimum gradient algorithm. The specific method is: The updated carrier phase state corresponding to BPSK modulation is calculated using the minimum gradient method w k , the formula is: in, Respectively represent the complex signal rc k,n 2 The real part information and imaginary part information of is the derivative operation, μ is the update step size; Or: Use the minimum gradient method to calculate the updated carrier phase state corresponding to QPSK modulation w k , the formula is: in, Respectively represent the complex signal rc k,n 4 The real part information and imaginary part information.
6. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 3, characterized in that: The updated carrier phase state value is used to calculate the iteratively updated parallel signal carrier phase information {φ k,1 ,φ k,2 ,...,φ k,n }, the specific method is: in, w k is the update state, φ end,k is the carrier phase information of the Nth sampling signal at time k.
7. The method for coherent optical communication carrier recovery based on the minimum gradient algorithm according to claim 1, characterized in that: The method of compensating the parallel sampling signal by using the carrier phase information to obtain the carrier recovery processing output signal is as follows: Among them, {r k,1 ,r k,2 ,...,r k,N } is the sampling signal, {φ k,1 ,φ k,2 ,...,φ k,n } is the carrier phase information, {s k,1 ,s k,2 ,...,s k,N } is the output signal of N-channel carrier recovery processing at time K.
Citation Information
Patent Citations
Frequency offset compensation method and system
CN106411494A
Carrier phase recovery method and device based on Kalman filtering algorithm
CN106936513A
