A channel damage compensation method for strong lightning current based on Hodler function model

Abstract

The present invention relates to a channel damage compensation method based on a strong lightning current of a Hodler function model, comprising the following steps: step 1: pre-processing two polarization state signals in an optical fiber and inputting them into a butterfly filter to obtain a balanced demultiplexed signal; step 2: calculating an error value of the balanced demultiplexed signal obtained in step 1; step 3: a Kalman filter updates the tap coefficient of the butterfly filter according to the error value obtained in step 2; step 4: the butterfly filter uses the updated tap coefficient to realize balanced demultiplexing of the code element at the next moment, thereby compensating for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model. The present invention uses two Kalman filters to respectively realize balanced demultiplexing of two polarization states in a polarization multiplexing system, and effectively compensates for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model.

Description

Technical Field

[0001] The invention belongs to the technical field of long-distance high-speed optical communication transmission, and relates to a channel damage compensation method, in particular to a channel damage compensation method for strong lightning current based on a Hodler function model. Background Art

[0002] With the increasing number of ultra-long-distance communication scenarios in power systems, the site span generally exceeds 300km, while the general site span is about 100km. This has caused the power system and ultra-long-distance optical fiber communication to show a wide-area and complex trend, and has put forward higher requirements for the reliability and stability of ultra-long-distance optical fiber communication. This brings two challenges to the supporting optical fiber communication system: on the one hand, it is necessary to increase the relay transmission distance of the optical fiber communication system to more than 300km while ensuring the communication quality; on the other hand, the physical damage suffered by the ultra-long-distance OPGW overhead optical cable has been greatly aggravated, such as a significant increase in dispersion and more serious polarization mode dispersion, especially the rapid polarization rotation damage caused by lightning strikes. There is even a consensus in the industry: OPGW optical cables under the influence of lightning strikes will cause communication interruptions in metropolitan area networks and long-distance networks.

[0003] As a lightning current function model proposed by the International Electrotechnical Commission, the Hodler function has been widely used in the field of lightning research. According to the research of Peter M. Krummrich and others, there is a linear relationship between the polarization rotation angle in the optical fiber and the lightning current intensity. The time-varying lightning current intensity will cause the time-varying polarization rotation angle in the optical fiber and also cause time-varying RSOP damage.

[0004] As a powerful mathematical tool for solving non-stationary signal filtering problems, Kalman filter has been used by researchers in recent years to compensate for rapid polarization rotation damage in optical fibers.

[0005] The following prior art was found after searching:

[0006] Prior art 1: In 2010, T. Marshall proposed an extended Kalman filter (EKF) that can simultaneously track the polarization state and phase of a signal, and applied the Kalman filter algorithm to the field of equalization demultiplexing in optical fiber communications. Compared with the CMA / MMA algorithm, the EKF algorithm has the advantages of fast convergence speed, no singularity problem, and can compensate for rapid polarization rotation damage. Experiments have shown that it can compensate for RSOP damage with a rotation rate of 6.8 Mrad / s.

[0007] Existing technology 2: Yang Yanfu et al. proposed a radius-guided linear Kalman algorithm (RD-LKF) based on the CMA scheme. The RD-LKF balanced demultiplexing scheme can effectively compensate for fast RSOP damage and can compensate for 2.5Mrad / s polarization rotation damage under the polarization multiplexing 16QAM modulation format. In addition, RD-LKF is not affected by frequency offset and phase noise.

[0008] Prior art 3: The joint compensation Kalman scheme proposed by Zhang Xiaoguang et al. can effectively compensate for polarization rotation, polarization mode dispersion, and polarization-related impairments at the same time. The simulation results in a 28Gbaud PDM-QPSK / 16QAM coherent system show that the joint compensation scheme can cope with high-speed RSOP (up to 3Mrad / s), large PMD (over 200ps), and large RCD tolerance.

[0009] In the designs of the above Kalman equalization schemes, a single Kalman filter is used to realize the equalization demultiplexing of two polarization states in the polarization multiplexing system, which will bring high computational complexity and is not conducive to DSP implementation. In addition, the above schemes do not consider the time-varying RSOP damage under lightning conditions. Summary of the invention

[0010] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a channel damage compensation method for strong lightning current based on the Hodler function model. Two Kalman filters are used to respectively realize balanced demultiplexing of two polarization states in a polarization multiplexing system, and effectively compensate for the time-varying rapid RSOP damage caused by strong lightning current based on the Hodler function model.

[0011] The present invention solves the practical problem by adopting the following technical solutions:

[0012] A channel damage compensation method for strong lightning current based on the Hodler function model comprises the following steps:

[0013] Step 1: Pre-process the two polarization state signals in the optical fiber and input them into the butterfly filter to obtain balanced demultiplexed signals;

[0014] Step 2: Calculate the error value of the balanced demultiplexed signal obtained in step 1;

[0015] Step 3, the Kalman filter updates the tap coefficients of the butterfly filter according to the error value obtained in step 2;

[0016] Step 4: The butterfly filter uses the updated tap coefficients to achieve balanced demultiplexing of the codewords at the next moment, thereby compensating for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model.

[0017] Moreover, the specific method of step 1 is:

[0018] The two polarization state signals in the optical fiber are sent to the butterfly filter after digital signal processing, and multiplied by the tap coefficient of the butterfly filter, that is:

[0019]

[0020]

[0021] Among them, the butterfly filter center tap coefficient The initial value is set to The initial values ​​of the non-center tap coefficients are all set to 0, and X' and Y' are output signals.

[0022] Moreover, the specific method of step 2 is:

[0023] Subtract the equalized demultiplexed signals X' and Y' output from step 1 from the reference radius of the signal modulation format:

[0024]

[0025]

[0026] Among them, ε x , ε y are the error values ​​of the two signal polarization states respectively; r x 、r y are the values ​​of the nearest reference radius of the output signals X' and Y', respectively. For the normalized 16QAM signal,

[0027] Moreover, the specific method of step 3 is:

[0028] The error value ε of the polarization state of the two signals x , ε y Input two Kalman filters respectively to update the tap coefficients of the butterfly filter:

[0029] [h xx (k) h xy (k)] T =[h xx (k-1) h xy (k-1)] T +K x (k)·ε x (k) (7)

[0030] [h yx (k) h yy (k)] T =[h yx (k-1) h yy(k-1)] T +K y (k)·ε y (k) (8)

[0031] Among them, (k) represents the current moment, (k-1) represents the previous moment; K x (k), K y (k) are the Kalman gain values ​​of the two Kalman filters, and their derivation formula is:

[0032] H(k)=[E v (k) E h (k)] T (9)

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039] Where H(k) is the conversion matrix; Q is the measurement noise; R is the process noise; is the a priori estimated noise covariance; P x (k), P y (k) is the posterior estimated noise covariance.

[0040] Moreover, the specific method of step 4 is:

[0041] Updated tap coefficients The equalization demultiplexing of the code element at the next moment is returned to step 1. When the two Kalman filters converge, channel damage compensation is realized.

[0042] Advantages and beneficial effects of the present invention:

[0043] The present invention proposes a channel damage compensation method based on the Hodler function model of strong lightning current, adopts a relatively low-complexity Kalman equalization scheme, and uses two Kalman filters to respectively realize the balanced demultiplexing of two polarization states in the polarization multiplexing system, thereby effectively compensating for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model. Due to its relatively lower complexity, the Kalman equalization method is more conducive to DSP implementation and is expected to ensure the normal operation of the optical fiber communication system under lightning conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a working principle diagram of the present invention;

[0045] Figure 2 FIG. 1 is a diagram of a butterfly balanced demultiplexer of the present invention;

[0046] Figure 3 A reference graph for calculating the error value of the present invention;

[0047] Figure 4 This is a Kalman filter flow chart of the present invention. DETAILED DESCRIPTION

[0048] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings:

[0049] A channel damage compensation method for strong lightning current based on the Hodler function model, such as Figure 4 As shown, the following steps are included:

[0050] Step 1: Pre-process the two polarization state signals in the optical fiber and input them into the butterfly filter to obtain balanced demultiplexed signals;

[0051] The specific method of step 1 is:

[0052] The two polarization state signals in the optical fiber are processed by digital signal processing (IQ orthogonalization compensation, sampling offset compensation, dispersion compensation, sampling clock recovery, frequency offset compensation) to obtain the signal (E v E h ) T Send in Figure 2 The butterfly filter is multiplied by the tap coefficient of the filter, that is:

[0053]

[0054]

[0055] Among them, the butterfly filter center tap coefficient The initial value is set to The initial values ​​of the non-center tap coefficients are all set to 0, and X' and Y' are output signals.

[0056] Step 2: Calculate the error value of the balanced demultiplexed signal obtained in step 1;

[0057] The specific method of step 2 is:

[0058] Subtract the equalized demultiplexed signals X' and Y' output from step 1 from the reference radius of the signal modulation format;

[0059] In this embodiment, the polarization multiplexing 16QAM modulation format is taken as an example. Figure 3 As shown:

[0060]

[0061]

[0062] Among them, ε x , ε y are the error values ​​of the two signal polarization states respectively; r x 、r y are the values ​​of the nearest reference radius of the output signals X' and Y', respectively. For the normalized 16QAM signal,

[0063] Step 3, the Kalman filter updates the tap coefficients of the butterfly filter according to the error value obtained in step 2;

[0064] The specific method of step 3 is:

[0065] The error value ε of the polarization state of the two signals x , ε y Input two Kalman filters respectively to update the tap coefficients of the butterfly filter:

[0066] [h xx (k) h xy (k)] T =[h xx (k-1) h xy (k-1)] T +K x (k)·ε x (k) (7)

[0067] [h yx (k) h yy (k)] T =[h yx (k-1) h yy (k-1)] T +K y (k)·ε y (k) (8)

[0068] Among them, (k) represents the current moment, (k-1) represents the previous moment; K x (k), K y (k) are the Kalman gain values ​​of the two Kalman filters, and their derivation formula is:

[0069] H(k)=[E v (k) E h (k)] T (9)

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] Where H(k) is the conversion matrix; Q is the measurement noise; R is the process noise; is the a priori estimated noise covariance; P x (k), P y (k) is the posterior estimated noise covariance.

[0077] Step 4: The butterfly filter uses the updated tap coefficients to achieve balanced demultiplexing of the codewords at the next moment, thereby compensating for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model.

[0078] The specific method of step 4 is:

[0079] Updated tap coefficients The equalization demultiplexing of the code element at the next moment is returned to step 1. When the two Kalman filters converge (the error value ε x , ε y Close to 0), channel damage compensation is achieved.

[0080] Where S x (k) = [h xx (k) h xy (k)] T , S y (k) = [h yx (k) h yy (k)] T ,

[0081] By comparing the computational complexity of the technical solution (S-RD-LKF) of the present invention with that of the original solution (RD-LKF), it can be found that under the same conditions, the complexity of the technical solution (S-RD-LKF) of the present invention is only 25%-30% of that of the original solution (RD-LKF).

[0082] Real number multiplication Addition of real numbers S-RD-LKF <![CDATA[(32n 3 +48n 2 +40n+4)N]]> <![CDATA[(32n 3 +48n 2 +24n-1)N]]> RD-LKF <![CDATA[(128n 3 +192n 2 +74n+4)N]]> <![CDATA[(128n 3 +192n 2 +48n-1)N]]>

[0083] Where n represents the number of taps of the butterfly filter, and N represents the length of the codeword sequence.

[0084] The working principle of the present invention is as follows Figure 1 As shown:

[0085] At the receiving end, the two polarization state signals (X in , Y in ) After a certain amount of digital signal processing (IQ orthogonalization compensation, sampling offset compensation, dispersion compensation, sampling clock recovery, frequency offset compensation), two Kalman filters are used in the balanced demultiplexing module to achieve balanced demultiplexing of two different polarization states. The output signal (X out , Y out )The original digital signal can be obtained after phase recovery and decision decoding.

[0086] Among them, the principle explanation of balanced demultiplexing is as follows:

[0087] When only polarization rotation damage is considered, the transmission function of the optical fiber channel is a 2×2 matrix, and the relationship between the received signal and the transmitted signal is:

[0088]

[0089] Where E h and E v represents the polarization multiplexing signal at the receiving end, E x and E y represents the polarization multiplexing signal at the transmitting end, h xx 、h xy 、h yx 、h yy are the tap coefficients of the four FIR filters, h 11 、h 12 、h 21 、h 22 It constitutes the transmission function of the optical fiber channel. As shown in equation (1), due to the non-ideal characteristics of the transceiver device such as limited bandwidth and the polarization rotation of the optical fiber, the received signal is affected by polarization crosstalk, that is, the received polarization light may carry the original information of the two polarization lights at the transmitting end at the same time. This can be achieved by using a butterfly structure cross-polarization interference canceller (XPIC) composed of 4 FIR filters to achieve polarization demultiplexing (RSOP compensation), as shown in Figure 2 shown.

[0090] Then, the butterfly equalizer demultiplexer output signal should be:

[0091]

[0092] in,

[0093] This completes the balanced demultiplexing of the signal.

[0094] The explanations of the abbreviations involved in the present invention are as follows:

[0095] RSOP: rotation of the state of polarization

[0096] OPGW: optical ground wire

[0097] EKF: Extended Kalman filter

[0098] RD-LKF: radius-directed linear Kalman filter

[0099] S-RD-LKF:simplified radius-directed linear Kalman filter

[0100] CMA:constant modulus algorithm

[0101] MMA: modified constant modulus algorithm

[0102] PMD: polarization mode dispersion

[0103] RCD: residual chromatic dispersion

[0104] DSP:digital signal processing

[0105] PDM: polarization division multiplexing

[0106] QPSK: quadrature phase shift keying

[0107] 16QAM:16quadrature amplitude modulation

[0108] IQ: In-phase Quadrature-phase

[0109] It should be emphasized that the embodiments of the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation modes. Any other implementation modes derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. A channel damage compensation method for strong lightning current based on the Hodler function model, characterized by: The following steps are involved: Step 1: Pre-process the two polarization state signals in the optical fiber and input them into the butterfly filter to obtain balanced demultiplexed signals; Step 2: Calculate the error value of the balanced demultiplexed signal obtained in step 1; Step 3, the Kalman filter updates the tap coefficients of the butterfly filter according to the error value obtained in step 2; Step 4: The butterfly filter uses the updated tap coefficients to achieve balanced demultiplexing of the code element at the next moment, thereby compensating for the time-varying fast RSOP damage caused by the strong lightning current based on the Hodler function model; The specific method of step 1 is: The two polarization state signals in the optical fiber are sent to the butterfly filter after digital signal processing, and multiplied by the tap coefficient of the butterfly filter, that is: Among them, the butterfly filter center tap coefficient The initial value is set to The initial values ​​of the non-center tap coefficients are all set to 0, and X' and Y' are output signals; The specific method of step 3 is: The error value ε of the polarization state of the two signals x , ε y Input two Kalman filters respectively to update the tap coefficients of the butterfly filter: [h xx (k) h xy (k)] T =[h xx (k-1) h xy (k-1)] T +K x (k)·ε x (k) (7) [h yx (k) h yy (k)] T =[h yx (k-1) h yy (k-1)] T +K y (k)·ε y (k) (8) Among them, (k) represents the current moment, (k-1) represents the previous moment; K x (k), K y (k) are the Kalman gain values ​​of the two Kalman filters, and their derivation formula is: H(k)=[E v (k)E h (k)] T (9) Where H(k) is the conversion matrix; Q is the measurement noise; R is the process noise; is the a priori estimated noise covariance; P x (k), P y (k) is the posterior estimated noise covariance.

2. The channel damage compensation method for strong lightning current based on the Hodler function model according to claim 1 is characterized in that: The specific method of step 2 is: Subtract the equalized demultiplexed signals X' and Y' output from step 1 from the reference radius of the signal modulation format: Among them, ε x , ε y are the error values ​​of the two signal polarization states respectively; r x 、r y are the values ​​of the nearest reference radius of the output signals X' and Y', respectively. For the normalized 16QAM signal, 3. The channel damage compensation method for strong lightning current based on the Hodler function model according to claim 1 is characterized in that: The specific method of step 4 is: Updated tap coefficients The equalization demultiplexing of the code element at the next moment is returned to step 1. When the two Kalman filters converge, channel damage compensation is realized.

Images