A single pilot based polarization crosstalk elimination method and DSCM system

Abstract

The present application relates to the technical field of optical signal processing, and more particularly to a single pilot-based polarization de-cross-talk method and a DSCM system. Polarization de-cross-talk is achieved by using a single pilot, so that the equalizer used for each sub-carrier equalization is simplified to two polarization-independent multi-tap SISO equalizers, which significantly reduces the computational complexity of the receiver equalizer. At the same time, the guard band between adjacent sub-carriers is eliminated, and the spectral efficiency of the signal is improved. The present application considers the influence of the XY time delay at the receiving end on the pilot-based low-complexity polarization de-cross-talk scheme, and through the estimation and calibration of the XY time delay at the receiving end, the effectiveness of the pilot-based polarization de-cross-talk method is further ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical signal processing, and more particularly to a single pilot-based polarization deinterleaving method and a DSCM system. BACKGROUND

[0002] To meet the increasing demand for data traffic, coherent technology will sink into data center optical interconnection application scenarios in the future to support large-capacity data transmission. However, due to the high complexity of the digital signal processing (DSP) algorithm of the traditional coherent optical communication system, it cannot be directly applied to short-distance data center optical interconnection application scenarios. Therefore, the simplification of the coherent optical communication algorithm is crucial. Compared with the traditional single-carrier coherent system, the signal based on digital subcarrier multiplexing modulation (DSCM) has better robustness to fiber dispersion and can effectively reduce the complexity of dispersion compensation, and is considered as a very promising solution. However, in addition to dispersion compensation, the complexity of the equalizer is also a major factor leading to the high complexity of the traditional coherent DSP. In the traditional coherent DSP algorithm, a multi-tap 2x2 multiple-input multiple-output (MIMO) equalizer is usually used to realize polarization deinterleaving and channel equalization, and the butterfly-shaped MIMO operation also brings the problem of high complexity.

[0003] In order to reduce the complexity of the equalizer, some solutions have been proposed. Since the influence of polarization mode dispersion (PMD) can be ignored in short-distance application scenarios, the existing solutions propose to split the multi-tap MIMO equalizer into a single-tap MIMO equalizer and two polarization-independent multi-tap single-input single-output (SISO) equalizers, and to realize polarization deinterleaving and channel equalization through single-tap MIMO and multi-tap SISO equalizers, respectively. However, the polarization tracking speed that can be achieved by this adaptive equalizer is limited, and it cannot cope with scenarios with fast polarization changes, and in the DSCM system, single-tap MIMO operation is required in the processing of each subcarrier. In order to improve the polarization tracking speed, a polarization deinterleaving scheme based on three pilots is proposed to be applied in the coherent DSCM system, which can realize polarization deinterleaving before subcarrier demultiplexing, so that the subsequent equalization of each subcarrier only needs two polarization-independent SISO equalizers. However, in this scheme, when the amplitude of the received Y-polarization pilot component approaches zero, the problem of arithmetic overflow occurs. In order to avoid the problem of arithmetic overflow, another polarization deinterleaving scheme based on two pilots is proposed, which reduces the number of required pilots and also reduces the computational complexity. However, the above-mentioned pilot-based schemes need to reserve a certain guard interval between adjacent subcarriers to place the pilot, thus reducing the spectral efficiency of the signal.

[0004] In summary, in the scheme based on single-tap MIMO combined with multi-tap SISO equalizers, each subcarrier signal is processed independently after DSCM signal demultiplexing. Therefore, MIMO operations are included in the DSP processing module of each subcarrier, resulting in computational complexity. Furthermore, its polarization tracking speed is limited, making it unsuitable for applications with rapidly changing polarization states. In polarization decrosstalk descrambling schemes based on multiple pilots, current proposals all require reserving a certain width of guard band between adjacent subcarriers, thus reducing the signal's spectral efficiency. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies, such as complex calculation of the receiver equalizer and low signal transmission spectral efficiency. It provides a method for depolarization crosstalk based on a single pilot and a DSCM system, which significantly reduces the calculation complexity of the receiver equalizer, eliminates the guard band between adjacent subcarriers, and improves the spectral efficiency of the signal.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A method for depolarization crosstalk based on a single pilot is provided. A pilot is inserted into the X-polarized DSCM signal to track the polarization state changes experienced by the signal. At the receiving end, the changing Jones matrix is ​​estimated using the pilot components received on both polarizations, thereby achieving depolarization crosstalk removal. Specifically, the method includes the following steps:

[0008] S1. Insert a pilot signal P into the X-polarized DSCM signal. X (t), to obtain the signal E on the X-polarization after inserting the pilot signal. X (t) and the signal E on Y polarization Y (t);

[0009] S2. The polarization of the transmitted signal is rotated using a polarization rotator, causing the pilot signal originally located in X-polarization to appear in Y-polarization. The polarization is then compared between the pilot signal received at the receiver and the pilot signal P at the transmitter. X The frequency of (t) is used to obtain the frequency offset Δf;

[0010] S3. Perform frequency offset compensation to obtain the compensated X-polarized signal R' X (t) and the signal R' on Y polarization Y (t);

[0011] S4. After converting the pilot signal received at the receiving end to a lower frequency, filter it out using a low-pass filter to obtain the pilot components extracted from the X-polarization and Y-polarization. and The parameters of the Jones matrix are obtained from the extracted pilot components, thus yielding the complete Jones matrix.

[0012] S5. De-polarization crosstalk is performed on the signal obtained in step S4 according to the Jones matrix.

[0013] The application provides a single-pilot-based de-polarization crosstalk method, which realizes polarization de-crosstalk through a single pilot, thereby simplifying an equalizer used for equalization of each sub-carrier into two polarization-independent multi-tap SISO equalizers, and significantly reducing the calculation complexity of the equalizer at the receiving end. Meanwhile, the guard band between adjacent sub-carriers is eliminated, and the spectral efficiency of the signal is improved.

[0014] Further, the transmitting end inserts a pilot signal P X (t) is represented as:

[0015]

[0016] In the formula, S X (t) and S Y (t) are DSCM signals modulated on the X polarization and the Y polarization respectively, E X (t) and E Y (t) are signals on the X polarization and the Y polarization after the pilot is inserted, P X (t) is the pilot signal inserted on the X polarization, and the pilot P X (t) is specifically represented as:

[0017] P X (t) = A·exp(j2πf1t)

[0018] In the formula, A and f1 are the amplitude and the frequency of the pilot signal respectively, and t represents the time.

[0019] Further, step S2 specifically comprises:

[0020] The polarization rotator is used to rotate the polarization of the transmission signal, so that the pilot originally located on the X polarization appears a pilot component on the Y polarization, which is represented by the following formula:

[0021]

[0022] In the formula, J XX (t), J XY (t), J YX (t) and J YY (t) are four parameters of the Jones matrix, Δf and respectively represent the frequency offset and the phase noise, R X (t) and R Y (t) respectively represent the received X polarization and Y polarization signals; and j is the imaginary unit. That is, after polarization rotation, the pilot signal PX (t) the pilot component projected onto Y polarization; by comparing the received pilot and the pilot signal P X (t) at the transmitting end, the frequency offset Δf is obtained.

[0023] Further, in step S3, the frequency offset compensated X polarization signal R' X (t) and Y polarization signal R' Y (t) are obtained by the following equations:

[0024]

[0025] Further, in step S4, the pilot filtered by the low pass filter is represented by the following equation:

[0026]

[0027] wherein, and represent the pilot components extracted from X polarization and Y polarization respectively, is the phase noise carried;

[0028] Since the Jones matrix is a unitary matrix, the parameters of the Jones matrix satisfy the following relationship:

[0029]

[0030] According to the obtained Jones matrix parameters and the complete Jones matrix is obtained:

[0031]

[0032] wherein, and represent the four parameters of the estimated Jones matrix respectively.

[0033] Further, the depolarization crosstalk of the signal in step S5 is realized by the following equation:

[0034]

[0035] wherein, and are the X polarization and Y polarization signals after depolarization crosstalk respectively.

[0036] Further, when there is a receiving end XY delay, firstly, the receiving end XY delay is estimated, then signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration, and finally steps S1 to S5 are executed. The application considers the influence of the receiving end XY delay on the pilot-based low complexity polarization deinterleaving scheme, and through estimation and calibration of the receiving end XY delay, the effectiveness of the pilot-based polarization deinterleaving method is further ensured.

[0037] The application also provides a single pilot polarization deinterleaving-based DSCM system, comprising:

[0038] Firstly, the receiving end XY delay is tested, and the transceiver end digital signal processing DSP flow comprises:

[0039] Transmitting end DSP: firstly, a single frequency cosine signal is generated, and after resampling, the single frequency cosine signal is modulated on the X polarization optical carrier;

[0040] Receiving end DSP: receiving pilot signals on two polarizations, extracting the real parts of the two pilot signals, and converting them to the frequency domain, then extracting the phase information of the pilot, and finally calculating the receiving end XY delay;

[0041] After the receiving end XY delay is estimated, signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration, so as to ensure the effectiveness of the polarization deinterleaving algorithm;

[0042] Then, data transmission is performed, and the transceiver end digital signal processing DSP flow comprises:

[0043] Transmitting end DSP: firstly, a pseudo-random bit sequence is mapped to a QAM symbol, and a root-raised cosine filter is used for Nyquist shaping of the QAM symbol; the shaped symbol is used to generate a DSCM signal with four subcarriers; after resampling, a pilot is inserted into the X polarization DSCM signal; finally, different polarization variation speeds are simulated in the transmitting end DSP;

[0044] Receiving end DSP: firstly, the receiving signal is compensated for the receiving end XY delay using the estimated receiving end XY delay value; then, frequency offset compensation and polarization deinterleaving are realized by using the above-mentioned polarization deinterleaving method, then the DSCM signal is subcarrier demultiplexed; for each demultiplexed subcarrier signal, matching filtering, dispersion compensation, clock recovery, synchronization, simplified equalization and phase noise compensation are sequentially performed, and finally the bit error rate is calculated.

[0045] The application provides a DSCM system based on single pilot polarization crosstalk demodulation, which utilizes a single pilot in a coherent DSCM system, realizes low-complexity polarization crosstalk demodulation before subcarrier demultiplexing, simplifies an equalizer used in each subsequent subcarrier processing from a traditional multi-tap 2*2 MIMO structure to two polarization-independent multi-tap SISO structures, and greatly reduces the complexity of the equalizer. Meanwhile, compared with an existing pilot-based polarization crosstalk demodulation scheme, the proposed scheme needs less pilots, only one pilot, and does not need to reserve a certain width of a guard band between adjacent subcarriers. In addition, the application also considers the influence of a receiving end XY time delay on the effectiveness of the pilot-based polarization crosstalk demodulation algorithm, and proposes a receiving end XY time delay estimation algorithm based on a single pilot training sequence, calibrates the system by using the estimated receiving end XY time delay, so as to ensure the effectiveness of the pilot-based low-complexity polarization crosstalk demodulation algorithm, and further realizes the simplification of the subsequent equalizer. Therefore, the application realizes a low-complexity high-performance coherent DSCM system based on single pilot polarization crosstalk demodulation.

[0046] Further, when the receiving end time delay test is performed, for the transmitting end DSP, a single frequency cosine pilot signal on the X polarization is represented as: A is the amplitude of the pilot signal, and f2 is the frequency of the pilot signal; after considering the transmitting end time delay and polarization rotation, the pilot signal is represented as:

[0047]

[0048] Wherein, τ TXI is the I path time delay of the X polarization of the transmitting end, a and α are the amplitude and phase of the parameter J XX (t) respectively, and b and β are the amplitude and phase of the parameter J XY (t) respectively.

[0049] Further, when the receiving end time delay test is performed, for the receiving end DSP: after considering the receiving end time delay, the received pilot signal is represented as:

[0050]

[0051] Wherein, and are the I path and Q path signals of the X polarization and the I path and Q path signals of the Y polarization respectively, τ RXI , τ RXQ , τ RYI and τ RYQ are the X polarization I path time delay, the X polarization Q path time delay, the Y polarization I path time delay and the Y polarization Q path time delay of the receiving end respectively;

[0052] The receiving end XY time delay τ RXY is defined as: τRXY = τ RYI - τ RXI ;

[0053] Transform the formula and the formula to the frequency domain:

[0054]

[0055]

[0056] Then the receiving end XY time delay is calculated as follows:

[0057]

[0058] Wherein, and respectively represent the received X polarization I route signal Pilot components in the positive half frequency and negative half frequency, and respectively represent the received Y polarization I route signal Pilot components in the positive half frequency and negative half frequency, and represent the conjugate operation.

[0059] Compared with the prior art, the beneficial effects of the present application are:

[0060] The single pilot depolarization crosstalk method and DSCM system based on the present application can further simplify the equalizer to only contain two multi-tap SISO equalizer structures compared with the current single-tap MIMO combined with multi-tap SISO equalizer scheme, further reduce the complexity of the equalizer, and significantly improve the polarization tracking capability. Compared with the current polarization depolarization crosstalk scheme based on multiple pilots, the number of pilots required by the present application is less, only one is required, and there is no need to reserve a certain width of guard band between adjacent subcarriers, which improves the spectral efficiency of the signal. In addition, the present application also considers the influence of the receiving end XY time delay on the effectiveness of the pilot-based polarization depolarization crosstalk algorithm, and proposes a receiving end XY time delay estimation algorithm based on a single pilot training sequence. Based on the proposed receiving end XY time delay algorithm, the receiving end XY time delay can be accurately calibrated, thereby ensuring the effectiveness of the proposed pilot-based low complexity polarization depolarization crosstalk algorithm, further simplifying the subsequent equalizer, and thus realizing a low complexity and high performance coherent DSCM system. BRIEF DESCRIPTION OF DRAWINGS

[0061] Figure 1 It is a flowchart of a single pilot depolarization crosstalk method.

[0062] Figure 2Spectrum diagram for the single pilot de-polarization crosstalk method of embodiment 1.

[0063] Figure 3 Pilot training sequence structure in embodiment 2. Figure 3 a represents the pilot structure of the transmitting end, Figure 3 b represents the pilot structure of the receiving end.

[0064] Figure 4 Simulation structure diagram of the DSCM system in embodiment 3.

[0065] Figure 5 Bit error rate performance curve of the single pilot de-polarization crosstalk method under different polarization rotation speeds.

[0066] Figure 6 Influence of different receiving end XY time delays on the performance of the single pilot de-polarization crosstalk method.

[0067] Figure 7 Estimated error under different preset time delay values.

[0068] Figure 8 Bit error rate curve comparison under different receiving end XY time delays. DETAILED DESCRIPTION

[0069] The application will be further described below in conjunction with specific embodiments. The accompanying drawings are only used for illustrative description, and represent only schematic diagrams, not physical diagrams, and should not be understood as a limitation on the patent. In order to better illustrate the embodiments of the application, some components in the drawings may be omitted, enlarged or reduced, and do not represent the actual size of the product. It is understandable for those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.

[0070] The same or similar reference numerals in the drawings of the embodiments of the application correspond to the same or similar components; in the description of the application, it should be understood that the orientations or positional relationships indicated by terms such as "upper", "lower", "left", "right" are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for illustrative description, and should not be understood as a limitation on the patent, and for those skilled in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0071] Embodiment 1:

[0072] As Figure 1As shown, the embodiment is a single pilot depolarization crosstalk method, a pilot is inserted in the X polarization DSCM signal to track the polarization state change experienced by the signal, and the changed Jones matrix is estimated by using the pilot components received on the two polarizations at the receiving end to achieve signal depolarization crosstalk; as shown Figure 2 As shown, the spectral diagram of the single pilot depolarization crosstalk scheme is given. Specifically, the following steps are included:

[0073] Step 1. Insert a pilot signal P X (t) into the X polarization DSCM signal to obtain the signal E X (t) on the X polarization and the signal E Y (t) on the Y polarization after inserting the pilot signal.

[0074] The signal after inserting the pilot at the transmitting end can be represented as:

[0075]

[0076] Where S X (t) and S Y (t) are the DSCM signals modulated on the X polarization and the Y polarization respectively, E X (t) and E Y (t) are the signals on the X polarization and the Y polarization after inserting the pilot respectively, P X (t) is the pilot signal inserted on the X polarization, and the pilot P X (t) can be specifically represented as:

[0077] P X (t) = A·exp(j2πf1t)

[0078] Where A and f1 are the amplitude and frequency of the pilot signal respectively, and t represents the time.

[0079] Step 2. Use the polarization rotator to rotate the polarization of the transmission signal, so that the pilot originally on the X polarization appears on the Y polarization. By comparing the frequency of the pilot signal received at the receiving end with the frequency of the pilot signal P X (t) at the transmitting end, the frequency offset Δf is obtained.

[0080] As shown, after polarization rotation, the pilot originally on the X polarization will appear on the Y polarization. This process can be represented as follows: Figure 2

[0081]

[0082] Where J XX (t), J XY (t), J YX ​(t) and J YY (t) are four parameters of Jones matrix, Δf and respectively represent frequency offset and phase noise, R X (t) and R Y (t) represent received X-polarization and Y-polarization signals respectively. j is imaginary unit; is the pilot component projected on Y-polarization by pilot located on X-polarization after polarization rotation. By comparing received pilot with pilot P X (t) transmitted by the transmitting end, frequency offset Δf can be obtained.

[0083] S3. Frequency offset compensation is performed to obtain frequency offset compensated X-polarization signal R' X (t) and Y-polarization signal R' Y (t).

[0084] Compensated X-polarization signal R' X (t) and Y-polarization signal R' Y (t) can be represented as:

[0085]

[0086] Step 4. After transforming received pilot signal at the receiving end to low frequency, low-pass filter is used to filter out pilot components extracted from X-polarization and Y-polarization and According to the extracted pilot components, parameters of Jones matrix are obtained, and thus complete Jones matrix is obtained.

[0087] After transforming pilot to low frequency and using low-pass filter to filter out, the extracted pilot can be represented as follows:

[0088]

[0089] wherein, and respectively represent pilot components extracted from X-polarization and Y-polarization, is phase noise carried. According to the above formula, the extracted pilot component is the Jones matrix parameter J carrying phase noise XX (t) and J YX (t). Since Jones matrix is a unitary matrix, i.e. its parameters satisfy the following relationship:

[0090]

[0091] Therefore, according to obtained Jones matrix parameters and The complete Jones matrix can be obtained as:

[0092]

[0093] wherein, and respectively represent four parameters of the estimated Jones matrix.

[0094] Step 5. Perform depolarization crosstalk on the signal according to the Jones matrix obtained in step 4.

[0095] The process of depolarization crosstalk on the signal can be expressed as follows:

[0096]

[0097] wherein, and are the signals of the X polarization and the Y polarization after depolarization crosstalk respectively.

[0098] The method for depolarization crosstalk based on a single pilot provided in the embodiment realizes polarization depolarization crosstalk through a single pilot, thereby simplifying the equalizer used for equalization of each subcarrier into two polarization-independent multi-tap SISO equalizers, significantly reducing the calculation complexity of the equalizer at the receiving end. At the same time, the guard band between adjacent subcarriers is eliminated, and the spectral efficiency of the signal is improved.

[0099] Embodiment 2:

[0100] The embodiment provides a method for estimating the XY time delay at the receiving end based on a single pilot training sequence. For the method for depolarization crosstalk in embodiment 1, when there is an XY time delay at the receiving end, the estimated Jones matrix will be expressed as:

[0101]

[0102] wherein, τ RXY is the XY time delay at the receiving end. At this time, the estimated Jones matrix cannot correctly reflect the change of polarization, thereby causing the failure of the depolarization crosstalk algorithm.

[0103] In order to ensure the effectiveness of the low-complexity polarization depolarization crosstalk algorithm based on the pilot, the XY time delay at the receiving end needs to be calibrated. Therefore, the embodiment provides an algorithm for estimating the XY time delay at the receiving end based on a single pilot training sequence, Figure 3 the structure of the training sequence is given, Figure 3 wherein a gives the pilot structure at the transmitting end, that is, a single-frequency cosine pilot signal on the X polarization, which can be expressed as:

[0104]

[0105] where A is the amplitude of the pilot signal, and f2 is the frequency of the pilot signal.

[0106] Considering the transmit end delay and polarization rotation, it can be expressed as:

[0107]

[0108] where τ TXI is the I-path delay of the X-polarization at the transmit end, and a and a are the amplitude and phase of the parameter J XX (t), respectively, and b and β are the amplitude and phase of the parameter J XY (t), respectively.

[0109] Considering the receive end delay, the received pilot signal can be expressed as:

[0110]

[0111] where and are the I-path and Q-path signals of the X-polarization and the I-path and Q-path signals of the Y-polarization received, respectively, τ RXI , τ RXQ , τ RYI , and τ RYQ are the I-path delay of the X-polarization, the Q-path delay of the X-polarization, the I-path delay of the Y-polarization, and the Q-path delay of the Y-polarization at the receive end, respectively.

[0112] The XY delay τ RXY at the receive end is defined as: τ RXY = τ RYI - τ RXI .

[0113] Therefore, the XY delay at the receive end can be obtained using the formula and the formula First, transform the above formula to the frequency domain:

[0114]

[0115] Then the XY delay at the receive end can be calculated as follows:

[0116]

[0117] where and represent the signal in the positive half frequency and the negative half frequency of the pilot component of the received X-polarization I-path, respectively, and represent the signal in the positive half frequency and the negative half frequency of the pilot component of the received Y-polarization I-path, respectively, and * represents the conjugate operation.

[0118] This embodiment considers the impact of receiver XY delay on the effectiveness of pilot-based polarization de-crosstalk algorithm, and proposes a receiver XY delay estimation method based on a single pilot training sequence. The estimated receiver XY delay is used to calibrate the system, thereby ensuring the effectiveness of the pilot-based low-complexity polarization de-crosstalk algorithm and further simplifying the subsequent equalizer.

[0119] Example 3:

[0120] This embodiment proposes a single pilot-based polarization crosstalk depolarization scrambling (DSCM) system. For example... Figure 4 The diagram shows a DSCM simulation system. The transmitting optical carrier is generated by a 100kHz laser with an output power of 16dBm. A polarization beam splitter (PBS) splits the carrier into X-polarized and Y-polarized light, which are then input to two IQ modulators (IQMs) for signal modulation. Two 50Gbaud DSCM signals generated offline at the transmitting end are loaded into the two IQMs for electro-optical conversion. The output signals of the two modulators are coupled by a polarization beam combiner (PBC). For XY delay estimation at the receiving end, tests were conducted in a back-to-back optical scenario, with the local oscillator (LO) provided by the transmitting laser. Additionally, a polarization rotator is needed to introduce polarization crosstalk. For data transmission, the signal is transmitted through an 80km optical fiber, and an optical signal-to-noise ratio (OSNR) module is used to adjust the system's OSNR. A separate laser serves as the LO at the receiving end. The XY delay estimation and the rest of the data transmission stage are the same: the LO and the signal are each split into two orthogonal polarization states by a PBS, then input into two 90° optical mixers, and detected by four pairs of photodetectors (PDs). The detected data is processed in the receiver's DSP. The system uses a sampling rate of 64 GSa / s.

[0121] When performing data signal processing, the XY delay of the system receiver is first tested and estimated, and the XY delay of the receiver is obtained by using the receiver delay estimation method provided in Example 2.

[0122] The DSP algorithm flow for the transceiver end is as follows:

[0123] (1) Transmitter DSP:

[0124] In the transmitter DSP, a single-frequency cosine signal with a frequency of 2GHz is first generated, and after resampling, it is modulated onto an X-polarized optical carrier.

[0125] (2) Receiver DSP:

[0126] At the receiving end, the pilot signals on two polarizations are received, the real parts of the two pilot signals are extracted and converted to the frequency domain. Then the phase information of the pilot is extracted, and finally the XY delay at the receiving end is calculated.

[0127] After the XY delay at the receiving end is estimated, signal transmission is performed, and the estimated delay value is used for XY delay calibration at the receiving end to ensure the effectiveness of the polarization demultiplication algorithm.

[0128] Then the DSP in the data transmission stage is performed:

[0129] (1) Transmitter DSP:

[0130] In the transmitter DSP, first, a pseudo-random bit sequence is mapped to 16QAM symbols, and a root-raised cosine filter with a roll-off factor of 0.1 is used to perform Nyquist shaping on the QAM symbols. The shaped symbols are used to generate a 50Gbaud DSCM signal with four subcarriers. After resampling, a pilot is inserted into the DSCM signal on the X polarization. Finally, different polarization variation speeds are simulated in the transmitter DSP.

[0131] (2) Receiver DSP:

[0132] At the receiving end, first, the received signal is compensated for the XY delay at the receiving end using the estimated XY delay value at the receiving end. Then, the polarization crosstalk demultiplication method described in Example 1 is used to perform frequency offset compensation and polarization crosstalk demultiplication, and then the DSCM signal is subcarrier demultiplexed. For each demultiplexed subcarrier signal, in turn, matched filtering, dispersion compensation, clock recovery, synchronization, simplified equalization, and phase noise compensation are performed, and finally the bit error rate is calculated.

[0133] Result analysis:

[0134] Based on the simulation device of Figure 4 , first, the performance of the proposed single-pilot polarization crosstalk demultiplication scheme is evaluated. Figure 5 The bit error rate curves of the proposed single-pilot polarization crosstalk demultiplication scheme under different polarization variation speeds are given, and the bit error rate curves without polarization crosstalk are given for comparison. From the simulation results, even if the polarization variation speed reaches 10 Mrad / s, there is almost no OSNR loss compared to the case without polarization crosstalk, which shows that the proposed single-pilot polarization crosstalk demultiplication algorithm can effectively solve the signal crosstalk under high-speed variation of the polarization state.

[0135] Figure 6The influence of different receiving end XY time delays on the performance of the proposed single pilot polarization deinterleaving scheme is given. As can be seen from the figure, with the increase of the receiving end XY time delay, the performance of the pilot-based polarization deinterleaving algorithm gradually deteriorates. When there is a 2ps receiving end XY time delay, there is an OSNR loss of 3.2dB at the hard decision forward error correction threshold. When there is a 3ps receiving end XY time delay, the pilot-based polarization deinterleaving algorithm will fail. Therefore, compensation for the receiving end XY time delay damage is a prerequisite to ensure the effectiveness of the pilot-based polarization deinterleaving algorithm, thereby further simplifying the subsequent equalizer.

[0136] Then the accuracy of the proposed single pilot training sequence-based receiving end XY time delay estimation algorithm is evaluated. Figure 7 The estimated time delay value and the corresponding estimation error under different preset receiving end XY time delays are given. As can be seen from the simulation results, the estimation error that can be achieved by the proposed receiving end XY time delay estimation algorithm is within ±0.15ps.

[0137] After estimating the receiving end XY time delay value, the calibration of the receiving end XY time delay can be achieved. Figure 8 The bit error rate performance curves of the single pilot-based polarization deinterleaving algorithm under different receiving end XY time delays without time delay compensation and with time delay compensation are given, and the polarization change speed is 10Mrad / s. As can be seen from the figure, without receiving end XY time delay compensation, the performance deteriorates rapidly with the increase of the time delay value. After receiving end XY time delay compensation is performed on the signal, the bit error rate curve is almost flat, and is basically consistent with the performance without receiving end XY time delay.

[0138] Compared with the current single-tap MIMO combined with multi-tap SISO equalizer scheme, the present application can further simplify the equalizer to a structure containing only two multi-tap SISO equalizers, further reduce the complexity of the equalizer, and significantly improve the polarization tracking capability. Compared with the current multi-pilot-based polarization deinterleaving scheme, the present application requires fewer pilots, only one, and does not need to reserve a certain width of guard band between adjacent subcarriers, thereby improving the spectral efficiency of the signal. In addition, the present application also considers the influence of the receiving end XY time delay on the effectiveness of the pilot-based polarization deinterleaving algorithm, and proposes a single pilot training sequence-based receiving end XY time delay estimation algorithm. Based on the proposed receiving end XY time delay algorithm, accurate calibration of the receiving end XY time delay can be achieved, thereby ensuring the effectiveness of the proposed pilot-based low complexity polarization deinterleaving algorithm, further simplifying the subsequent equalizer, and thereby realizing a low complexity high performance coherent DSCM system.

[0139] In the specific contents of the foregoing specific embodiments, each technical feature can be combined arbitrarily without contradiction. In order to make the description simple, all possible combinations of the foregoing technical features are not described, but as long as the combinations of the technical features do not contradict, they should be considered as the scope of the present disclosure.

[0140] Obviously, the above embodiments of the present application are merely exemplary and are not intended to limit the implementation modes of the present application. Based on the above description, other different forms of changes or variations can be made by those of ordinary skill in the art. It is not necessary or possible to exhaust all the implementation modes. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the claims of the present application.

Claims

1. A single pilot based method for depolarization crosstalk mitigation, the method comprising: A pilot is inserted into the X-polarized DSCM signal to track the polarization state variation experienced by the signal, and the Jones matrix of the variation is estimated at the receiving end using the pilot components received on the two polarizations to achieve depolarization crosstalk of the signal; specifically comprising the following steps: S1. Insert a pilot signal P into the X-polarized DSCM signal. X (t), to obtain the signal E on the X-polarization after inserting the pilot signal. X (t) and the signal E on Y polarization Y (t); S2. Polarization rotation is performed on the transmission signal by using a polarization rotator, so that the pilot originally located on the X polarization appears a pilot component on the Y polarization. By comparing the pilot signal received at the receiving end and the pilot signal P X (t) to obtain the frequency offset Δf; S3. Perform frequency offset compensation to obtain compensated X-polarized signal R' X (t) and Y-polarized signal R' Y (t); S4. Transforming the pilot signal received by the receiving end to low frequency, and filtering out with a low-pass filter to obtain pilot components extracted from X polarization and Y polarization and According to the extracted pilot components, parameters of the Jones matrix are obtained, so as to obtain the complete Jones matrix; S5. Depolarization crosstalk is performed on the signal according to the Jones matrix obtained in step S4.

2. The single pilot de-polarization crosstalk method according to claim 1, wherein, In step S1, the transmitting end inserts a pilot signal P X (t) is represented as: where S X (t) and S Y (t) are DSCM signals modulated on X-polarization and Y-polarization, respectively, E X (t) and E Y (t) are signals on X-polarization and Y-polarization after inserting pilot, respectively, P X (t) is the pilot signal inserted on X-polarization, and the pilot P X (t) is specifically expressed as: P X (t) = A - exp(j2πf1t) In the formula, A and f1 are the amplitude and frequency of the pilot signal respectively, and t represents the time.

3. The single pilot de-polarization crosstalk method according to claim 2, wherein, Step S2 specifically comprises: The polarization rotator is used to rotate the polarization of the transmission signal, so that the pilot originally located on the X-polarization appears on the Y-polarization, which is represented by the following formula: where J XX (t), J XY (t), J YX (t) and J YY (t) are four parameters of the Jones matrix, Δf and respectively represent frequency offset and phase noise, R X (t) and R Y (t) represent received X-polarization and Y-polarization signals respectively; j is the imaginary unit; is the pilot component projected onto Y-polarization from the pilot signal P X (t) located at X-polarization after polarization rotation; frequency offset Δf is obtained by comparing the frequency of received pilot with the frequency of the pilot signal P X (t) at the transmitting end.

4. The single pilot de-polarization crosstalk method according to claim 3, wherein, In step S3, the frequency offset-compensated X-polarized signal R'(t) is obtained by the following equation X (t) and Y-polarized signal R'(t) are obtained by the following equations Y (t):

5. The single pilot de-polarization crosstalk method according to claim 4, wherein, In step S4, the pilot filtered by the low-pass filter is represented by the following formula: wherein and denote pilot components extracted from the X-polarization and Y-polarization, respectively, is the phase noise carried. Since the Jones matrix is a unitary matrix, the parameters of the Jones matrix satisfy the following relationship: From the obtained Jones matrix parameters and the complete Jones matrix is obtained: where and denote the four parameters of the estimated Jones matrix, respectively, and * stands for the conjugate operation.

6. The single pilot de-polarization crosstalk method according to claim 5, wherein, In step S5, the depolarization crosstalk of the signal is achieved by the following formula: wherein and are the X-polarized and Y-polarized signals, respectively, after depolarization crosstalk.

7. The single pilot de-polarization crosstalk method according to any one of claims 1 to 6, wherein, When the receiving end XY delay exists, first, the receiving end XY delay is estimated, then the signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration, and finally steps S1 to S5 are executed.

8. A DSCM system based on single pilot de-polarization crosstalk, characterized in that, Comprise: First, the receiving end XY delay is tested, and the transceiver digital signal processing DSP process comprises: Transmitting end DSP: first, a single-frequency cosine signal is generated, and after resampling, the single-frequency cosine signal is modulated on the X-polarized optical carrier; Receiving end DSP: receiving the pilot signals on the two polarizations, extracting the real parts of the two pilot signals, and converting them to the frequency domain, then extracting the phase information of the pilot, and finally calculating the receiving end XY delay; After the receiving end XY delay is estimated, the signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration to ensure the effectiveness of the depolarization algorithm; Then, data transmission is performed, and the transceiver digital signal processing DSP process comprises: Transmitting end DSP: first, the pseudo-random bit sequence is mapped to QAM symbols, and the root-raised cosine filter is used for Nyquist shaping of the QAM symbols; the shaped symbols are used to generate a DSCM signal with four subcarriers; after resampling, a pilot is inserted into the X-polarized DSCM signal; finally, different polarization variation speeds are simulated in the transmitting end DSP; Receiving end DSP: first, the receiving end XY delay compensation is performed on the received signal using the estimated receiving end XY delay value; then, the depolarization crosstalk method of any one of claims 1 to 6 is used to achieve frequency offset compensation and polarization de-crosstalk, and then the subcarrier demultiplexing of the DSCM signal is performed; for each demultiplexed subcarrier signal, matching filtering, dispersion compensation, clock recovery, synchronization, simplified equalization and phase noise compensation are performed in turn, and finally the bit error rate is calculated.

9. The single pilot de-polarization crosstalk based DSCM system of claim 8, wherein, When the receiving end delay test is performed, for the transmitting end DSP, a single frequency cosine pilot signal on the X polarization is represented as: A is the amplitude of the pilot signal, and f2 is the frequency of the pilot signal; after considering the transmitting end delay and polarization rotation, the pilot signal is represented as: where τ TXI is the I-channel delay of the transmit end X, a and a are the amplitude and phase of the parameter J XX (t) respectively, and b and β are the amplitude and phase of the parameter J XY (t) respectively.

10. The single pilot de-polarization crosstalk based DSCM system of claim 9, wherein, When the receiving end delay test is performed, for the receiving end DSP: after considering the receiving end delay, the received pilot signal is represented as: wherein, and are the received I and Q channel signals of the X polarization, I and Q channel signals of the Y polarization, respectively, τ RXI , τ RXQ , τ RYI and τ RYQ are the X polarization I channel delay, X polarization Q channel delay, Y polarization I channel delay and Y polarization Q channel delay at the receiving end, respectively; the XY delay τ RXY at the receiving end is defined as: τ RXY = τ RYI - τ RXI ; Transforming the equation and the equation to the frequency domain: The receiving end XY delay is calculated as follows: When the receiving end XY delay exists, first, the receiving end XY delay is estimated, then the signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration, and finally steps S1 to S5 are executed. Comprise: First, the receiving end XY delay is tested, and the transceiver digital signal processing DSP process comprises: Transmitting end DSP: first, a single-frequency cosine signal is generated, and after resampling, the single-frequency cosine signal is modulated on the X-polarized optical carrier; Receiving end DSP: receiving the pilot signals on the two polarizations, extracting the real parts of the two pilot signals, and converting them to the frequency domain, then extracting the phase information of the pilot, and finally calculating the receiving end XY delay; After the receiving end XY delay is estimated, the signal transmission is performed, and the estimated delay value is used for receiving end XY delay calibration to ensure the effectiveness of the depolarization algorithm; Then, data transmission is performed, and the transceiver digital signal processing DSP process comprises: Transmitting end DSP: first, the pseudo-random bit sequence is mapped to QAM symbols, and the root-raised cosine filter is used for Nyquist shaping of the QAM symbols; the shaped symbols are used to generate a DSCM signal with four subcarriers; after resampling, a pilot is inserted into the X-polarized DSCM signal; finally, different polarization variation speeds are simulated in the transmitting end DSP; Receiving end DSP: first, the receiving end XY delay compensation is performed on the received signal using the estimated receiving end XY delay value; then, the depolarization crosstalk method of any one of claims 1 to 6 is used to achieve frequency offset compensation and polarization de-crosstalk, and then the subcarrier demultiplexing of the DSCM signal is performed; for each demultiplexed subcarrier signal, matching filtering, dispersion compensation, clock recovery, synchronization, simplified equalization and phase noise compensation are performed in turn, and finally the bit error rate is calculated. When the receiving end delay test is performed, for the receiving end DSP: after considering the receiving end delay, the received pilot signal is represented as: The receiving end XY delay is calculated as follows: where and represent the received X-polarized I-channel signal at the pilot components at the positive and negative half frequencies, respectively and represent the received Y-polarized I-channel signal at the pilot components at the positive and negative half frequencies, respectively

Images