WO2023246749A1

WO2023246749A1 - Data sending method, data receiving method, dual-polarization transmitter and single-polarization receiver

Info

Publication number: WO2023246749A1
Application number: PCT/CN2023/101296
Authority: WO
Inventors: 邢祯平
Original assignee: 华为技术有限公司
Priority date: 2022-06-21
Filing date: 2023-06-20
Publication date: 2023-12-28
Also published as: CN117318826A

Abstract

The present application belongs to the technical field of optical communications. Provided are a data sending method, a data receiving method, a dual-polarization transmitter and a single-polarization receiver. The data sending method is applied to a coherent optical communication system, and comprises: acquiring a symbol sequence corresponding to a first optical carrier and training sequences respectively corresponding to two polarization states which are orthogonal to each other; respectively performing differential polarization time coding on the symbol sequence and the training sequences, so as to obtain a dual-polarization complex digital signal corresponding to the first optical carrier, wherein in the dual-polarization complex digital signal, the coded training sequences are located at an initial position of each data frame; on the basis of dual-polarization IQ modulation, modulating the dual-polarization complex digital signal onto the first optical carrier; and sending the first optical carrier onto which the dual-polarization complex digital signal is modulated. By means of the present application, dual-polarization-state transmission and single-polarization-state reception can be realized in a coherent optical communication system, and different signals are transmitted in a dual-polarization state, thereby improving the utilization rate of a transmission spectrum.

Description

Methods of data transmission and reception, dual polarization transmitters and single polarization receivers

This application claims priority to the Chinese patent application submitted to the State Intellectual Property Office of China on June 21, 2022, with application number 202210708907.8 and the application name "Method for transmitting and receiving data, dual polarization transmitter and single polarization receiver" , the entire contents of which are incorporated herein by reference.

Technical field

This application relates to the field of optical communication technology, and in particular to a method of data transmission and reception, a dual-polarization transmitter and a single-polarization receiver.

Background technique

In the passive optical network (PON), the coherent optical communication system uses local oscillator (LO) at the receiving end, so the forward error correction (FEC) threshold is The required receiver has higher sensitivity and is expected to support a larger splitting ratio, that is, a central node (such as an optical line terminal (OLT)) can be connected to more leaf nodes (such as an optical network unit) , ONU)), so coherent optical communication systems have attracted much attention.

In a PON based on a coherent optical communication system, considering the cost of leaf nodes, and the leaf nodes can receive data, consider that the central node uses two orthogonal polarization states to send data, and the leaf nodes receive data in a single polarization state. For example, the central node uses two orthogonal polarization states to alternately zero out and transmit the same signal to ensure that the receiver performance is independent of the polarization state at the receiving end. In this way, it is equivalent to two orthogonal polarization states transmitting a single polarization signal with a baud rate halved, making the transmission spectrum utilization rate relatively low.

Contents of the invention

This application provides a method of data transmission and reception, a dual-polarization transmitter and a single-polarization receiver, which can realize dual-polarization transmission and single-polarization reception in a coherent optical communication system, and use dual-polarization states to simultaneously transmit signals. , improve emission spectrum utilization.

In a first aspect, the present application provides a data transmission method, which is applied in a coherent optical communication system. The method includes: obtaining a symbol sequence corresponding to a first optical carrier and training sequences corresponding to two polarization states. The two The polarization state is orthogonal, and the symbol sequence is obtained by performing symbol mapping on the data to be sent corresponding to the first optical carrier; differential polarization time coding is performed on the symbol sequence and the training sequence respectively to obtain the dual polarization corresponding to the first optical carrier. Complex digital signal. In the dual-polarization complex digital signal, the encoded training sequence is located at the starting position of each data frame; according to the dual-polarization in-phase quadrature-phase (IQ) modulation, the dual-polarization The complex digital signal is modulated onto the first optical carrier; and the first optical carrier modulated with the dual-polarization complex digital signal is sent.

In the solution shown in this application, in the coherent optical communication system, two orthogonal polarization states correspond to training sequences respectively. Using the training sequences can enable the single-polarization receiver at the receiving end to determine the frame head position. Perform differential polarization time coding on the symbol sequence and the training sequence corresponding to the two polarization states respectively to obtain the dual-polarization complex digital signal corresponding to the first optical carrier. According to the dual-polarization IQ modulation, the dual-polarization complex digital signal is modulated to the first optical carrier. On the other hand, two orthogonal polarization states are used to transmit different signals at the same time, which can improve the utilization of the emission spectrum.

In one example, performing differential polarization time coding on the symbol sequence and the training sequence respectively to obtain a dual-polarization complex digital signal corresponding to the first optical carrier includes: based on the symbol sequence and each time slot in the training sequence The symbols of the two corresponding polarization states generate the coding matrix corresponding to each time slot. The symbols of the two polarization states corresponding to each time slot are respectively one row or one column in the coding matrix corresponding to each time slot; based on the Tth The coding matrix corresponding to the T-th time slot, the differential polarization time coding matrix corresponding to the T-1th time slot and the stretching factor corresponding to the T-th time slot are obtained to obtain the differential polarization time coding matrix corresponding to the T-th time slot. The dual-polarization complex digital signal is obtained by sequentially connecting the differential polarization time encoding matrices corresponding to each time slot, and the stretching factor corresponding to the T-th time slot is determined by the encoding matrix or differential polarization time encoding corresponding to the T-1th time slot. The value of the determinant of the matrix is determined, and T is an integer greater than or equal to 2.

The solution shown in this application provides a method for differential polarization time encoding.

In an example, the two polarization states include a first polarization state and a second polarization state; performing differential polarization time encoding on the symbol sequence and the training sequence includes: performing differential polarization time encoding on the first polarization state and the second polarization state. Differential polarization time encoding is performed on the symbol sequence corresponding to the first polarization state and the second polarization state; differential polarization time encoding is performed on the training sequence corresponding to the first polarization state and the second polarization state.

In the solution shown in this application, the symbol sequences and training sequences corresponding to the two polarization states can be separated for differential polarization time encoding.

In an example, the training sequences respectively corresponding to the two polarization states include a first sub-training sequence and a second sub-training sequence, and the periods of the first sub-training sequence and the second sub-training sequence are different.

In the solution shown in this application, the training sequence corresponding to each polarization state includes two-period sub-training sequences. The periods of the two-period sub-training sequences are different, which enables the receiving end to use the two-period sub-training sequence for processing. Frequency offset estimation.

In a second aspect, the present application provides a data receiving method, which method is applied to a coherent optical communication system. The method includes: acquiring a first complex digital signal of a third polarization state corresponding to a first optical carrier; The complex digital signal is subjected to frequency offset compensation processing; the signal after frequency offset compensation processing is subjected to single-in single-output (SISO) equalization processing; the equalized signal is subjected to differential polarization time decoding to obtain The symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, the third polarization state and the fourth polarization state are orthogonal; based on the symbol sequence, the data corresponding to the third polarization state and the fourth polarization state are restored .

In the solution shown in this application, the data receiving method is applied to the coherent optical communication system, and the execution subject may be a single polarization receiver. The single-polarization receiver receives an optical signal of a single polarization state, and uses the optical signal of a single polarization state to recover the data on the two polarization states sent by the data transmitter, so that the single-polarization receiver only needs to receive data of its own corresponding bandwidth. Low demand, minimizing the complexity and cost of single polarization receivers.

In one example, performing differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state includes: based on each time slot group in the equalized signal including: The symbols of the two third polarization states and the Alamouti coding rules generate two symbols of the fourth polarization state included in each time slot group. Each time slot group includes two symbols of the third polarization state and two symbols of the third polarization state. Symbols in four polarization states form the receiving end matrix corresponding to each time slot group. Each time slot group is obtained by grouping the equalized signals in units of two consecutive time slots; based on the nth time slot group corresponding The receiving end matrix, the receiving end matrix corresponding to the n-1th time slot group, and the scaling factor corresponding to the nth time slot group are used to obtain the decoded matrix corresponding to the nth time slot group, and the scaling factor corresponding to the nth time slot group. The factor is determined by the value of the determinant of the receiving end matrix corresponding to the n-1th time slot group or the value of the determinant of the decoded matrix. n is an integer greater than or equal to 2; in the n-th time slot group corresponding to In the decoded matrix, the symbols of the two polarization states of the time slots before differential polarization time encoding corresponding to the nth time slot group are obtained.

Among them, Alamouti is a personal name, and the Alamouti encoding rule is the encoding rule named after Alamouti.

The solution shown in this application provides a method for differential polarization time decoding.

In one example, the method further includes: making a decision on symbol sequences corresponding to the third polarization state and the fourth polarization state, and determining a first symbol error, where the first symbol error is used to characterize the corresponding sequence of the third polarization state. The error between the symbol sequence after differential polarization time decoding and the symbol sequence before differential polarization time encoding corresponding to the first polarization state and the difference between the symbol sequence after differential polarization time decoding corresponding to the fourth polarization state and the second polarization state The error of the symbol sequence before polarization time encoding; based on the first symbol error and the chain rule, determine the second symbol error, the second symbol error is used to characterize the symbol sequence before differential polarization time decoding corresponding to the third polarization state The error between the symbol sequence after differential polarization time encoding corresponding to the first polarization state and/or the symbol sequence before differential polarization time encoding corresponding to the fourth polarization state and the differential polarization time encoding corresponding to the second polarization state. The error of the symbol sequence; based on the second symbol error, update the tap coefficient used for the SISO equalization process; wherein the first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier polarization state.

In the solution shown in this application, the symbol error after differential polarization time decoding is first calculated, and then the error after differential polarization time decoding and the chain rule are used to calculate the error before differential polarization time decoding. Based on the error before differential polarization time decoding, the tap coefficients used in the SISO equalization processing are updated, making the tap coefficients of the SISO equalization processing more accurate, thereby making the data recovered by the single-polarization receiver more accurate.

In one example, the method further includes: obtaining a constellation diagram of the symbol sequence corresponding to the fourth polarization state, determining a rotation angle corresponding to the constellation diagram, and obtaining a frequency offset value corresponding to the first optical carrier based on the rotation angle. ; Based on the frequency offset value, update the frequency offset value used for frequency offset compensation processing.

In the solution shown in this application, the constellation diagram of the symbol sequence corresponding to the recovered polarization state is used to determine the frequency offset value. Using this frequency offset value, the frequency offset value used for frequency offset compensation processing is updated. In this way, the frequency offset value can be updated during the data receiving process, making the data recovered by the single-polarization receiver more accurate.

In one example, before acquiring the first complex digital signal of the third polarization state corresponding to the first optical carrier, the method further includes: acquiring the second complex digital signal of the third polarization state corresponding to the first optical carrier, the third Two complex digital signals are received before the first complex digital signal, and the second complex digital signal includes a differential polarization time-encoded training sequence corresponding to the third polarization state; determining the frame in the second complex digital signal head position; based on the frame head position and the length of the differential polarization time-encoded training sequence corresponding to the third polarization state, in the second complex digital signal, obtain the second received sequence corresponding to the training sequence; based on the third polarization state The second receiving sequence and the Alamouti coding rule are used to recover the third receiving sequence corresponding to the fourth polarization state, and the third receiving sequence is the training sequence corresponding to the fourth polarization state; perform the second receiving sequence and the third receiving sequence. Differential polarization time decoding, obtaining the decoded training sequence corresponding to the third polarization state and the fourth polarization state respectively; determining the decoded training sequence corresponding to the third polarization state and the differential polarization time code corresponding to the first polarization state The error of the previous training sequence and the decoded training sequence corresponding to the fourth polarization state and the error of the training sequence before differential polarization time encoding corresponding to the second polarization state are obtained to obtain a third symbol error, the first polarization state and the The second polarization state is the two orthogonal polarization states used when transmitting the first optical carrier; based on the third symbol error and the chain rule, a fourth symbol error is determined, and the fourth symbol error is used to characterize the second reception The error of the differential polarization time-coded training sequence corresponding to the first polarization state and/or the error of the differential polarization time-coded training sequence corresponding to the third received sequence and the second polarization state; based on the fourth The symbol error updates the tap coefficient used for this SISO equalization process.

In the solution shown in this application, a known training sequence is used to pre-converge the tap coefficients used at the beginning of the SISO equalization process, so that the initial tap coefficients are more accurate.

In a third aspect, this application provides a device for data transmission, which device is used in a coherent optical communication system. The device includes one or more units for realizing the data described in the first aspect or the examples of the first aspect. method of sending.

In a fourth aspect, this application provides a data receiving device, which is used in a coherent optical communication system. The device includes one or more units for realizing the data described in the second aspect or examples of the second aspect. Receive method.

In a fifth aspect, the application provides a dual-polarization transmitter, which includes a digital signal processor, a digital-to-analog converter, a modulation driver, a laser, and a dual-polarization modulator; the digital signal processor is used to execute The data sending method described in the first aspect or an example of the first aspect.

In a sixth aspect, the application provides a single polarization receiver, characterized in that the single polarization receiver includes a local oscillator laser, a coupler, a detector, a transimpedance amplifier, an analog-to-digital converter, and a digital signal processor. ;

The digital signal processor is configured to perform the data receiving method described in the second aspect or an example of the second aspect.

Description of the drawings

Figure 1 is a schematic structural diagram of a PON based on point to multipoint (PtMP) provided by an exemplary embodiment of the present application;

Figure 2 is an architectural schematic diagram of single polarization heterodyne reception provided by an exemplary embodiment of the present application;

Figure 3 is a schematic structural diagram of a dual-polarization transmitter provided by an exemplary embodiment of the present application;

Figure 4 is a schematic structural diagram of a digital signal processor (DSP) in a dual-polarization transmitter provided by an exemplary embodiment of the present application;

Figure 5 is a schematic structural diagram of a DSP in a dual-polarization transmitter provided by an exemplary embodiment of the present application;

Figure 6 is a schematic structural diagram of a single polarization heterodyne receiver provided by an exemplary embodiment of the present application;

Figure 7 is a schematic structural diagram of a single polarization heterodyne receiver provided by an exemplary embodiment of the present application;

Figure 8 is a schematic structural diagram of a DSP in a single-polarization heterodyne receiver provided by an exemplary embodiment of the present application;

Figure 9 is a schematic structural diagram of a DSP in a single polarization heterodyne receiver provided by an exemplary embodiment of the present application;

Figure 10 is an architectural schematic diagram of single polarization homodyne reception provided by an exemplary embodiment of the present application;

Figure 11 is a schematic structural diagram of a single polarization homodyne receiver provided by an exemplary embodiment of the present application;

Figure 12 is a schematic structural diagram of a DSP in a single-polarization homodyne receiver provided by an exemplary embodiment of the present application;

Figure 13 is a schematic structural diagram of a DSP in a single-polarization homodyne receiver provided by an exemplary embodiment of the present application;

Figure 14 is a schematic structural diagram of a single polarization single sideband receiver provided by an exemplary embodiment of the present application;

Figure 15 is a schematic structural diagram of a single polarization single sideband receiver provided by an exemplary embodiment of the present application;

Figure 16 is a schematic structural diagram of a DSP in a single-polarization single-sideband receiver provided by an exemplary embodiment of the present application;

Figure 17 is a schematic structural diagram of a DSP in a single-polarization single-sideband receiver provided by an exemplary embodiment of the present application;

Figure 18 is a schematic flowchart of a data sending method provided by an exemplary embodiment of the present application;

Figure 19 is a schematic diagram of a dual-cycle training sequence provided by an exemplary embodiment of the present application;

Figure 20 is a schematic structural diagram of a data frame provided by an exemplary embodiment of the present application;

Figure 21 is a schematic flowchart of a data receiving method provided by an exemplary embodiment of the present application;

Figure 22 is a schematic diagram of the impact of the residual frequency offset value on differential polarization time decoding provided by an exemplary embodiment of the present application;

Figure 23 is a schematic diagram of system performance simulation provided by an exemplary embodiment of the present application when the differential polarization time decoding and dynamic update of tap coefficients of the DSP in a single polarization receiver are enabled;

Figure 24 is a schematic diagram of the optical signal to noise ratio (OSNR)-bit error rate (bit error rate/ratio, BER) performance curve of the system provided by an exemplary embodiment of the present application;

Figure 25 is a schematic diagram of the frequency offset value after the frequency offset tracking algorithm is turned on and the frequency offset tracking algorithm is turned off according to an exemplary embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

The application scenarios of the embodiments of this application are described below.

In the PtMP network based on the coherent optical communication system, it can be considered that the leaf nodes use single polarization receivers to receive to reduce the cost of the leaf nodes. However, if the central node transmits data in a single polarization state through a single polarization transmitter, the leaf nodes receive through single polarization. The machine receives data from a single polarization state. However, since optical fiber transmission will introduce polarization rotation, as long as the polarization states of the receiving end and the transmitting end are not aligned, the signal-to-noise ratio of the single-polarization receiver will be reduced. The performance of the single-polarization receiver is related to the rotation coefficient of the polarization state. In extreme cases, if the polarization states of the receiving end and the transmitting end are orthogonal, then the single-polarization receiver cannot receive data at all.

Since polarization rotation is uncontrollable, it is hoped that the performance of a single-polarization receiver will be independent of the polarization state when receiving the signal, but at the same time, the transmission rate will not be lost. Therefore, consider that the central node uses a dual-polarization transmitter, the leaf nodes use a single-polarization receiver, the center node transmits two polarization states through the dual-polarization receiver, and the leaf nodes receive a single polarization state signal through a single-polarization receiver, using a single polarization state signal, and the signals of two polarization states are recovered. In this way, the single polarization receiver at the leaf node still only requires one digital to analog converter (DAC) and one analog to digital converter (analog to digital converter, ADC). Although the dual polarization receiver at the center node requires four DAC and ADC, but it serves multiple leaf nodes, so even if the central node includes multi-channel DAC and multi-channel ADC, the cost of the central node can be shared by multiple leaf nodes.

In this embodiment of the present application, the PtMP network may be a PON network, the central node may be an OLT, and the leaf nodes may be an optical network terminal (optical network terminal, ONT) or ONU.

The following describes the system architecture, the method flow of data sending, and the method flow of data receiving in order.

The system architecture of the embodiment of the present application is first described below.

1. The PON network includes a central node OLT, a passive optical splitter and multiple leaf nodes ONT or ONU.

In the PON based on the coherent optical communication system, based on the different networking methods of optical transceivers, there are architectures based on point to point (PtP) networking and PtMP networking.

For example, consider a central node and N leaf nodes. In an architecture based on PtP networking, the central node requires N optical transceivers, and each leaf node requires an optical transceiver. A total of 2N optical transceivers, 2N PtP link.

In an architecture based on PtMP networking, the central node only needs a large-bandwidth optical transceiver to send and receive multiple subcarriers. Each leaf node needs an optical transceiver, and each leaf node uses different subcarriers, totaling N +1 optical transceiver. In the downlink based on the PtMP networking architecture, the subcarriers corresponding to multiple leaf nodes are distinguished in the digital dimension. Multiple leaf nodes use corresponding local oscillator optical receivers to receive the corresponding See Figure 1 for subcarriers.

2. In PON based on coherent optical communication systems, depending on the reception methods of single-polarization receivers, there are single-polarization heterodyne reception architectures, single-polarization homodyne reception architectures, and single-polarization single sideband. SSB) receiving architecture, that is to say, single polarization receivers are single polarization heterodyne receivers, single polarization homodyne receivers and single polarization single sideband receivers.

Among them, (1) Single polarization heterodyne receiving architecture.

Referring to the single-polarization heterodyne receiving architecture shown in Figure 2, the architecture includes a central node, passive beam splitter and multiple leaf nodes. In Figure 2, the architecture based on PtMP networking is used as an example. In Figure 2, only four leaf nodes (leaf node_1 to leaf node_4) are shown. The solid border triangle represents the subcarriers used. , the dotted border triangle indicates the subcarriers vacated by heterodyne reception, and the dotted arrow indicates the position of the center frequency of the local oscillator laser of each leaf node, They are LO_1 to LO_4 respectively, f0 represents the position of the center frequency of the laser of the central node, and G* represents the spectrum corresponding to the time domain conjugate of subcarrier G (G=1, 2, 3, 4). For the scenario with four leaf nodes in Figure 2, the central node supports the modulated bandwidth of 8 subcarriers, but only uses four of the subcarriers to carry data, and the remaining four subcarriers do not modulate any data. After the single polarization receiver of the leaf node performs heterodyne reception, it will generate a spectrum corresponding to the time domain conjugate of the target subcarrier signal at the vacant subcarrier position. The target subcarrier is the subcarrier carrying data. Since the spectrum does not alias with the spectrum of the target subcarrier, the data carried on the target subcarrier can be recovered through frequency down conversion.

(1.1)Dual polarization transmitter

In the architecture shown in Figure 2, the structural diagram of the dual-polarization transmitter in the central node is shown in Figure 3. The dual-polarization transmitter includes a DSP, a digital-to-analog converter, a modulator driver, a laser, and a dual-polarization modulator. The dual-polarization modulator The device is a dual polarization IQ modulator.

The DSP processes the data to be sent corresponding to each subcarrier to obtain a dual-polarization complex digital signal corresponding to each subcarrier. The dual-polarization complex digital signal corresponding to each subcarrier includes four digital signals, each of which is channel I of the first polarization state. signal, the Q-channel signal of the first polarization state, the I-channel signal of the second polarization state, and the Q-channel signal of the second polarization state, respectively represented as XI, pay. The four digital signals corresponding to each subcarrier are converted into four analog signals through a digital-to-analog converter, and different digital signals are input to different digital-to-analog converters. The four analog signals corresponding to each subcarrier are loaded into the four electrical signal input ports of a dual-polarization IQ modulator through four modulator drivers. The laser inputs an optical signal with a center frequency of f0 to the dual-polarization IQ modulator. The dual-polarization IQ The modulator modulates the four analog signals corresponding to each subcarrier onto the optical signal to obtain the modulated subcarriers for data to be sent.

The dual-polarization IQ modulator includes a polarization beam splitter, two IQ modulators and a polarization coupler. The polarization coupler can also be called a polarization beam combiner. The polarization beam splitter divides the optical signal output from the laser into the optical signal in the first polarization state and the second polarization state, and outputs them to the two IQ modulators respectively. The first IQ modulation modulates the two analog signals of the first polarization state onto the optical signal of the first polarization state and outputs them to the polarization coupler. The second IQ modulation modulates the two analog signals of the second polarization state onto the optical signal of the second polarization state and outputs them to the polarization coupler. The polarization coupler combines the two polarized signal lights and outputs them.

In an example, in a dual-polarization transmitter, when the PON adopts an architecture based on PtMP networking and an architecture based on PtP networking, the structure of the DSP is different, which will be described separately below.

When the PON adopts an architecture based on PtMP networking, it is assumed that the PON includes N leaf nodes, and the DSP of the dual-polarization transmitter includes N FEC coding units, N bit-symbol mapping units, N selection units, and N differential polarization units. For the time encoding unit, N filter units and digital subcarrier multiplexing unit, see the DSP shown in Figure 4.

Among them, each subcarrier corresponds to a forward error correction (FEC) coding unit, a bit-symbol mapping unit, a selection unit, a differential polarization time coding unit and a filtering unit. The FEC coding unit, a A bit-to-symbol mapping unit, a selection unit, a differential polarization time encoding unit and a filtering unit are connected in sequence.

For the data to be sent corresponding to subcarrier 1, the data to be sent can be considered as a bit stream. The FEC encoding unit encodes the data to be sent, and outputs the encoded signal to the bit-symbol mapping unit. The bit-symbol mapping unit The encoded signal is subjected to bit-to-symbol mapping to obtain a symbol sequence, which is output to the selection unit. The mapping is based on a preset modulation format mapping rule. The preset modulation format can be quadrature phase shift keying (quadrature phase shift keying). QPSK) or 16-ary quadrature amplitude modulation (16-QAM), etc. The selection unit can select according to the length of the data frame whether to output the symbol sequence corresponding to the data to be sent to the differential polarization time encoding unit, or to output the symbol sequence of the training sequence corresponding to the two polarization states to the differential polarization time encoding unit. . The differential polarization time encoding unit performs differential polarization time encoding on the input symbol sequence, and encodes the differential polarization time The signal of Pulse shaping, digital pre-emphasis processing, IQ damage pre-compensation and dispersion pre-compensation, etc.

The digitally filtered signal corresponding to each subcarrier will be output to the digital subcarrier multiplexing unit. The digital subcarrier multiplexing unit will process the received signal into a dual-polarization complex digital signal to be transmitted by the digital-to-analog converter.

When the PON adopts an architecture based on PtP networking, the DSP includes the FEC coding unit, bit-symbol mapping unit, selection unit, differential polarization time coding unit and filtering unit. See the DSP shown in Figure 5, FEC coding unit, bit- The symbol mapping unit, selection unit, differential polarization time encoding unit and filtering unit are connected in sequence.

Among them, for the data to be sent corresponding to carrier 1, the data to be sent can be considered as a bit stream. The FEC encoding unit encodes the data to be sent, and outputs the encoded signal to the bit-symbol mapping unit. The bit-symbol mapping unit Perform bit-to-symbol mapping on the encoded signal to obtain a symbol sequence, and output the symbol sequence to the selection unit. The mapping is based on a preset modulation format mapping rule. The preset modulation format can be QPSK or 16-QAM, etc. The selection unit can select according to the length of the data frame whether to output the symbol sequence corresponding to the data to be sent to the differential polarization time encoding unit, or to output the symbol sequence of the dual-period training sequence corresponding to the two polarization states to the differential polarization time. coding unit. The differential polarization time encoding unit performs differential polarization time encoding on the input symbol sequence, and outputs the differential polarization time encoded signal to the filter unit. The filter unit performs digital filtering on the baseband differential polarization time encoded signal corresponding to carrier 1. , obtain the dual-polarization complex digital signal corresponding to carrier 1.

(1.2) Single polarization receiver

In the architecture shown in Figure 2, a structural diagram of a single polarization heterodyne receiver in a leaf node is shown in Figure 6. The single-polarization heterodyne receiver includes a local oscillator laser, a 2×2 coupler, a balanced optical detector, a DC blocking capacitor, a transimpedance amplifier (TIA), an analog-to-digital converter and a DSP.

A single-polarization heterodyne receiver receives optical signals in two polarization states. The local oscillation laser outputs the local oscillation light of a single polarization state to the 2×2 coupler, and the optical signals of the two polarization states and the local oscillation light of the single polarization state are input to the 2×2 coupler. The 2×2 coupler couples the signal light of two polarization states with the local oscillator light of a single polarization state, and outputs the two optical signals to the balanced light detector. The balanced light detector performs photoelectric conversion on two optical signals and outputs an electrical analog signal. The DC blocking capacitor isolates the DC component of the electrical analog signal and outputs it to the TIA. TIA amplifies the electrical analog signal of the isolated DC component, and then obtains the digital signal through the analog-to-digital converter. DSP processes the digital signal to obtain the data sent by the dual-polarization transmitter.

It should be noted that, assuming that the polarization state of the local oscillation light is the third polarization state, only the component of the third polarization state of the two polarization optical signals produces a beating frequency with the local oscillation light, and the components of the other polarization states There is no beat frequency with the local oscillation light, and the third polarization state is orthogonal to the fourth polarization state. After passing through the balanced light detector, the self-timer frequency of the local oscillator light, the self-timer frequency of the optical signal component in the third polarization state, and the self-timer frequency of the optical signal component in the fourth polarization state are offset by the balanced light detector, leaving only the third polarization state. The optical signal component of is proportional to the beat frequency of the local oscillator light, and this term is proportional to the optical signal component of the third polarization state. The 2×2 coupler is a polarization-independent coupler, that is, for the third polarization state and the fourth polarization state, the 2×2 coupler behaves as a 3dB power splitter. This type of 2×2 coupler can be based on Optical fiber coupler. This is not necessarily true for 2×2 optical couplers with integrated photonic devices on-chip. For example, silicon-based photonic devices often exhibit polarization-sensitive characteristics due to the unequal width and height of single-mode waveguides. Therefore, single-polarization heterodyne receivers must be When the optical part is integrated on-chip, the structure of a single-polarization heterodyne receiver shown in Figure 7 can be used.

Referring to Figure 7, the single-polarization heterodyne receiver includes a local oscillator laser, a polarization beam splitter, a single-polarization 2×2 coupler, a balanced light detector, a DC blocking capacitor, a transimpedance amplifier, an analog-to-digital converter, and a DSP. Optionally, polarization beam splitters, single polarization 2×2 couplers and balanced light detectors can be integrated on-chip.

A single-polarization heterodyne receiver receives optical signals in two polarization states. The local oscillator laser outputs local oscillator light in a single polarization state to 2×2 coupler, assuming that the single polarization state is the third polarization state. The optical signals of two polarization states are input to the polarization beam splitter. The polarization beam splitter obtains an optical signal of the same third polarization state from the two polarization states of optical signals and outputs it to a single polarization 2×2 coupler. The polarization beam splitter If the beam splitter does not output another optical signal, it can also be considered that the other optical signal is cut off. The single polarization 2×2 coupler couples the two input optical signals and outputs the two optical signals to the balanced light detector. The balanced light detector performs photoelectric conversion on two optical signals and outputs an electrical analog signal. The DC blocking capacitor isolates the DC component of the electrical analog signal and outputs it to the transimpedance amplifier. The transimpedance amplifier amplifies the electrical analog signal with the isolated DC component, and then obtains the digital signal through the analog-to-digital converter. DSP processes the digital signal to obtain the data sent by the dual-polarization transmitter.

It should be noted that in the single-polarization heterodyne receiver shown in Figure 7, the polarization components of the two polarization optical signals that are orthogonal to the polarization state of the local light are originally canceled during the balanced detection process. , so using a polarization beam splitter to filter out the orthogonal polarization component will not introduce additional power loss in addition to the insertion loss of the polarization beam splitter itself.

In one example, when a single polarization heterodyne receiver is deployed and put online, considering the large frequency deviation of the optical signal, it is necessary to introduce a double period structure into the training sequence before differential polarization time encoding, so that the training sequence after differential polarization time encoding There is still a dual-period structure, and the receiving end can use the receiving-end sequence corresponding to the dual-period training sequence to complete the frame fixation and estimate the frequency offset under the condition that there is a large frequency offset. That is, the two polarization states respectively correspond to the training sequence before differential polarization time encoding. The training sequence corresponding to each polarization state includes a subsequence of two periods. For each polarization state, the two periods are different. In the embodiment of this application, "frequency offset" refers to frequency offset, which refers to the offset between the frequency of the optical signal emitted by the transmitting end and the center frequency of the local oscillator light of the receiving end. In this embodiment of the present application, the receiving end is a leaf node, and the transmitting end is a central node.

In this example, the DSP in the single-polarization heterodyne receiver includes a frequency shift unit, a matched filter unit, a clock recovery unit, a fixed frame unit, a first frequency offset estimation unit, a frequency offset compensation unit, a SISO equalization unit, and differential polarization time decoding. unit, symbol-bit mapping unit, FEC decoding unit, second frequency offset estimation unit and tap coefficient update unit, see Figure 8. Among them, the DSP receives the real digital signal output by the analog-to-digital converter in the single-polarization heterodyne receiver. The real digital signal is an intermediate frequency signal, not a baseband signal. The frequency shifting unit performs frequency shift processing on the real digital signal to obtain a baseband signal. Output the baseband signal to the matched filter unit. The matched filter unit filters out the beat frequency of other subcarriers and local oscillator light, as well as the conjugate image of the signal introduced by heterodyne detection, so that the leaf node only receives a single subcarrier, and outputs the processed signal to the clock recovery unit. The recovery unit performs clock recovery on the processed signal.

The signal output by the clock recovery unit will be input into the frame fixing unit to determine the frame header position of the data frame. The frame fixing unit uses the correlation operation between the receiving end sequence and its own delayed sequence to determine the frame header position of the data frame. The first frequency offset estimation unit determines the initial frequency offset value based on the bi-periodic characteristics of the training sequence. The initial frequency offset value of the first frequency offset estimation unit can be fully output to the frequency offset compensation unit, or part of it can be output to the frequency offset compensation unit, and the other part can be output to the local oscillator laser. That is to say, the DSP can output a part of the frequency offset value. A control circuit is provided for the local oscillation laser, and the control circuit adjusts the temperature and/or current of the local oscillation laser, thereby controlling the center frequency of the local oscillation light output by the local oscillation laser. After the single-polarization heterodyne receiver is deployed and online, the framing unit and the first frequency offset estimation unit can stop working, and the clock recovery unit directly outputs the clock-recovered signal to the Frequency offset compensation unit. The frequency offset compensation unit is used to perform frequency offset compensation on the input signal. The SISO equalization unit is used to equalize the signal after frequency offset compensation. The differential polarization time decoding unit is used to perform differential polarization time decoding on the SISO equalized signal and recover the symbol sequence of the two polarization states. The symbol-bit mapping unit is used to map symbol sequences of two polarization states into bit streams. The FEC decoding unit is used to FEC decode the bit stream into data sent by the central node.

In another example, by default when a single-polarization heterodyne receiver is deployed and online, considering that the optical signal frequency deviation is small, the training sequence before differential polarization time encoding corresponding to the two polarization states can be arbitrary autocorrelation. Better sequence.

In this example, the DSP in the single-polarization heterodyne receiver includes a frequency shift unit, a matched filter unit, a clock recovery unit, a fixed frame unit, a frequency offset compensation unit, a SISO equalization unit, a differential polarization time decoding unit, and a symbol-bit mapping unit. unit, FEC decoding unit, frequency offset estimation unit and tap coefficient update unit, see Figure 9.

Among them, the DSP receives the real digital signal output by the analog-to-digital converter in the single-polarization heterodyne receiver. The frequency shifting unit performs frequency shift processing on the real digital signal to obtain a baseband signal, and outputs the baseband signal to the matched filtering unit. The matched filter unit filters out the beat frequency of other subcarriers and local oscillator light, as well as the conjugate image of the signal introduced by heterodyne detection, and outputs the processed signal to the clock recovery unit, which clocks the processed signal. recover.

The signal output by the clock recovery unit is input into the frame fixing unit to determine the frame header of the data frame. The frame-fixing unit uses the receiver sequence and the locally stored differential polarization time-encoded training sequence to perform correlation operations to determine the frame header position of the data frame. After the single-polarization heterodyne receiver is deployed and online, the frame-fixing unit can stop working and the clock recovers. The unit outputs the clock recovered signal directly to the frequency offset compensation unit without passing through the framing unit.

The frequency offset compensation unit is used to perform frequency offset compensation on the input signal. The SISO equalization unit is used to equalize the signal after frequency offset compensation. The differential polarization time decoding unit is used to perform differential polarization time decoding on the SISO equalized signal and recover the symbol sequence of the two polarization states. The symbol-bit mapping unit is used to map symbol sequences of two polarization states into bit streams. The FEC decoding unit is used to FEC decode the bit stream into data sent by the central node.

(2) Architecture of single polarization homodyne reception.

Refer to the architecture of single-polarization homodyne reception shown in Figure 10. The architecture includes a central node, a passive splitter and multiple leaf nodes. In Figure 10, only four leaf nodes are shown, and the solid triangle box represents the subcarrier. , the dotted arrow represents the position of the center frequency of the local oscillator laser of each leaf node, f0 represents the position of the center frequency of the transmitting end laser of the central node, LO_1 to LO_4 respectively represent the local frequencies corresponding to leaf node_1 to leaf node_4. The frequency of vibrating light. For the scenario with four leaf nodes in Figure 10, the central node supports the modulation bandwidth of four subcarriers. The single polarization receiver of the leaf node performs single polarization homodyne reception, receives a single subcarrier, and recovers data on a single subcarrier.

In the architecture shown in Figure 10, the architecture based on PtMP networking is used as an example for explanation. The architecture of single polarization homodyne reception shown in Figure 10 can also be applied to the architecture based on PtP networking.

In the architecture shown in Figure 10, the structural diagram of the dual-polarization transmitter in the central node is shown in Figure 3, which will not be described again here.

In the architecture shown in Figure 10, a schematic structural diagram of a single polarization homodyne receiver in a leaf node is shown in Figure 11. The single polarization homodyne receiver includes a local oscillator laser, a polarization beam splitter, a single polarization 2×4 coupler, a first balanced light detector, a second balanced light detector, a first DC blocking capacitor, and a second DC blocking capacitor. , a first transimpedance amplifier, a second transimpedance amplifier, a first analog-to-digital converter, a second analog-to-digital converter and a DSP.

A single-polarization homodyne receiver receives optical signals in two polarization states. The local oscillator laser outputs the local oscillation light of a single polarization state to a single polarization 2×4 coupler, assuming that the single polarization state is the third polarization state. The two polarization optical signals are input to the polarization beam splitter. The polarization beam splitter obtains an optical signal that is the same as the third polarization state from the two polarization optical signals and outputs it to a single polarization 2×4 coupler. If the beam splitter does not output another optical signal, it can also be considered that the other optical signal is cut off. The single polarization 2×4 coupler couples the two input optical signals, outputs the two optical signals to the first balanced optical detector, and the other two optical signals to the second balanced optical detector. The first balanced light detector performs photoelectric conversion on the two optical signals and outputs a first electrical analog signal. The first DC blocking capacitor isolates the DC component of the first electrical analog signal and outputs it to the first transimpedance amplifier. The first transimpedance amplifier amplifies the first electrical analog signal of the isolated DC component, and then obtains the first digital signal through the first analog-to-digital converter. The second balanced light detector performs photoelectric conversion on the two optical signals and outputs a second electrical analog signal. The second DC blocking capacitor isolates the DC component of the second electrical analog signal and outputs it to the second transimpedance amplifier. The second transimpedance amplifier amplifies the second electrical analog signal of the isolated DC component, and then obtains the second digital signal through the second analog-to-digital converter. The DSP processes the first digital signal and the second digital signal to obtain data sent by the dual-polarization transmitter.

In an example, when the single polarization receiver is a single polarization homodyne receiver, the DSP includes a merging unit, a matched filtering unit, a clock recovery unit, a framing unit, a first frequency offset estimation unit, a frequency offset compensation unit, a SISO Balanced unit, differential For the polarization time decoding unit, symbol-bit mapping unit, FEC decoding unit, second frequency offset estimation unit and tap coefficient update unit, see Figure 12. This example corresponds to the situation where the training sequence has a double period structure.

In another example, when the single polarization receiver is a single polarization homodyne receiver, the DSP includes a merging unit, a matched filter unit, a clock recovery unit, a framing unit, a frequency offset compensation unit, a SISO equalization unit, and a differential polarization time Decoding unit, symbol-bit mapping unit, FEC decoding unit, frequency offset estimation unit and tap coefficient update unit, see Figure 13. This example corresponds to a situation where the training sequence only has good autocorrelation and does not require a double period structure.

Among them, two channels of real digital signals are input to the DSP, and the two channels of real data signals are respectively the first real digital signal and the second real digital signal. The merging unit is used to combine two real digital signals into one complex signal and output it to the matched filtering unit. The subsequent process is the same as the DSP processing process in a single-polarization heterodyne receiver, and will not be described again here.

It should be noted that the reason why the DSP of the single-polarization homodyne receiver does not include a frequency shifting unit is that the signal received by the single-polarization homodyne receiver is at the baseband, and no frequency shifting operation is required.

(3) Architecture of single-polarization single-sideband reception.

The architecture of single-polarization single-sideband reception is the same as that of single-polarization heterodyne reception. See Figure 2 for a detailed description. See Figure 3 for the structure of a dual-polarization transmitter, which will not be described again here.

The single-polarization single-sideband receiving architecture can be applied to the PtMP networking architecture or the PtP networking architecture.

A schematic structural diagram of a single-polarization single-sideband receiver in a leaf node is shown in Figure 14. The single-polarization single-sideband receiver includes a local oscillator laser, a 2×2 coupler, a photodetector, a DC blocking capacitor, a transimpedance amplifier, an analog-to-digital converter, and a DSP.

A single-polarization single-sideband receiver receives optical signals in two polarization states. The local oscillation laser outputs the local oscillation light of a single polarization state to the 2×2 coupler, and the optical signals of the two polarization states and the local oscillation light of the single polarization state are input to the 2×2 coupler. The 2×2 coupler couples the signal light of two polarization states with the local oscillator light of a single polarization state, divides it into two optical signals, outputs one optical signal to the optical detector, and does not output the other optical signal. The photodetector performs photoelectric conversion on an optical signal and outputs an electrical analog signal. The DC blocking capacitor isolates the DC component of the electrical analog signal and outputs it to the transimpedance amplifier. The transimpedance amplifier amplifies the electrical analog signal with the isolated DC component, and then obtains the digital signal through the analog-to-digital converter. DSP processes the digital signal to obtain the data sent by the dual-polarization transmitter.

It should be noted that, assuming that the local oscillation light is in the third polarization state, only the component of the third polarization state of the two polarization optical signals produces a beat frequency with the local oscillator light, and the optical signal component of the fourth polarization state has a beating frequency with the local oscillator light. Light does not produce a beat frequency, and the third polarization state is orthogonal to the fourth polarization state. After passing through the balanced light detector, the Selfie frequency of the local oscillator light, the light signal component of the third polarization state, and the self-timer frequency of the light signal component of the fourth polarization state are offset by the balanced light detector, leaving only the light signal of the third polarization state. The component is proportional to the beat frequency of the local oscillator light, and this term is proportional to the optical signal component of the third polarization state. The 2×2 coupler is a polarization-independent coupler, that is, for the third polarization state and the fourth polarization state, the 2×2 coupler behaves as a 3dB power splitter. This type of 2×2 coupler can be based on Optical fiber coupler. This is not necessarily true for 2×2 optical couplers with integrated photonic devices on-chip. For example, silicon-based photonic devices often exhibit polarization-sensitive characteristics due to the unequal width and height of single-mode waveguides, so single-polarization single-sideband reception is required. When the optical part of the receiver is integrated on-chip, the structure of a single-polarization single-sideband receiver shown in Figure 15 can be used.

Referring to Figure 15, the single-polarization single-sideband receiver includes a local oscillator laser, a polarization beam splitter, a single-polarization 2×2 coupler, a photodetector, a DC blocking capacitor, a transimpedance amplifier, an analog-to-digital converter, and a DSP.

A single-polarization single-sideband receiver receives optical signals in two polarization states. The local oscillator laser outputs the local oscillation light of a single polarization state to the 2×2 coupler, assuming that the single polarization state is the third polarization state. The two polarization optical signals are input to the polarization beam splitter. The polarization beam splitter obtains an optical signal that is the same as the third polarization state from the two polarization optical signals and outputs it to the single polarization 2×2 coupler. If the beam splitter does not output another optical signal, it can also be considered that the other optical signal is cut off. The single polarization 2×2 coupler couples the two input optical signals into two optical signals, outputs one optical signal to the photodetector, and does not output the other. A light signal. The photodetector performs photoelectric conversion on the two optical signals and outputs an electrical analog signal. The DC blocking capacitor isolates the DC component of the electrical analog signal and outputs it to the transimpedance amplifier. The transimpedance amplifier amplifies the electrical analog signal with the isolated DC component, and then obtains the digital signal through the analog-to-digital converter. DSP processes the digital signal to obtain the data sent by the dual-polarization transmitter.

It should be noted that after detection by the photodetector, the electrical analog signal obtained includes the beat frequency term of the local oscillator light and the local oscillator light, the beat frequency term of the local oscillator light and the received optical signal, and the received optical signal. The signal-signal beating interference (SSBI) item with the optical signal. Among them, the beat frequency term of the local oscillator light and the local oscillator light will be filtered by the DC blocking capacitor. The beat frequency term of the local oscillator light and the received optical signal is the term that you want to keep, and the SSBI term is the term you want to eliminate.

In an example, when the single-polarization receiver is a single-polarization single-sideband receiver, the DSP includes a single-sideband signal recovery unit, a frequency shifting unit, a matched filtering unit, a clock recovery unit, a framing unit, and a first frequency offset unit. Estimation unit, frequency offset compensation unit, SISO equalization unit, differential polarization time decoding unit, symbol-bit mapping unit, FEC decoding unit, second frequency offset estimation unit and tap coefficient update unit, see Figure 16. This example corresponds to the situation where the training sequence has a double period structure.

In another example, when the single-polarization receiver is a single-polarization single-sideband receiver, the DSP includes a single-sideband signal recovery unit, a frequency shifting unit, a matched filtering unit, a clock recovery unit, a framing unit, and a frequency offset compensation unit. unit, SISO equalization unit, differential polarization time decoding unit, symbol-bit mapping unit, FEC decoding unit, frequency offset estimation unit and tap coefficient update unit, see Figure 17. This example corresponds to a situation where the training sequence only has good autocorrelation and does not require a double period structure.

In this example, the single sideband signal recovery unit is used to recover the single sideband signal without SSBI terms.

In PON, the power of the optical signal received by the single-polarization single-sideband receiver is generally much smaller than the power of the local oscillator light, so the SSBI term is much smaller than the local oscillator light - the beat frequency term of the received optical signal, and the SSBI term It is likely to be submerged in the single-polarization single-sideband receiver noise. In this case, there will be no performance impact if the single-sideband signal recovery unit is not used. Therefore, the DSP of a single-polarization single-sideband receiver can be exactly the same as that of a single-polarization heterodyne receiver, and the single-sideband signal recovery unit is an optional unit.

The following describes the data transmission method in the embodiment of the present application. The method is executed by the dual-polarization transmitter, specifically by the DSP in the dual-polarization transmitter. See step 1801 to step 1804 in Figure 18.

Step 1801: Obtain the symbol sequence corresponding to the first optical carrier and the training sequences corresponding to the two polarization states. The two polarization states are orthogonal. The symbol sequence is obtained by performing symbol mapping on the data to be sent corresponding to the first optical carrier.

Among them, when the PON adopts an architecture based on PtP networking, the first optical carrier is a single carrier. When PON adopts an architecture based on PtMP networking, the first optical carrier is a single subcarrier.

In this embodiment, when applied to an architecture based on PtMP networking, the dual-polarization transmitter can obtain the symbol sequence corresponding to the first optical carrier according to the process shown in Figure 4, which is the output of the bit-symbol mapping unit result. When applied to an architecture based on PtP networking, the dual-polarization transmitter can obtain the symbol sequence corresponding to the first optical carrier according to the process shown in Figure 5, which is the output result of the bit-symbol mapping unit. In addition, the dual-polarization transmitter can also obtain training sequences corresponding to the two polarization states. The training sequence obtained by the dual-polarization transmitter is also a symbol sequence.

The symbol sequence corresponding to the first optical carrier is obtained by performing FEC encoding and symbol mapping on the data to be sent corresponding to the first optical carrier.

The training sequence is used for single-polarization receivers to achieve pre-convergence of equalization coefficients and determine the frame header position of the data frame. The training sequences corresponding to the two polarization states can be preset and are known in both dual-polarization transmitters and single-polarization receivers. The training sequence may be a symbol sequence obtained based on QPSK or a symbol sequence obtained based on QAM.

As mentioned above, in one example, the training sequences corresponding to the two polarization states are any sequences with good autocorrelation.

In another example, the training sequence introduces a two-period structure. The two-period sequence can be used to achieve pre-collection of equalization coefficients Convergence, frequency offset estimation in a wide range, and determining the frame header position of the data frame under the condition of large frequency offset.

Assume that the two polarization states include a first polarization state and a second polarization state. The dual-period training sequence corresponding to the first polarization state includes two periods of training sequences. The lengths of the two periods are different, and the number of repetitions may be the same or different. The dual-period training sequence corresponding to the second polarization state also includes a training sequence of two periods. The lengths of the two periods are different, and the number of repetitions may be the same or different. The period length and repetition number of the training sequence of the first period corresponding to the first polarization state and the second polarization state are the same, and the period length and repetition number of the training sequence of the second period corresponding to the first polarization state and the second polarization state are the same. The times are the same. The period and number of repetitions of the dual-period training sequence can be set according to actual needs. For example, they can be set according to the period and number of repetitions required for a single polarization receiver to correctly decode data. For example, referring to FIG. 19 , a dual-period training sequence before differential polarization time encoding corresponding to two polarization states is shown. The dual-period training sequence corresponding to the first polarization state may be the A training sequence and the C training sequence, and the periods are N and M respectively. The dual-period training sequence corresponding to the second polarization state can be a B training sequence and a D training sequence. The periods are N and M respectively. The values of N and M are different. The training sequence with period N is repeated P times, and the training sequence with period M is repeated P times. The training sequence is repeated Q times, the values of N and M are different, and the values of P and Q can be the same or different. The values in each training sequence can be randomly selected from the symbol sequence obtained by QPSK or the symbol sequence obtained by QAM.

It should be noted that after differential polarization time coding is performed on the dual-period training sequences corresponding to the two polarization states, it can be ensured that the power of the corresponding part of the training sequence received by the single-polarization receiver is the same when any polarization rotation is introduced in the transmission channel. , and maintain the double period structure. In addition, the two-period training sequence corresponding to each subcarrier can be the same.

Step 1802: Perform differential polarization time coding on the symbol sequence and the training sequence corresponding to the first optical carrier to obtain a dual-polarization complex digital signal corresponding to the first optical carrier. In the dual-polarization complex digital signal, the encoded training The sequence is at the beginning of each data frame.

In this embodiment, the dual-polarization transmitter performs differential polarization time encoding on the training sequence corresponding to the first polarization state and the second polarization state. If the training sequence is a bi-periodic training sequence, the encoded training sequence is still a bi-periodic training sequence. sequence. The dual-polarization transmitter performs differential polarization time encoding on the symbol sequence corresponding to the first polarization state and the second polarization state to obtain an encoded symbol sequence.

If the PON adopts an architecture based on PtMP networking, if the coded symbol sequence and coded training sequence are digitally filtered by the filtering unit, they are output to the digital subcarrier multiplexing unit. The digital subcarrier multiplexing unit processes the received signals corresponding to each subcarrier into dual polarization complex digital signals that will be output by the digital-to-analog converter. A certain subcarrier among each subcarrier is the first optical carrier.

If the PON adopts an architecture based on PtP networking, if the encoded symbol sequence and the encoded dual-period training sequence are digitally filtered by the filtering unit, a dual-polarization complex digital signal corresponding to the first optical carrier is output.

In one example, the process of differential polarization time encoding is as follows.

The symbol sequence corresponding to the first optical carrier and the symbols in the training sequence are sent according to time slots. Two symbols are sent in the same time slot, and the two symbols are sent in the first polarization state and the second polarization state respectively. The symbols of the first polarization state and the second polarization state corresponding to each time slot are obtained in the symbol sequence and the dual-period training sequence. For the T-th time slot, obtain the symbol of the first polarization state and _the symbol of the second polarization state as X _T and Y _T , and construct the coding matrix corresponding to the T-th time slot according to the Alamouti coding rules _. A row or column in the coding matrix corresponding to the T-th time slot. For example, the coding matrix corresponding to the T-th time slot is expressed as For the coding matrix corresponding to the T-th time slot, the matrix obtained by performing any row or column exchange operation on A _T meets the requirements. Subsequently, Take an example to illustrate.

When T takes the value 1, take any 2×2 unitary matrix as the initial matrix. For example, the 2×2 unitary matrix can be Right multiply the initial matrix by the coding matrix corresponding to the first time slot (such as ) to obtain a product, which is then multiplied by the scaling factor corresponding to the first time slot to obtain the differential polarization time encoding matrix corresponding to the first time slot. The differential polarization time encoding matrix is also a 2×2 matrix. The scaling factor corresponding to the first time slot is determined by the value of the determinant of the initial matrix, which is equal to the reciprocal of the square root of the value of the determinant of the initial matrix multiplied by any known constant.

When the value of T is not 1, right multiply the differential polarization time coding matrix corresponding to the T-1th time slot by the coding matrix corresponding to the Tth time slot (such as ) to obtain a product, which is then multiplied by the stretching factor corresponding to the T-th time slot to obtain the differential polarization time encoding matrix corresponding to the T-th time slot. The differential polarization time encoding matrix corresponding to the T-th time slot is expressed as Equation (0).

Among them, in formula (0), C _T represents the differential polarization time encoding matrix corresponding to the T-th time slot, C _T-1 represents the differential polarization time encoding matrix corresponding to the T-1th time slot, α _T-1 represents The scaling factor corresponding to the T-th time slot, det(C _T-1 ) represents the value of the determinant of matrix C _T-1 , det(A _T-1 ) represents the value of the determinant of matrix A _T-1 , A _{T -1} represents the coding matrix corresponding to the T-1th time slot. The scaling factor corresponding to the T-th time slot is determined by the value of the determinant of the coding matrix corresponding to the T-1th time slot or the value of the determinant of the differential polarization time coding matrix corresponding to the T-1th time slot, such as The reciprocal of the square root of the determinant of the coding matrix corresponding to the T-1th time slot.

The differential polarization time encoding matrix corresponding to each time slot is a 2×2 matrix. The differential polarization time encoding matrix corresponding to each time slot is sequentially connected to obtain an encoded dual-polarization symbol sequence. In the encoded dual-polarization symbol sequence, the symbols in the first row correspond to the first polarization state, and the number of symbols becomes twice that before encoding. The symbols in the second row correspond to the second polarization state, and the number of symbols becomes 2 times before encoding. times. It can also be understood that the differential polarization time encoding unit encodes the pre-encoding dual-polarization symbol sequence of length N as a unit. Since the dual-polarization symbol is two symbols, the pre-encoding dual-polarization symbols total 2N symbols, which are The encoding is an encoded dual-polarization symbol sequence of length 2N. The encoded symbol sequence can be divided into N 2×2 encoded matrices in units of two consecutive time slots, totaling 4N symbols. Each matrix corresponds to a pre-encoded time slot, that is to say, the encoded symbol sequence Each two time slots in carries the information of one time slot in the symbol sequence before encoding.

Perform digital filtering and other processing on the dual-polarization symbol sequence to obtain a dual-polarization complex digital signal.

It should be noted that in an optical communication system, differential polarization time encoding can be implemented in parallel.

It should also be noted that after differential polarization time encoding is performed, the symbols corresponding to the first polarization state will be divided into multiple data frames, and the symbols corresponding to the second polarization state will also be divided into multiple data frames. In the data frame , the encoded training sequence is located at the beginning of the data frame, that is to say, the encoded training sequence and payload are arranged from front to back, see Figure 20. In the data frames transmitted simultaneously by two polarization states, the data frame corresponding to the first polarization state includes a part of the encoded training sequence, and the data frame corresponding to the second polarization state includes another part of the encoded training sequence.

In another example, the DSP may not generate the coding matrix corresponding to the T-th time slot group, but directly use the two symbols corresponding to the T-th time slot to determine the differential polarization time coding corresponding to the T-th time slot. matrix. For example, consider situation. Multiply the two symbols corresponding to the T-th time slot by the differential offset corresponding to the T-1th time slot. The two elements in the first row of the polarization time matrix are then added to obtain the first value. The two symbols corresponding to the T-th time slot are multiplied by the second element of the differential polarization time matrix corresponding to the T-1th time slot. The two elements of the row are then added to get the second value. Multiply the first value and the second value by the scaling factor corresponding to the T-th time slot to obtain the third value and the fourth value. The third value and the fourth value are determined as the first element and the second element of the first column in the differential polarization time encoding matrix corresponding to the T-th time slot. Then use the Alamouti coding rule to take the conjugate of the third value to obtain the second element of the second column in the differential polarization time encoding matrix corresponding to the T-th time slot. Take the inverse of the conjugate of the fourth value to obtain the th The first element in the second column of the differential polarization time encoding matrix corresponding to T time slots. Here, the two symbols corresponding to the T-th time slot are used as the first column in the coding matrix corresponding to the T-th time slot as an example for explanation.

Step 1803: Modulate the dual-polarization complex digital signal onto the first optical carrier according to dual-polarization IQ modulation.

In this embodiment, the dual-polarization complex digital signal corresponding to the first optical carrier includes four digital signals, which are the I signal in the first polarization state, the Q signal in the first polarization state, and the I signal in the second polarization state. The Q signal and the second polarization state signal are represented as XI, XQ, YI and YQ respectively. The four digital signals pass through a digital-to-analog converter and are converted into four analog signals. Different digital signals are input into different digital-to-analog converters. Four analog signals are loaded to the four electrical signal input ports of a dual-polarization IQ modulator through four modulator drivers. The laser in the dual-polarization transmitter inputs optical signals to the dual-polarization IQ modulator. The dual-polarization IQ modulator converts the four An analog signal is modulated onto the optical signal to obtain a first optical carrier modulated with the dual-polarization complex digital signal.

It should be noted here that if the PON adopts an architecture based on PtMP networking, when the subcarriers are symmetrical and there is no frequency offset in the laser, the optical signal output by the laser in the dual-polarization transmitter is as shown in Figure 2 or At the center frequency f0 in Figure 10, the dual-polarization complex digital signals corresponding to all subcarriers are modulated onto the optical signal at this center frequency, and each subcarrier modulated with data is obtained. When the frequency of the subcarrier corresponding to each leaf node is not symmetrical about f0, or the laser has a frequency offset, the center frequency of the optical signal output by the laser in the dual-polarization transmitter is not f0.

If the PON adopts an architecture based on PtP networking, when there is no frequency offset in the laser, the frequency of the optical signal output by the laser in the dual-polarization transmitter is the center frequency of the first optical carrier (the frequency of the single-polarization receiver is Single polarization homodyne receiver), or the frequency of the optical signal output by the laser in the dual polarization transmitter is the intermediate frequency of the first optical carrier and the optical carrier that does not modulate data (the single polarization receiver is a single polarization heterodyne receiver or Single polarization single sideband receiver).

Step 1804: Send the first optical carrier modulated with the dual-polarization complex digital signal.

In this embodiment, if the PON adopts an architecture based on PtMP networking, subcarriers modulated with dual-polarization complex digital signals are sent to all leaf nodes, and the first optical carrier is one of the subcarriers.

If the PON adopts an architecture based on PtP networking, the first optical carrier modulated with a dual-polarization complex digital signal is sent to a single leaf node.

The following describes the data receiving method in the embodiment of the present application. The method is executed by a single polarization receiver, specifically by the DSP in the single polarization receiver. See steps 2101 to 2105 in Figure 21.

Step 2101: Obtain the first complex digital signal of the third polarization state corresponding to the first optical carrier.

In this embodiment, it is assumed that the polarization state of the local light of the single-polarization receiver is the third polarization state. The third polarization state has nothing to do with the first polarization state and the second polarization state mentioned above, and may be related to the first polarization state. Alternatively, the second polarization state may be the same, or may be different from both the first polarization state and the second polarization state.

As can be seen from the previous description, since the polarization state of the local oscillator light is the third polarization state, among the optical signals of the two polarization states received by the single-polarization receiver, only the optical signal component of the third polarization state can be compared with the local oscillation light. A beat frequency is generated, so each ADC receives a real digital signal of the third polarization state corresponding to the first optical carrier.

When using a single-polarization homodyne receiver, the real digital signals received by the two ADCs recover the first complex digital signal of the third polarization state corresponding to the first optical carrier. When using a single polarization heterodyne receiver or a single polarization single sideband receiver, the The real digital signal received by one ADC passes through the single sideband signal recovery unit (this step is optional). The frequency shifting unit recovers the first complex digital signal of the third polarization state corresponding to the first optical carrier. The first complex digital signal is The signal is a baseband complex digital signal.

Step 2102: Perform frequency offset compensation processing on the complex digital signal.

In this embodiment, the frequency offset compensation unit in the DSP performs frequency offset compensation processing on the complex digital signal according to the set frequency offset value.

Step 2103: Perform SISO equalization processing on the signal after frequency offset compensation processing.

In this embodiment, the SISO equalization unit in the DSP performs SISO equalization processing on the signal after frequency offset compensation processing, and outputs the equalized signal.

It should be noted that since the subsequent differential polarization time decoding process eliminates the effects of polarization rotation and phase noise, when the framing is accurate and the clock recovery is perfect, the SISO equalization unit only needs to compensate for dispersion and insufficient device bandwidth. Inter-symbol interference (ISI) introduced by impairment, otherwise, the SISO equalization unit can also absorb a certain degree of framing error and sampling point deviation.

Step 2104: Perform differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state. The third polarization state and the fourth polarization state are orthogonal.

In this embodiment, the differential polarization time decoding unit in the DSP performs differential polarization time decoding processing on the equalized signal to obtain a symbol sequence corresponding to the third polarization state and a symbol sequence corresponding to the fourth polarization state.

In one example, the process of performing differential polarization time decoding is as follows. The encoding and decoding processes described in this application are discussed in units of one coding block, and one coding block is a block composed of consecutively encoded symbols.

It can be seen from the coding principle of the differential polarization time coding unit mentioned above that two consecutive time slots of the coded symbol sequence correspond to one time slot of the pre-coding symbol sequence, so the differential polarization time decoding unit performs operation in units of two consecutive time slots. Grouping: divide the equalized signal into multiple time slot groups, and each time slot group includes two symbols of the third polarization state.

In the Alamouti coding rule, the two third polarization state symbols included in each time slot group can be used as any row of the receiving end matrix corresponding to each time slot group. The receiving end matrix is a 2×2 matrix. For the nth time slot group, the differential polarization time decoding unit uses the two symbols of the third polarization state included in the nth time slot group according to the Alamouti coding rules to generate the two symbols of the fourth polarization state included in the nth time slot group. . The two symbols of the third polarization state and the two symbols of the fourth polarization state included in the n-th time slot group form a receiving end matrix corresponding to the n-th time slot group. Among them, when the symbols of the two third polarization states are the first row of the receiving end matrix corresponding to the nth time slot group, the two symbols of the fourth polarization state are the second row of the receiving end matrix corresponding to the nth time slot group. OK. When the two symbols of the third polarization state are the second row of the receiving end matrix corresponding to the nth time slot group, the two symbols of the fourth polarization state are the first row of the receiving end matrix corresponding to the nth time slot group, n is greater than or equal to 2. For example, what a single polarization receiver receives is r _1,n and r _2,n , which are the symbols of the two third polarization states included in the nth time slot group. They are the first row of the receiving end matrix and the receiving end matrix. The first element of the second row is The second element of the second row is For another example, what the single polarization receiver receives is r _1,n and r _2,n , which are the two third polarization state symbols included in the nth time slot group, and are the second row of the receiving end matrix. The receiving end matrix The first element of the first row is The second element of the first row is In the following description, the symbols of two third polarization states are used as the first row of the receiving end matrix as an example.

In a coding block, when n is 1, the receiving end matrix corresponding to the n-1th time slot group belongs to the previous coding block. Therefore, the time slot before differential polarization time coding corresponding to the nth time slot group is The symbols of the two polarization states cannot be obtained, and the symbols of the two polarization states in the time slot before differential polarization time encoding may not be valid information.

When n is not 1, left multiply the receiving end matrix corresponding to the nth time slot group by the conjugate transpose of the receiving end matrix corresponding to the n-1th time slot group to obtain a result, and multiply the result by the conjugate transpose of the receiving end matrix corresponding to the n-1th time slot group The scaling factor corresponding to the n time slot group is used to obtain the nth time slot group. the corresponding encoding matrix. For example, if the ISI is completely compensated, the receiving end matrix corresponding to the nth time slot group is expressed as Equation (1).

Among them, in formula (1), R _n and C _n represent the receiving end matrix of the n-th time slot group recovered by the single-polarization receiver and the differential polarization of the dual-polarization transmitter (transmitting end) corresponding to the n-th time slot group. Time encoding matrix, the single polarization receiver actually receives r _1,n and r _2,n (r _1,n and r _2,n are the symbols of the two third polarization states of the nth time slot group), and Non-R _n , R _n is the receiving end matrix corresponding to the nth time slot group reconstructed according to the characteristics of the Alamouti coding rule, and equation (8) is the receiving end matrix with r _1,n and r _2,n One row is reconstructed and decoded according to differential polarization time to obtain the matrix. The first row corresponds to the first polarization state of the dual-polarization transmitter, and the second row corresponds to the second polarization state of the dual-polarization transmitter. In addition, r _1,n and r _2,n are used as the second row of the receiving end matrix for reconstruction. The receiving end matrix The first row of the matrix obtained by decoding according to the differential polarization time corresponds to the second polarization state of the dual-polarization transmitter, and the second row corresponds to the first polarization state of the dual-polarization transmitter. In the following text, Take an example to illustrate. p _Rx and p _Tx are scalars, representing the frequency offset and the unknown phase introduced by the phase noise of the laser of the dual-polarization transmitter and the laser of the single-polarization receiver respectively. H is a 2×2 matrix, representing the polarization effect during the transmission process. . p _Rx , p _Tx and H are unchanged between four consecutive symbols, so after compensating for ISI, equation (2) holds.

Among them, in formula (2), is the conjugate transpose of the receiving end matrix of the n-1th time slot group, det(C _n-1 ) is the value of the determinant of matrix C _n-1 , and S _n is the decoded value corresponding to the n-th time slot group. Matrix, S _n corresponds to the coding matrix of the time slot corresponding to the nth time slot group at the transmitting end. When there is no channel impairment other than polarization rotation, S _n corresponds to the coding matrix of the time slot corresponding to the nth time slot group at the transmitting end. The matrices are the same.

On the single polarization receiver side, there is:

From formula (2) and formula (3) we can know:

Equation (4) is the principle of differential polarization time decoding. In equation (4), Indicates the scaling factor corresponding to the nth time slot group. It is worth noting that |h ₁ | ² + |h ₂ | ² changes slowly relative to the phase change introduced by the frequency offset value and phase noise, so the contribution of |h ₁ | ² + |h ₂ | ² is essentially can be absorbed by the SISO equalization unit before differential polarization time decoding. Therefore, when a single polarization receiver uses the DSP in the embodiment of the present application, there is no need to perform |h ₁ | ² + |h ₂ | ² during the differential polarization time decoding process. estimate. Therefore, the scaling factor corresponding to the n-th time slot group is determined by the value of the determinant of the receiving end matrix of the n-1th time slot group or the value of the determinant of the decoded matrix of the n-1th time slot group.

It can be seen from the Alamouti coding rules that the coding matrix corresponding to the nth time slot group is obtained from the symbols of the two polarization states of the corresponding time slot before differential polarization time encoding. Therefore, in the decoded matrix corresponding to the nth time slot group, Can get to the nth The symbols of the two polarization states of the time slot before differential polarization time encoding corresponding to the time slot group. For example, before differential polarization time encoding, the symbols of the two polarization states in the same time slot are the first column of the encoding matrix, then the two elements of the first column are obtained in the decoded matrix corresponding to the nth time slot group , that is, the symbols of the two polarization states of the time slot before differential polarization time encoding corresponding to the nth time slot group.

In this way, the symbols of the two polarization states of the time slots before differential polarization time encoding corresponding to each time slot group are sequentially connected together to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state respectively.

Step 2105: Based on the symbol sequence, restore the data corresponding to the third polarization state and the fourth polarization state.

In this embodiment, the symbol-bit mapping unit in the DSP performs symbol bit mapping on the symbol sequence corresponding to the third polarization state to obtain a bit stream corresponding to the symbol sequence. The FEC decoding unit in the DSP performs FEC decoding on the bit stream and obtains the data sent by the central node in one of the two polarization states.

The symbol-bit mapping unit in the DSP performs symbol bit mapping on the symbol sequence corresponding to the fourth polarization state to obtain a bit stream corresponding to the symbol sequence. The FEC decoding unit in the DSP performs FEC decoding on the bit stream to obtain the data sent by the central node on the other of the two polarization states.

Using the process shown in Figure 21, the single-polarization receiver receives the optical signal in a single polarization state, and the data on the optical signal in the dual-polarization state can be recovered using the optical signal in the single polarization state. In this way, dual-polarization transmission and single-polarization reception can be achieved in a coherent optical communication system.

The following is a supplementary explanation of the process shown in Figure 21.

In one example, when a single polarization receiver is deployed and online, which is the start-up stage of data signal processing by DSP, the frequency offset value of the laser may be relatively large. A dual-cycle training sequence can be used to initially estimate the frequency offset value. , that is, the initial frequency offset value is obtained. The process is as follows.

When the single polarization receiver is deployed and online, the DSP obtains the second complex digital signal, and the DSP performs the following operations on each data frame in the second complex digital signal.

Among them, in equation (5), assuming that the single-polarization receiver receives the X polarization state (i.e., the third polarization state), α(l) represents the value corresponding to the l-th symbol, which is an intermediate quantity, r _x ( k) represents the k-th symbol in a data frame, and 2M is a period value in the dual-period training sequence after differential polarization time encoding. Generally, a larger period value is used, that is, M is greater than N, r _x (k+2M ) represents the k+2M symbol in a data frame, assuming 1≤l≤L-2M, and L represents the length of a single data frame.

Calculate the value corresponding to each l value according to equation (5), determine the peak position of the value corresponding to each l value, and determine the frame header position of the data frame based on the peak position.

From the frame head position, according to the length of the training sequence after differential polarization time encoding of a single polarization state, the length of the training sequence after differential polarization time encoding is equal to twice the length of the training sequence before differential polarization time encoding, from the received sequence, Get the first received sequence of corresponding length and perform the operation of equation (6).

In formula (6), b ₁ (k) and b ₂ (k) are two intermediate quantities.

Assume that the dual-period training sequences after differential polarization time encoding corresponding to the _two polarization states are _S

Among them, in equation (7), S _X,Y (k) represents the k-th symbol of the third polarization state and the k-th symbol of the fourth polarization state.

Taking into account the influence of polarization rotation and frequency offset value Δf ₁ , there is equation (8).
r _x ₍ k ₎ _＝ [ _h ₁ _S

Among them, in equation (8), T _s represents the duration of each symbol in the double-period training sequence, which is the reciprocal of the baud rate, h ₁ and h ₂ respectively represent the two elements of the first row of the transmission matrix, j Represents an imaginary unit.

Combining Equation (6) to Equation (8), Equation (9) can be obtained.

Taking the phase of b ₁ (k) and b ₂ (k), the frequency offset values Δf ₁₁ and Δf ₁₂ can be calculated. Obviously, the relationship between Δf ₁₁ and Δf ₁₂ and the real frequency offset value Δf ₁ satisfies the equation (10) .

Among them, in formula (10), n ₁ and n ₂ are integers. Generally, the frequency offset value range is known, which determines the value range of n ₁ and n ₂ . We can exhaustively enumerate all value combinations of n ₁ and n ₂ , and take formula (11) is the minimum value combination. Then calculate the initial frequency offset value according to equation (10).

After determining the initial frequency offset value, frequency offset compensation can be implemented in digital signal processing, that is, frequency offset compensation is performed before SISO equalization processing. Frequency offset compensation can also be implemented in analog signal processing, that is, the DSP outputs a frequency offset value to the control circuit of the local oscillator laser, and the control circuit adjusts the temperature and current of the local oscillator laser according to the frequency offset value, thereby Control the center wavelength of the local oscillator laser. Frequency offset compensation can also be implemented jointly in digital signal processing and analog signal processing.

In another example, when the single-polarization receiver is deployed and online, which is the startup stage of data signal processing by DSP, the frequency offset value of the laser is considered to be small. At this time, there is no need to make a rough estimate of the frequency offset and can be completely relied on. The post-equalization frequency offset tracking mechanism described later completes frequency offset compensation and/or laser frequency locking. At this time, the data signal processing start-up phase only needs to complete the framing, and the framing does not need to have a large tolerance for frequency offset.

In this case, there is no need to introduce a double-periodic structure into the training sequence. You only need to use any training sequence with good autocorrelation to perform differential polarization time encoding to obtain the training sequence after differential polarization time encoding. At the start-up stage of signal processing, it is only necessary to perform correlation operations on the corresponding parts of the single-frame length sequence in the receiving end sequence and the first polarization state and the second polarization state of the locally stored differential polarization time-encoded sequence, and obtain the correlation peaks respectively. Select the position of the larger correlation peak to determine the head position of the data frame.

In one example, the SISO equalization unit is a single-input single-output equalizer. Since the process of differential polarization time decoding eliminates the effects of polarization rotation and phase noise, there is no need to use the classic 2x2 multiple-input multiple-output in coherent optical communications. (multi-input multi-output, MIMO) equalizer tracks polarization effects. In the embodiment of this application, the influence of polarization mode dispersion (PMD) in the optical fiber is ignored. This is acceptable in short-distance communication systems with low baud rates. The link length of PON is generally less than 20km, which is short-distance. Communication Systems.

After the single-polarization receiver is deployed and online, the tap coefficients of the SISO equalization unit can be dynamically updated. The update process is as follows.

Determine the symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, and determine the first symbol error. The first symbol error is used to characterize the differential polarization time-decoded symbol sequence corresponding to the third polarization state and the first polarization state. The error between the corresponding symbol sequence before differential polarization time encoding and the error between the symbol sequence after differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence before differential polarization time encoding corresponding to the second polarization state. Based on the first symbol error and the chain rule, the second symbol error is determined. The second symbol error is used to characterize the symbol sequence before differential polarization time decoding corresponding to the third polarization state and the differential polarization time encoding corresponding to the first polarization state. The error in the symbol sequence and/or the error between the symbol sequence before differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence after differential polarization time encoding corresponding to the second polarization state. Based on the second symbol error, the tap coefficients used for SISO equalization processing are updated. The first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier.

In this embodiment, since the SISO equalization unit absorbs the contribution of |h ₁ | ² + |h ₂ | ² that changes with time, the tap coefficient of the SISO equalization unit needs to be dynamically updated. The symbol sequence after differential polarization time decoding is a signal with a preset modulation format (the preset modulation format is QPSK or 16-QAM, etc.), and the influence of phase noise and polarization effects is eliminated, so there are distinguishable constellation points, which can Sign errors are directly determined and calculated. But the problem is that when using the least mean square (LMS) algorithm to update the tap coefficients, the symbol error before differential polarization time decoding needs to be used. In the embodiment of the present application, the chain rule and a certain approximation are used to calculate the second symbol error before differential polarization time decoding from the first symbol error after differential polarization time decoding.

Assume that a single polarization receiver receives a one-dimensional vector After SISO equalization processing by the SISO equalization unit, we obtain Expressed by formula (12).

Among them, in equation (12), u _1,n-1 and u _2,n-1 represent the symbols of the two third polarization states corresponding to the n-1th time slot after SISO equalization processing.

Construct the receiving end matrix corresponding to the n-1th time slot and the receiving end matrix corresponding to the n-1th time slot, which are expressed by Equation (13) and Equation (14) respectively.

Considering that |h ₁ | ² +|h ₂ | ² can be absorbed by the SISO equalization unit, ignoring |h ₁ | ² +|h ₂ | ² , so the decoded matrix recovered by the single polarization receiver is expressed in equation (15) express.

In equation (15), Represents the reciprocal of the square root of the determinant of U _n-1 . U _n-1 is a unitary matrix, so there is

The error of S _n can be obtained from the judgment and is expressed by equation (16).

Among them, in equation (16), Dec(s) represents the decision on symbol s. In this example, s _{1, n} is the decoded symbol of the first symbol corresponding to the third polarization state in the n-th time slot group. , s _2,n is the second symbol solution corresponding to the third polarization state in the nth time slot group The coded symbol, Dec(s _1,n )-s _1,n represents the error between the decoded first symbol corresponding to the third polarization state and the corresponding symbol before differential polarization time encoding corresponding to the first polarization state, Dec (s _2,n )-s _2,n represents the error between the decoded second symbol corresponding to the third polarization state and the corresponding symbol before differential polarization time encoding corresponding to the first polarization state, Represents the error between the decoded first symbol corresponding to the fourth polarization state and the corresponding symbol before differential polarization time encoding corresponding to the second polarization state, Indicates the error between the decoded second symbol corresponding to the fourth polarization state and the corresponding symbol before differential polarization time encoding corresponding to the second polarization state.

The LMS algorithm itself is an approximation of gradient descent, and for gradients, there is the relationship of equation (17) according to the chain rule.

From formula (15) we can see that, The term is only related to U _n-1 . An approximation is made here. U _n-1 is fixed each time the coefficient is updated, that is, the gradient is 0. Only the change of U _n is considered, and then err(S _n ), err(U _n ) are used. replace and Then there is formula (18).

In err(U _n ) in equation (18), the elements err(u _1,n ) and err(u _2,n ) are the symbol errors before decoding corresponding to the third polarization state. The elements and is the symbol error before decoding corresponding to the fourth polarization state, Taking the opposite number after conjugation is equal to err(u _2,n ), After taking the conjugate, it is equal to err(u _1,n ). When using the LMS algorithm to update the tap coefficients of the SISO equalization unit, some or all four elements of err(U _n ) can be used. For example, in the embodiment of this application, only the upper-left element err(u _1,n ) in err(U _n ) can be used, and the process of updating the tap coefficients of the SISO equalization unit using the LMS algorithm is shown in Equation (19).

In equation (19), the vector represents the tap coefficient of the SISO equalization unit, μ represents the step coefficient of the LMS algorithm, means taking the conjugate of each element, is the one-dimensional vector received by the single-polarization receiver, corresponding to the third polarization state.

In addition, the second symbol error is used to represent the error between the symbol sequence before differential polarization time decoding corresponding to the third polarization state and the symbol sequence after differential polarization time encoding corresponding to the first polarization state and the differential polarization corresponding to the fourth polarization state. When there is an error between the symbol sequence before time decoding and the symbol sequence after differential polarization time encoding corresponding to the second polarization state, in one example, it can be determined in Equation (18) Take the opposite of the conjugate and use The updated tap coefficient is obtained, and the updated tap coefficient is also calculated using equation (19). The two updated tap coefficients are averaged to obtain the updated tap coefficient.

In one example, when a single polarization receiver is deployed and online, the training sequence can be used to pre-converge the tap coefficients of the SISO equalization unit. The pre-convergence method of the tap coefficients is as follows. The training sequence can be a sequence with relatively good autocorrelation, or a training sequence with a double period structure.

The DSP acquires the second complex digital signal of the third polarization state corresponding to the first optical carrier, and the second complex digital signal includes the differential polarization time-encoded training sequence corresponding to the third polarization state.

The DSP determines the frame header position in the second complex digital signal using the method described above. Using the frame head position and the length of the differential polarization time-encoded training sequence corresponding to the third polarization state, in the second complex digital signal, the second received sequence corresponding to the training sequence is obtained. Based on the second receiving sequence and the Alamouti coding rule, the third receiving sequence corresponding to the fourth polarization state is restored. The third receiving sequence is the training sequence corresponding to the fourth polarization state. For the restoration method, refer to the previous description.

The DSP performs differential polarization time decoding on the second receiving sequence and the third receiving sequence, and obtains decoded training sequences corresponding to the third polarization state and the fourth polarization state respectively.

The DSP determines the error between the decoded training sequence corresponding to the third polarization state and the training sequence before differential polarization time encoding corresponding to the first polarization state. And determine the error of the decoded training sequence corresponding to the fourth polarization state and the training sequence before differential polarization time encoding corresponding to the second polarization state. These two errors are called third symbol errors. The first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier.

The DSP determines the fourth symbol error based on the third symbol error and the chain rule. The fourth symbol error is used to characterize the error of the differential polarization time-encoded training sequence corresponding to the second received sequence and the first polarization state and/or the third symbol error. The error between the received sequence and the differential polarization time-encoded training sequence corresponding to the second polarization state. The DSP updates the tap coefficients used for SISO equalization processing based on the fourth symbol error.

It should be noted that although the above processes are all about updating tap coefficients, there are differences between them. The difference between the pre-convergence process and the dynamic update process is that the calculation of the symbol error in the dynamic update process relies on the judgment of the symbols after differential polarization time decoding, while the calculation of the symbol error in the pre-convergence process can directly use the training after differential polarization time encoding. sequence, which contributes to the correct convergence of the equalizer tap coefficients during the startup phase of digital signal processing and gives the equalizer the ability to absorb possible framing errors.

In one example, in the above process of updating the tap coefficients, the default frequency offset value has been fully compensated. However, laser frequency offsets in optical communication systems may cause the assumption that the channel response remains unchanged between four consecutive symbols before differential polarization time decoding to be invalid. Although the dual-period training sequence can be used for deviation offset estimation compensation, the frequency offset may not be completely eliminated in the following scenarios. Scenario 1: In low signal-to-noise ratio scenarios (such as in PON due to large link insertion loss), the frequency offset estimation based on the dual-cycle training sequence may be biased and the frequency offset value cannot be completely eliminated. Scenario 2: The frequency offset value of the laser changes with temperature and laser driving current, that is, the frequency offset value of the laser is time-varying. If the frequency offset value is estimated only by relying on the two-period training sequence, the training sequence needs to be inserted frequently, which is equivalent to reducing the proportion of payload in each data frame and increasing the overhead. Therefore, in the embodiment of the present application, the frequency offset value can also be dynamically tracked.

For convenience of description, consider here the situation where ISI has been compensated by the SISO equalization unit. Consider the symbols u _1,n-1 ,u _2,n-1 ,u _1,n ,u _2,n of four consecutive receivers designed by a differential polarization time decoding operation. It is assumed that there are still residual frequencies before differential polarization time decoding. Offset value Δf ₂ , then the actual symbols entering the differential polarization time decoding unit are: u _1,n-1 p _1,n-1 ,u _2,n-1 p _2,n-1 ,u _1,n p _{1 ,n} ,u _2,n p _2,n , and there exists p _2,n ＝p _1,n Δp＝p _2,n-1 (Δp) ² ＝p _1,n-1 (Δp) ³ , where Δp =exp(j2πΔf ₂ T _s ), T _s is the symbol duration (the reciprocal of the baud rate).

Perform differential polarization time decoding according to equation (15) to obtain equation (20).

Among them, in equation (20), the symbols of the two polarization states of the time slots before encoding corresponding to the nth time slot group of X _n and Y _n , and the two polarization states are the third polarization state and the fourth polarization state, The third polarization state is the polarization state of the local light of the single-polarization receiver, and the third polarization state and the fourth polarization state are represented by the X polarization state and the Y polarization state respectively.

After mathematical derivation, formula (21) can be obtained.

Figure 22 gives a specific example. Considering the 8G baud rate (baud) 16-QAM signal, the OSNR is 35dB and is not considered. Considering the phase noise of the laser, only the influence of frequency offset is needed. It can be seen that if there is no frequency offset, X _n and Y _n have clear constellation points under this channel condition, but if there is a residual frequency offset of 100MHz, Then the X _n constellation points are dispersed, and the Y _n constellation points are dispersed, and there is an overall constellation rotation.

The constellation point dispersion corresponds to the influence of the Δp correlation term in the brackets in Equation (21), while the overall rotation of the constellation diagram corresponds to the Δp term outside the brackets that can be extracted from the expression of Y _n in Equation (21). It can be seen from Figure 22 and Equation (21) that the frequency offset will bring a non-negligible performance cost to differential polarization time decoding, so the frequency offset should be compensated before decoding, and the constellation of the symbol sequence corresponding to Y _n can be The rotation angle of the graph estimates the frequency offset value. Specifically, for the situation where the signal after differential polarization time decoding is QPSK, 16-QAM and other standard QAM, the classic Viterbi-Viterbi (Viterbi-Viterbi) algorithm can be used to estimate the rotation angle, that is, estimate the value of Δp, and then according to Δp=exp(j2πΔf ₂ T _s ) calculates the frequency offset value Δf ₂ . In this process, there is no need to know the transmitter symbols. The payload part in the data frame can be used to estimate the real-time frequency offset value, and then the estimated frequency offset value can be used to perform frequency offset compensation in the digital domain, and/ Or adjust the center frequency of the local oscillator laser to complete dynamic tracking of frequency offsets.

In the embodiment of the present application, in the coherent optical communication system of dual-polarization transmission and single-polarization reception, symbols of a single polarization state are used to recover symbols of two polarization states, and taps for differential polarization time encoding and decoding adaptation are also provided. Coefficient update method. In addition, the frequency offset can also be estimated during the startup phase of digital signal processing, and the frequency offset can be dynamically updated during the normal working phase of digital signal processing, enabling the use of differential polarization time encoding and decoding schemes in coherent optical communication systems.

In order to explain more clearly the technical effects of the embodiments of the present application. Consider an example as follows: a leaf node receives a subcarrier with an 8G baud rate, the modulation format of the uncoded signal is QPSK, the OSNR is 12dB, the digital signal processing algorithm parallelism is N, and the loop delay is D. The system performance simulation results after using the technical solution of the embodiment of the present application are shown in Figure 23 (when obtaining the corresponding simulation of Figure 23, only the differential polarization time decoding of the DSP in the single polarization receiver was turned on, and the tap coefficients were dynamically updated, but the Turn on frequency offset estimation compensation and dynamic tracking of frequency offsets, that is, the cost of directly tolerating frequency offsets by differential polarization time decoding).

As can be seen from Figure 23, considering the typical system parameters in the PON scenario: 1MHz line width (meaning that the line width of the dual-polarization transmitter laser and the single-polarization receiver local oscillator laser are both 1MHz), the laser frequency offset value is 40MHz, and the parallelism is 16 /32 (corresponding to clock frequency 500/250MHz), the tap coefficient update loop delay is 48 beats, and the BER before error correction after using the technical solution of the embodiment of this application is lower than the soft decision FEC threshold 2e-2.

This application proposal proposes to use SISO equalization to compensate ISI before differential polarization time decoding, and proposes a tap dynamic update algorithm adapted to differential polarization time encoding, which solves the problem of traditional LMS algorithm in differential polarization time encoding single carrier/digital subcarrier Problems with failures in reused systems. (a) in Figure 24 shows the OSNR-BER performance curve of the system when the SISO equalization unit uses different numbers of taps. The simulation parameter settings are the same as Figure 23. This simulation only considers the dispersion of 20km fiber and does not consider the influence of device bandwidth. We have seen that increasing the number of taps within a certain range can significantly improve system performance, proving that the SISO balancing unit here has a beneficial effect in compensating for ISI.

Embodiments of the present application also propose a frequency offset estimation compensation scheme adapted to differential polarization time coding to support the application of this coding in optical communication systems. (b) in Figure 24 shows the system performance curve when residual frequency offsets of different sizes are before equalization. The simulation parameters are the same as Figure 23. It can be seen from (b) in Figure 24 that without frequency offset compensation, although differential polarization time coding itself can tolerate frequency offset within a certain range, if the frequency offset value is large, it will cause OSNR performance penalty, and The encoding cannot work properly when the frequency offset value is too large. Specific to the current example, considering the 8G baud rate signal, there is no obvious cost to the system if the frequency offset value is less than 40MHz. When the frequency offset value reaches 100MHz, there is about 1dB OSNR penalty at the soft decision FEC threshold, and the frequency offset value is greater than 500MHz. The system is not working properly.

The embodiment of the present application also provides a frequency offset dynamic tracking solution. Figure 25 shows a schematic diagram of the residual frequency offset when the frequency offset dynamic tracking algorithm is turned on and off. It is obvious that after the frequency offset dynamic tracking algorithm is turned on, residual frequency offset The shift ratio is small.

In the embodiment of this application, a coherent optical communication system based on PtMP networking is considered. The central node uses a high-bandwidth dual-polarization IQ transmitter to transmit multiple digital subcarriers, and the leaf nodes use a low-bandwidth polarization heterodyne receiver to receive a single digital subcarrier. subcarrier. Using this architecture, the bandwidth requirements of leaf nodes are low, which can minimize the complexity and cost of leaf nodes. Moreover, in the embodiment of the present application, frequency offset estimation and tap coefficient updating can be performed, so that dual-polarization transmission and single-polarization reception can be engineered.

In the embodiment of this application, it is considered that in a coherent optical communication system based on PtP networking, the central node only sends a single carrier signal, without the need for digital subcarrier multiplexing, and the signal baud rate can be increased.

In addition, the solution in the related art that transmits signals based on Alamouti coding based on two orthogonal polarization states requires relying on a feedback structure or a frequency domain pilot signal to track frequency offset and phase noise. However, there are loops in the feedback structure during actual implementation. The problem of bandwidth is usually not enough to track the rapidly changing phase noise. When the frequency domain pilot signal is actually implemented, the spectral resolution is limited by the number of points of the fast Fourier transform in digital signal processing, and the frequency domain pilot signal usually requires a high Power will occupy the dynamic range of digital/analog converters and analog/digital converters that could otherwise be used by signals. In actual systems, the existence of nonlinear damage to devices will bring additional costs, thereby reducing system performance. However, in the embodiment of the present application, there is no need to rely on a feedback structure to track phase noise, and no frequency domain pilot signal is needed.

The following describes the device of the embodiment of the present application.

In one example, the data sending device can be implemented as part or all of the device through software, hardware, or a combination of both. The device provided by the embodiment of this application can implement the process described in Figure 18 of the embodiment of this application.

An acquisition unit, configured to acquire a symbol sequence corresponding to the first optical carrier and a training sequence corresponding to two polarization states, the two polarization states being orthogonal, and the symbol sequence corresponding to the first optical carrier to be sent. The data is obtained through symbol mapping, which can be used to implement the acquisition function of step 1801 and execute the implicit steps included in step 1801;

A differential polarization time encoding unit, configured to perform differential polarization time encoding on the symbol sequence and the training sequence respectively to obtain a dual-polarization complex digital signal corresponding to the first optical carrier. In the dual-polarization complex digital signal, The encoded training sequence is located at the starting position of each data frame, and can be used to implement the encoding function of step 1802 and execute the implicit steps included in step 1802;

Modulation unit, used to modulate the dual-polarization complex digital signal onto the first optical carrier according to dual-polarization in-phase quadrature IQ modulation. Specifically, it can be used to implement the modulation function of step 1803 and perform the implicit instructions included in step 1803. step;;

A sending unit, configured to send the first optical carrier modulated with the dual-polarization complex digital signal, and may specifically be used to implement the sending function of step 1804.

The acquisition unit may logically include an FEC encoding unit and a bit symbol mapping unit.

The modulation unit may logically include a filter unit, a digital-to-analog converter, a modulator driver, a polarization fractionator, a polarization coupler, an in-phase quadrature modulator, etc.

In an example, the differential polarization time encoding unit is used for:

Based on the symbol sequence and the symbols of the two polarization states corresponding to each time slot in the training sequence, a coding matrix corresponding to each time slot is generated, and the symbols of the two polarization states corresponding to each time slot are respectively. Describe a row or column in the coding matrix corresponding to each time slot;

Based on the coding matrix corresponding to the T-th time slot, the differential polarization time coding matrix corresponding to the T-1th time slot, and the scaling factor corresponding to the T-th time slot, the differential polarization corresponding to the T-th time slot is obtained Time encoding matrix, the dual-polarization complex digital signal is obtained by sequentially connecting the differential polarization time encoding matrices corresponding to each time slot, and the scaling factor corresponding to the T-th time slot is obtained by the T-1th time slot The value of the determinant of the corresponding encoding matrix or differential polarization time encoding matrix is determined, and T is an integer greater than or equal to 2.

In an example, the two polarization states include a first polarization state and a second polarization state;

The differential polarization time encoding unit is used for:

Perform differential polarization time encoding on the symbol sequence corresponding to the first polarization state and the second polarization state;

Differential polarization time encoding is performed on the training sequences corresponding to the first polarization state and the second polarization state.

In one example, the training sequences respectively corresponding to the two polarization states include a first sub-training sequence and a second sub-training sequence, and the periods of the first sub-training sequence and the second sub-training sequence are different.

In one example, the device for receiving data can be implemented as part or all of the device through software, hardware, or a combination of both. The device provided by the embodiment of this application can implement the process described in Figure 21 of the embodiment of this application.

The acquisition unit is used to acquire the first complex digital signal of the third polarization state corresponding to the first optical carrier. Specifically, it can be used to implement the acquisition function of step 2101 and perform the implicit steps included in step 2101;

A frequency offset compensation unit is used to perform frequency offset compensation processing on the first complex digital signal. Specifically, it can be used to implement the frequency offset compensation function of step 2101 and perform the implicit steps included in step 2102;

The equalization processing unit is used to perform SISO equalization processing on the signal after frequency offset compensation processing. Specifically, it can be used to implement the equalization processing function of step 2103 and execute the implicit steps included in step 2103;

A differential polarization time decoding unit, configured to perform differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, and the third polarization state and the fourth polarization state. State orthogonal, specifically can be used to implement the decoding function of step 2104 and execute the implicit steps included in step 2104;

A data recovery unit, configured to recover data corresponding to the third polarization state and the fourth polarization state based on the symbol sequence. Specifically, it can be used to implement the data recovery function of step 2105 and perform the implicit steps included in step 2105. .

The acquisition unit may logically include a frequency shifting unit, a matched filtering unit, a clock recovery unit, a frame fixing unit, etc.

The equalization processing unit is the SISO equalization unit.

The restored data unit may logically include a symbol-bit mapping unit, an FEC decoding unit, and the like.

In an example, the differential polarization time decoding unit is used for:

Based on the symbols of the two third polarization states included in each time slot group in the equalized signal and the Alamouti coding rule, two symbols of the fourth polarization state included in each time slot group are generated. The two symbols of the third polarization state and the two symbols of the fourth polarization state constitute the receiving end matrix corresponding to each time slot group, and each time slot group is based on two consecutive time slots as a unit. The signals after equalization are grouped and obtained;

Based on the receiving end matrix corresponding to the nth time slot group, the receiving end matrix corresponding to the n-1th time slot group, and the scaling factor corresponding to the nth time slot group, the decoded matrix corresponding to the nth time slot group is obtained, The scaling factor corresponding to the n-th time slot group is determined by the value of the determinant of the receiving end matrix corresponding to the n-1th time slot group or the value of the determinant of the decoded matrix, where n is greater than or equal to 2. integer;

In the decoded matrix corresponding to the n-th time slot group, the symbols of the two polarization states of the time slot before differential polarization time encoding corresponding to the n-th time slot group are obtained.

In an example, the device further includes a tap coefficient updating unit, configured to:

Make a decision on the symbol sequences corresponding to the third polarization state and the fourth polarization state, and determine the first symbol error. The first symbol error is used to characterize the differential polarization time decoded symbol corresponding to the third polarization state. The error between the sequence and the symbol sequence before differential polarization time encoding corresponding to the first polarization state and the symbol sequence after differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence before differential polarization time encoding corresponding to the second polarization state error;

Based on the first symbol error and the chain rule, a second symbol error is determined. The second symbol error is used to characterize the symbol sequence before differential polarization time decoding corresponding to the third polarization state and the first polarization state. The corresponding differential polarization time The error between the encoded symbol sequence and/or the error between the symbol sequence before differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence after differential polarization time encoding corresponding to the second polarization state;

Based on the second symbol error, update the tap coefficients used for the SISO equalization process;

Wherein, the first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier.

In an example, the device further includes a frequency offset estimation unit, configured to:

Obtain the constellation diagram of the symbol sequence corresponding to the fourth polarization state, and determine the rotation angle corresponding to the constellation diagram;

Based on the rotation angle, obtain the frequency offset value corresponding to the first optical carrier;

Based on the frequency offset value, the frequency offset value used for frequency offset compensation processing is updated.

In one example, the acquisition unit is also used to:

Before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, obtain the second complex digital signal of the third polarization state corresponding to the first optical carrier, and the second complex digital signal is in the The first complex digital signal was previously received. The second complex digital signal includes a differential polarization time-encoded training sequence corresponding to the third polarization state. The training sequence includes a differential polarization time-encoded first sub-digit signal. The training sequence and the second sub-training sequence, the periods of the first sub-training sequence and the second sub-training sequence are different;

The device further includes a framing unit configured to determine the frame header position in the second complex digital signal based on the training sequence;

The device also includes a frequency offset estimation unit for:

Based on the frame header position and the length of the differential polarization time encoded training sequence, in the second complex digital signal, obtain the first received sequence corresponding to the training sequence;

Based on the first received sequence and the reference frequency offset value range, an initial frequency offset value is determined, and the initial frequency offset value is used to initially perform frequency offset compensation processing.

In one example, the acquisition unit is also used to:

Before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, obtain the second complex digital signal of the third polarization state corresponding to the first optical carrier, and the second complex digital signal is in the The first complex digital signal was previously received, and the second complex digital signal includes the differential polarization time-encoded training sequence corresponding to the third polarization state;

The device further includes a framing unit for determining the frame header position in the second complex digital signal;

The differential polarization time decoding unit is also configured to, based on the frame head position and the length of the differential polarization time encoded training sequence corresponding to the third polarization state, in the second complex digital signal, take the The second received sequence corresponding to the training sequence;

Based on the second reception sequence and the Alamouti coding rule, restore the third reception sequence corresponding to the fourth polarization state, where the third reception sequence is the training sequence corresponding to the fourth polarization state;

Perform differential polarization time decoding on the second receiving sequence and the third receiving sequence to obtain decoded training sequences corresponding to the third polarization state and the fourth polarization state respectively;

The device also includes a tap coefficient update unit for:

Determining the error of the decoded training sequence corresponding to the third polarization state and the training sequence before differential polarization time encoding corresponding to the first polarization state, and the decoded training sequence corresponding to the fourth polarization state and the second polarization state. The error of the corresponding training sequence before differential polarization time encoding is used to obtain a third symbol error, and the first polarization state and the second polarization state are the two orthogonal polarization states used when transmitting the first optical carrier;

Based on the third symbol error and the chain rule, a fourth symbol error is determined, and the fourth symbol error is used to characterize the The error between the second received sequence and the differential polarization time-encoded training sequence corresponding to the first polarization state and/or the third receive sequence and the differential polarization time-encoded training sequence corresponding to the second polarization state error;

Based on the fourth symbol error, the tap coefficients used for the SISO equalization processing are updated.

Those of ordinary skill in the art will appreciate that the method steps and units described in conjunction with the embodiments disclosed in this application can be implemented with electronic hardware, computer software, or a combination of both. In order to clearly illustrate the relationship between hardware and software Interchangeability, in the above description, the steps and compositions of each embodiment have been generally described according to functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. One of ordinary skill in the art may implement the described functionality using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

In the several embodiments provided in this application, it should be understood that the disclosed system architecture, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or may be Integrated into another system, or some features can be ignored, or not implemented. In addition, the coupling or direct coupling or communication connection between each other shown or discussed may be indirect coupling or communication connection through some interfaces, devices or modules, or may be electrical, mechanical or other forms of connection.

The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

In addition, each module in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software modules.

If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be an optical communication device, etc.) to execute all or part of the steps of the methods in various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program code.

In this application, the terms "first" and "second" are used to distinguish identical or similar items with substantially the same functions and functions. It should be understood that there is no logical or logical connection between "first" and "second". Timing dependencies do not limit the number and execution order. It should also be understood that, although the following description uses the terms first, second, etc. to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another. For example, a first symbol error may be referred to as a second symbol error, and similarly, a second symbol error may be referred to as a first symbol error, without departing from the scope of various examples. Both the first symbol error and the second symbol error may be symbol errors, and in some cases, may be separate and different symbol errors.

The term "at least one" in this application means one or more, the term "plurality" in this application means two or more, and the term "and/or" includes three situations, for example, A and/or B includes A, B, and A and B.

The above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any person familiar with the technical field can easily think of various equivalent modifications within the technical scope disclosed in the present application. Or replacement, these modifications or replacements should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A method of data transmission, characterized in that the method is applied to a coherent optical communication system, and the method includes:

Obtain the symbol sequence corresponding to the first optical carrier and the training sequences corresponding to the two polarization states. The two polarization states are orthogonal. The symbol sequence is obtained by performing symbol mapping on the data to be sent corresponding to the first optical carrier. ;

Perform differential polarization time encoding on the symbol sequence and the training sequence respectively to obtain a dual-polarization complex digital signal corresponding to the first optical carrier. In the dual-polarization complex digital signal, the encoded training sequence is located in each The starting position of the data frame;

Modulating the dual-polarization complex digital signal onto the first optical carrier according to dual-polarization in-phase quadrature IQ modulation;

The first optical carrier modulated with the dual-polarization complex digital signal is transmitted.
The method according to claim 1, characterized in that performing differential polarization time encoding on the symbol sequence and the training sequence respectively to obtain the dual-polarization complex digital signal corresponding to the first optical carrier includes:

Based on the symbol sequence and the symbols of the two polarization states corresponding to each time slot in the training sequence, a coding matrix corresponding to each time slot is generated, and the symbols of the two polarization states corresponding to each time slot are respectively. Describe a row or column in the coding matrix corresponding to each time slot;

Based on the coding matrix corresponding to the T-th time slot, the differential polarization time coding matrix corresponding to the T-1th time slot, and the scaling factor corresponding to the T-th time slot, the differential polarization corresponding to the T-th time slot is obtained Time encoding matrix, the dual-polarization complex digital signal is obtained by sequentially connecting the differential polarization time encoding matrices corresponding to each time slot, and the scaling factor corresponding to the T-th time slot is obtained by the T-1th time slot The value of the determinant of the corresponding encoding matrix or differential polarization time encoding matrix is determined, and T is an integer greater than or equal to 2.
The method according to claim 1, wherein the two polarization states include a first polarization state and a second polarization state;

Performing differential polarization time encoding on the symbol sequence and the training sequence respectively includes:

Perform differential polarization time encoding on the symbol sequence corresponding to the first polarization state and the second polarization state;

Differential polarization time encoding is performed on the training sequences corresponding to the first polarization state and the second polarization state.
The method according to any one of claims 1 to 3, characterized in that the training sequences respectively corresponding to the two polarization states include a first sub-training sequence and a second sub-training sequence, and the first sub-training sequence and The periods of the second sub-training sequences are different.
A method of data reception, characterized in that the method is applied to a coherent optical communication system, and the method includes:

Obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier;

Perform frequency offset compensation processing on the first complex digital signal;

Perform single-input single-output SISO equalization processing on the signal after frequency offset compensation processing;

Perform differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state, where the third polarization state and the fourth polarization state are orthogonal;

Based on the symbol sequence, data corresponding to the third polarization state and the fourth polarization state are recovered.
The method according to claim 5, characterized in that, performing differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state, including:

Based on the symbols of the two third polarization states included in each time slot group in the equalized signal and the Alamouti coding rule, two symbols of the fourth polarization state included in each time slot group are generated. The two symbols of the third polarization state and the two symbols of the fourth polarization state constitute the receiving end matrix corresponding to each time slot group, and each time slot group is composed of two consecutive Group the equalized signals to obtain them in units of time slots;

Based on the receiving end matrix corresponding to the nth time slot group, the receiving end matrix corresponding to the n-1th time slot group, and the scaling factor corresponding to the nth time slot group, the decoded matrix corresponding to the nth time slot group is obtained, The scaling factor corresponding to the n-th time slot group is determined by the value of the determinant of the receiving end matrix corresponding to the n-1th time slot group or the value of the determinant of the decoded matrix, where n is greater than or equal to 2. integer;

In the decoded matrix corresponding to the n-th time slot group, the symbols of the two polarization states of the time slot before differential polarization time encoding corresponding to the n-th time slot group are obtained.
The method according to claim 5 or 6, characterized in that, the method further includes:

Determine the symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, and determine the first symbol error. The first symbol error is used to characterize the differential polarization time corresponding to the third polarization state after decoding. The error between the symbol sequence before differential polarization time encoding corresponding to the first polarization state and the error between the symbol sequence after differential polarization time encoding corresponding to the fourth polarization state and the symbol sequence before differential polarization time encoding corresponding to the second polarization state Error in symbol sequence;

Based on the first symbol error and the chain rule, a second symbol error is determined. The second symbol error is used to characterize the symbol sequence before differential polarization time decoding corresponding to the third polarization state and the first polarization state. The error between the corresponding symbol sequence after differential polarization time encoding and/or the error between the symbol sequence before differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence after differential polarization time encoding corresponding to the second polarization state ;

Based on the second symbol error, update the tap coefficients used for the SISO equalization process;

Wherein, the first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier.
The method according to any one of claims 5 to 7, characterized in that the method further includes:

Obtain the constellation diagram of the symbol sequence corresponding to the fourth polarization state, and determine the rotation angle corresponding to the constellation diagram;

Based on the rotation angle, obtain the frequency offset value corresponding to the first optical carrier;

Based on the frequency offset value, the frequency offset value used for frequency offset compensation processing is updated.
The method according to any one of claims 5 to 8, characterized in that before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, it further includes:

Obtain a second complex digital signal of a third polarization state corresponding to the first optical carrier, the second complex digital signal is received before the first complex digital signal, and the second complex digital signal includes the The training sequence after differential polarization time encoding corresponding to the third polarization state, the training sequence includes a first sub-training sequence and a second sub-training sequence after differential polarization time encoding, the first sub-training sequence and the third sub-training sequence The periods of the two sub-training sequences are different;

Based on the training sequence, determine a frame header position in the second complex digital signal;

Based on the frame header position and the length of the differential polarization time encoded training sequence, in the second complex digital signal, obtain the first received sequence corresponding to the training sequence;

Based on the first received sequence and the reference frequency offset value range, an initial frequency offset value is determined, and the initial frequency offset value is used to initially perform frequency offset compensation processing.
The method according to any one of claims 5 to 9, characterized in that before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, it further includes:

Obtain a second complex digital signal of a third polarization state corresponding to the first optical carrier, the second complex digital signal is received before the first complex digital signal, and the second complex digital signal includes the The training sequence after differential polarization time encoding corresponding to the third polarization state;

Determine the frame header position in the second complex digital signal;

Based on the frame head position and the length of the differential polarization time-encoded training sequence corresponding to the third polarization state, in the second complex digital signal, obtain the second received sequence corresponding to the training sequence;

Based on the second reception sequence and the Alamouti coding rule, restore the third reception sequence corresponding to the fourth polarization state, where the third reception sequence is the training sequence corresponding to the fourth polarization state;

Perform differential polarization time decoding on the second receiving sequence and the third receiving sequence to obtain decoded training sequences corresponding to the third polarization state and the fourth polarization state respectively;

Determining the error of the decoded training sequence corresponding to the third polarization state and the training sequence before differential polarization time encoding corresponding to the first polarization state, and the decoded training sequence corresponding to the fourth polarization state and the second polarization state. The error of the corresponding training sequence before differential polarization time encoding is used to obtain a third symbol error, and the first polarization state and the second polarization state are the two orthogonal polarization states used when transmitting the first optical carrier;

Based on the third symbol error and the chain rule, a fourth symbol error is determined, and the fourth symbol error is used to characterize the differential polarization time-encoded training sequence corresponding to the second received sequence and the first polarization state. The error and/or the error of the training sequence after differential polarization time encoding corresponding to the third received sequence and the second polarization state;

Based on the fourth symbol error, the tap coefficients used for the SISO equalization processing are updated.
A device for data transmission, characterized in that the device is applied to a coherent optical communication system, and the device includes:

An acquisition unit, configured to acquire a symbol sequence corresponding to the first optical carrier and a training sequence corresponding to two polarization states, the two polarization states being orthogonal, and the symbol sequence corresponding to the first optical carrier to be sent. The data is obtained by symbol mapping;

A differential polarization time encoding unit, configured to perform differential polarization time encoding on the symbol sequence and the training sequence respectively to obtain a dual-polarization complex digital signal corresponding to the first optical carrier. In the dual-polarization complex digital signal, The encoded training sequence is located at the beginning of each data frame;

A modulation unit configured to modulate the dual-polarization complex digital signal onto the first optical carrier according to dual-polarization in-phase quadrature IQ modulation;

A sending unit, configured to send the first optical carrier modulated with the dual-polarization complex digital signal.
The device according to claim 11, characterized in that the differential polarization time encoding unit is used for:

Based on the symbol sequence and the symbols of the two polarization states corresponding to each time slot in the training sequence, a coding matrix corresponding to each time slot is generated, and the symbols of the two polarization states corresponding to each time slot are respectively. Describe a row or column in the coding matrix corresponding to each time slot;

Based on the coding matrix corresponding to the T-th time slot, the differential polarization time coding matrix corresponding to the T-1th time slot, and the scaling factor corresponding to the T-th time slot, the differential polarization corresponding to the T-th time slot is obtained Time encoding matrix, the dual-polarization complex digital signal is obtained by sequentially connecting the differential polarization time encoding matrices corresponding to each time slot, and the scaling factor corresponding to the T-th time slot is obtained by the T-1th time slot The value of the determinant of the corresponding encoding matrix or differential polarization time encoding matrix is determined, and T is an integer greater than or equal to 2.
The device according to claim 11, wherein the two polarization states include a first polarization state and a second polarization state;

The differential polarization time encoding unit is used for:

Perform differential polarization time encoding on the symbol sequence corresponding to the first polarization state and the second polarization state;

Differential polarization time encoding is performed on the training sequences corresponding to the first polarization state and the second polarization state.
The device according to any one of claims 11 to 13, wherein the training sequences respectively corresponding to the two polarization states include a first sub-training sequence and a second sub-training sequence, and the first sub-training sequence and The periods of the second sub-training sequences are different.
A data receiving device, characterized in that the device is applied to a coherent optical communication system, and the device includes:

An acquisition unit, configured to acquire the first complex digital signal of the third polarization state corresponding to the first optical carrier;

a frequency offset compensation unit, configured to perform frequency offset compensation processing on the first complex digital signal;

An equalization processing unit is used to perform single-input single-output SISO equalization processing on the signal after frequency offset compensation processing;

A differential polarization time decoding unit, configured to perform differential polarization time decoding on the equalized signal to obtain symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, and the third polarization state and the fourth polarization state. state orthogonality;

A data recovery unit, configured to recover data corresponding to the third polarization state and the fourth polarization state based on the symbol sequence.
The device according to claim 15, characterized in that the differential polarization time decoding unit is used for:

Based on the symbols of the two third polarization states included in each time slot group in the equalized signal and the Alamouti coding rule, two symbols of the fourth polarization state included in each time slot group are generated. The two symbols of the third polarization state and the two symbols of the fourth polarization state constitute the receiving end matrix corresponding to each time slot group, and each time slot group is based on two consecutive time slots as a unit. The signals after equalization are grouped and obtained;

Based on the receiving end matrix corresponding to the nth time slot group, the receiving end matrix corresponding to the n-1th time slot group, and the scaling factor corresponding to the nth time slot group, the decoded matrix corresponding to the nth time slot group is obtained, The scaling factor corresponding to the n-th time slot group is determined by the value of the determinant of the receiving end matrix corresponding to the n-1th time slot group or the value of the determinant of the decoded matrix, where n is greater than or equal to 2. integer;

In the decoded matrix corresponding to the n-th time slot group, the symbols of the two polarization states of the time slot before differential polarization time encoding corresponding to the n-th time slot group are obtained.
The device according to claim 15 or 16, characterized in that the device further includes a tap coefficient update unit for:

Determine the symbol sequences corresponding to the third polarization state and the fourth polarization state respectively, and determine the first symbol error. The first symbol error is used to characterize the differential polarization time corresponding to the third polarization state after decoding. The error between the symbol sequence before differential polarization time encoding corresponding to the first polarization state and the error between the symbol sequence after differential polarization time encoding corresponding to the fourth polarization state and the symbol sequence before differential polarization time encoding corresponding to the second polarization state Error in symbol sequence;

Based on the first symbol error and the chain rule, a second symbol error is determined. The second symbol error is used to characterize the symbol sequence before differential polarization time decoding corresponding to the third polarization state and the first polarization state. The error between the corresponding symbol sequence after differential polarization time encoding and/or the error between the symbol sequence before differential polarization time decoding corresponding to the fourth polarization state and the symbol sequence after differential polarization time encoding corresponding to the second polarization state ;

Based on the second symbol error, update the tap coefficients used for the SISO equalization process;

Wherein, the first polarization state and the second polarization state are two orthogonal polarization states used when transmitting the first optical carrier.
The device according to any one of claims 15 to 17, characterized in that the device further includes a frequency offset estimation unit for:

Obtain the constellation diagram of the symbol sequence corresponding to the fourth polarization state, and determine the rotation angle corresponding to the constellation diagram;

Based on the rotation angle, obtain the frequency offset value corresponding to the first optical carrier;

Based on the frequency offset value, the frequency offset value used for frequency offset compensation processing is updated.
The device according to any one of claims 15 to 18, characterized in that the acquisition unit is also used to:

Before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, obtain the second complex digital signal of the third polarization state corresponding to the first optical carrier, and the second complex digital signal is in the The first complex digital signal was previously received. The second complex digital signal includes a differential polarization time-encoded training sequence corresponding to the third polarization state. The training sequence includes a differential polarization time-encoded first sub-digit signal. The training sequence and the second sub-training sequence, the periods of the first sub-training sequence and the second sub-training sequence are different;

The device further includes a framing unit configured to determine the frame header position in the second complex digital signal based on the training sequence;

The device also includes a frequency offset estimation unit for:

Based on the frame header position and the length of the differential polarization time encoded training sequence, in the second complex digital signal, obtain the first received sequence corresponding to the training sequence;

Based on the first received sequence and the reference frequency offset value range, an initial frequency offset value is determined, and the initial frequency offset value is used to initially perform frequency offset compensation processing.
The device according to any one of claims 15 to 19, characterized in that the acquisition unit is also used to:

Before obtaining the first complex digital signal of the third polarization state corresponding to the first optical carrier, obtain the second complex digital signal of the third polarization state corresponding to the first optical carrier, and the second complex digital signal is in the The first complex digital signal was previously received, and the second complex digital signal includes the differential polarization time-encoded training sequence corresponding to the third polarization state;

The device further includes a framing unit for determining the frame header position in the second complex digital signal;

The differential polarization time decoding unit is also configured to, based on the frame head position and the length of the differential polarization time encoded training sequence corresponding to the third polarization state, in the second complex digital signal, take the The second received sequence corresponding to the training sequence;

Based on the second reception sequence and the Alamouti coding rule, restore the third reception sequence corresponding to the fourth polarization state, where the third reception sequence is the training sequence corresponding to the fourth polarization state;

Perform differential polarization time decoding on the second receiving sequence and the third receiving sequence to obtain decoded training sequences corresponding to the third polarization state and the fourth polarization state respectively;

The device also includes a tap coefficient update unit for:

Determining the error of the decoded training sequence corresponding to the third polarization state and the training sequence before differential polarization time encoding corresponding to the first polarization state, and the decoded training sequence corresponding to the fourth polarization state and the second polarization state. The error of the corresponding training sequence before differential polarization time encoding is used to obtain a third symbol error, and the first polarization state and the second polarization state are the two orthogonal polarization states used when transmitting the first optical carrier;

Based on the third symbol error and the chain rule, a fourth symbol error is determined, and the fourth symbol error is used to characterize the differential polarization time-encoded training sequence corresponding to the second received sequence and the first polarization state. The error and/or the error of the training sequence after differential polarization time encoding corresponding to the third received sequence and the second polarization state;

Based on the fourth symbol error, the tap coefficients used for the SISO equalization processing are updated.
A dual-polarization transmitter, characterized in that the dual-polarization transmitter includes a digital signal processor, a digital-to-analog converter, a modulation driver, a laser and a dual-polarization modulator;

The digital signal processor is used to perform the method described in any one of claims 1 to 4.
A single polarization receiver, characterized in that the single polarization receiver includes a local oscillator laser, a coupler, a detector, a transimpedance amplifier, an analog-to-digital converter and a digital signal processor;

The digital signal processor is used to perform the method according to any one of claims 5 to 10.