Optical forward error correction encoding method and apparatus
By performing forward error correction coding based on the mode dimension of optical signals, the time delay and cost problems caused by electronic FEC encoder conversion are solved, realizing efficient and low-loss optical forward error correction, which is suitable for matrix operations in optical fiber communication.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI JIAOTONG UNIV
- Filing Date
- 2025-05-14
- Publication Date
- 2026-06-23
AI Technical Summary
In existing technologies, electronic FEC encoders need to frequently perform optical-electrical-optical conversions in optical fiber communication, resulting in additional processing delays and increased system costs. Furthermore, existing optical matrix calculation schemes sacrifice time and bandwidth resources, affecting system performance.
Forward error correction coding is performed using the mode dimension based on optical signals. By utilizing mode beam splitters and demultiplexers, calculations are performed directly on the optical signals, reducing the number of signal state transitions. Low-density parity-check codes and Hamming codes are combined for encoding to achieve optical forward error correction.
It reduces power loss and additional processing latency during the computation process, supports homomorphic transmission of optical signals, has a small device size and high density, is easy to integrate on a large scale, has good controllability and flexibility, and has little impact on the overall system performance.
Smart Images
Figure CN120546792B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of optoelectronic chips and optical computing, and more specifically, to an optical forward error correction coding method and apparatus. Background Technology
[0002] With the booming development of data-driven applications such as autonomous driving, artificial intelligence, and cloud computing, various fields urgently need processing systems with higher computing speeds and higher information accuracy. However, the performance limits of integrated circuits are constrained by Moore's Law, making it impossible to continuously increase circuit density. Furthermore, higher density distribution of electronic devices leads to enhanced electromagnetic interference, causing increased losses, a sharp rise in heat dissipation requirements, and a series of other problems. Optical computing solutions based on optical signals can effectively alleviate these problems and provide considerable computing power. Linear matrix-vector multiplication operations, represented by convolution, dominate the computation process; therefore, linear computing devices that accelerate matrix operations can effectively improve the system's processing speed and reduce system computing costs.
[0003] In matrix operation-based applications, forward error correction (FEC) is considered in optical fiber communication. FEC is a commonly used technique for controlling error bits. This technique adds redundancy check bits, enabling the receiver to detect and recover correct symbols. Commercial FEC encoders are based on integrated electronic devices. In optical fiber communication, the electronic FEC encoder at the transceiver end needs to map the optical signal to the electrical domain for processing. The frequent access to electrical memory and other operations involved further increase system power loss. In some relay transmission scenarios, the electronic FEC encoder requires optical-to-electrical-to-optical conversion of the optical signal before processing can be completed. The conversion process leads to additional processing delays and increases system cost. The essence of matrix operation-based FEC coding schemes is the product of the generator matrix and the vector to be encoded in a finite field. Therefore, optical computation-based optical operation structures can be applied to the implementation of optical FEC encoders. Currently, most optical matrix computation schemes use time-division multiplexing, space-division multiplexing, and wavelength-division multiplexing to achieve parallel acceleration computation. These methods sacrifice time and bandwidth resources, affecting the overall system performance.
[0004] Patent application CN110263296A discloses a matrix-vector multiplier based on a photoelectric computing array and its operation method. The multiplier includes a photoelectric computing array and an analog-to-digital converter (ADC). The photoelectric computing array consists of m*n photoelectric computing units arranged periodically. Each photoelectric computing unit includes a light-emitting unit and a computing unit. Each computing unit includes a light input terminal and a result output terminal. Light emitted by the light-emitting unit is incident on the light input terminal of the corresponding computing unit. The result output terminals of each column of computing units are connected sequentially. The current signal output by each column of computing units is input to an ADC individually, or the current signals output by multiple columns of computing units are input to a single ADC. However, this patent cannot completely solve the existing technical problems, nor can it meet the needs of this invention. Summary of the Invention
[0005] In view of the deficiencies in the prior art, the purpose of this invention is to provide an optical forward error correction coding method and apparatus.
[0006] The optical forward error correction coding method provided by the present invention includes:
[0007] Step S1: Input the optical signal output from the laser and the electrical signal output from the signal generator into the electro-optic intensity modulator, and modulate the electrical signal carrying the vector to be processed onto the optical carrier to form an optical signal;
[0008] Step S2: Based on the matrix element values, change the phase of the phase shifter in the mode beam splitter to adjust the power ratio of different mode optical signals for transmission;
[0009] Step S3: The output optical signals of different modes are demultiplexed into multiple single-mode optical signals by a mode demultiplexer, and then transmitted to a photodetector to be converted into electrical signals.
[0010] Step S4: Sample, quantize, and sum all electrical signals to obtain the calculation results.
[0011] Preferably, step S1 includes: the signal generator transmitting signal x within the time interval [(i-1)τ, iτ). i , i = 1, 2, ..., n, x i Let x represent the i-th element of a 1×n input vector x, where x = [x1, x2, ..., xn]. n-1 ,x n ], τ represents the duration of a single bit, which is determined by the input signal rate of the signal generator.
[0012] Preferably, step S2 includes: the mode beam splitter changing the phase of the phase shifter inside the beam splitter, such that the optical signal power P of the j-th mode... j The following condition is satisfied within the time interval [(i-1)τ, iτ):
[0013]
[0014]
[0015] Where j = 1, 2, ..., k, m ij These are the element values of an n×k matrix M.
[0016] Preferably, step S3 includes: the number of output branches of the mode demultiplexer is equal to N, the light of each mode enters a single output branch, and a photodetector is connected after each output branch. The photodetector converts the optical signal into an electrical signal, and the intensity of the electrical signal output by the photodetector of the j-th branch is I within the time interval [(i-1)τ, iτ). ij ∝m ij x i .
[0017] Preferably, step S4 includes: a sampling interval equal to τ, a sampling sequence length equal to n, and a sampling sequence s for the j-th branch. j ′ ∝[m 1j x1,m 2j x1,…,m nj x n ], then for s j ′ The elements are summed, and the summation result is quantized and combined into a 1×k vector s = [s1, s2, ..., sk]. k-1 ,s k ],in s = xM, where s is the final result of matrix multiplication.
[0018] Preferably, the intensity of the output optical signal of the electro-optic modulator is proportional to the intensity of the input electrical signal;
[0019] The maximum number of modes N obtained by the mode beam splitter is not less than n, and the time for the mode beam splitter to change the mode power ratio is not greater than τ.
[0020] Preferably, optical forward error correction coding is performed using a low-density parity-check code generation matrix coding method and a Hamming code coding method.
[0021] The optical forward error correction coding device provided by the present invention includes a laser, a signal generator, an electro-optic intensity modulator, a mode beam splitter, a mode demultiplexer, and a photodetector.
[0022] The output terminal of the signal generator is electrically connected to the input terminal of the electro-optic intensity modulator, the output terminal of the laser is optically connected to the input terminal of the electro-optic intensity modulator, the output terminal of the electro-optic intensity modulator is optically connected to the input terminal of the mode beam splitter, the output terminal of the mode beam splitter is optically connected to the input terminal of the mode demultiplexer, the output terminal of the mode demultiplexer is optically connected to the input terminal of the photodetector, and the output terminal of the photodetector is the encoding result output terminal of the invention.
[0023] Preferably, the mode beam splitter is composed of alternating cascades of phase shifters and three-waveguide coupling structures;
[0024] The mode beam splitter achieves arbitrary power ratios of different mode optical signals by changing the phase of the corresponding phase shifter;
[0025] In the mode beam splitter, the first three-waveguide coupling structure is a single-mode optical waveguide, and the other three-waveguide coupling structures are multimode optical waveguides that can accommodate a corresponding number of optical signal modes.
[0026] Preferably, the mode beam splitter is fabricated based on a silicon optical waveguide platform, and the phase shifter employs thermo-optic modulation.
[0027] Compared with the prior art, the present invention has the following beneficial effects:
[0028] 1. This invention realizes a forward error correction encoder based on the mode dimension of optical signals. The encoder is based on an optoelectronic hardware platform, which reduces the number of signal state transitions and the required electronic storage devices, supports the homomorphic transmission of optical signals and the effect of transmission-computation fusion technology, thereby effectively reducing power loss and additional processing latency during the calculation process.
[0029] 2. This invention is mainly based on on-chip integrated devices, which can be easily fabricated using mature processes and have stable performance; the device is small in size and high in density, and is easy to integrate on a large scale.
[0030] 3. This invention allows users to set weight values by changing the power ratio between different modes of optical signals. It is simple to operate and has good controllability and flexibility.
[0031] 4. This invention utilizes the physical dimension of light patterns, requiring no additional resources such as system bandwidth, and has minimal impact on the overall system performance;
[0032] 5. By utilizing the technical characteristics of optical signals, such as their large bandwidth, this invention can achieve potential technical effects such as low error and high speed. Attached Figure Description
[0033] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0034] Figure 1 A schematic diagram of an optical matrix operation structure;
[0035] Figure 2 This is a schematic diagram of an embodiment of a forward error correction coding device based on a mode selection switch;
[0036] Figures 3a to 3d These are partial model field simulation results of the phase applied by the four phase shifters PS0, PS1, PS2, and PS3 and their usage modes;
[0037] Figures 4a to 4d These are the output curves of different photodetectors, representing the results of different modes of optical signals.
[0038] In the diagram, 100 - electrical connection, 110 - optical connection, 120 - laser, 130 - signal generator, 140 - electro-optic intensity modulator, 150 - mode beam splitter, 160 - mode demultiplexer, 170 - photodetector; 2000 - optical connection, 2010 - electrical connection, Laser - laser, MZI - electro-optic Mach-Zehnder intensity modulator, SG1 - signal generator to be encoded, Y_branch - Y-type 1:2 power divider waveguide, PS0, PS1, PS2, PS3 - phase shifter modules, SG2, SG3, SG4, SG5 - phase shifter module control signal generator, TWC1, TWC2, TWC3, TWC4 - three-waveguide coupling structure, DEMUX - mode demultiplexer, PD1, PD2, PD3, PD4 - photodetectors, LPF1, LPF2, LPF3, LPF4 - low-pass filters. Detailed Implementation
[0039] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.
[0040] Example
[0041] The purpose of this invention is to provide a forward error correction coding method and apparatus based on a mode selection switch. It is mainly based on on-chip integrated devices, which are small in size, high in density, and easy to integrate on a large scale. This invention utilizes the mode dimension of light, does not occupy additional frequency band resources, and has little impact on the overall performance of the system. This invention fully utilizes the advantages of high speed, low electromagnetic interference, etc. of optical signals, providing users with better controllability and flexibility.
[0042] A preferred embodiment of the present invention is illustrated in the figure below. Figure 1 and Figure 2As shown, it includes: a laser, an electro-optic intensity modulator (MZI), a signal generator (SG1) to be encoded, a Y-type 1:2 power divider waveguide (Y_branch), phase shifter modules (PS0, PS1, PS2, PS3), phase shifter module control signal generators (SG2, SG3, SG4, SG5), three-waveguide coupling structures (TWC1, TWC2, TWC3, TWC4), a mode demultiplexer (DEMUX), photodetectors (PD1, PD2, PD3, PD4), and low-pass filters (LPF1, LPF2, LPF3, LPF4).
[0043] Specifically, the laser is an activation device that inputs an initial optical signal to the optical forward error correction coding device based on the mode selection switch. The continuous input of optical signals maintains the operation of the device. In this embodiment, the center wavelength of the input optical signal is selected as 1550nm.
[0044] Specifically, the electro-optic intensity modulator MZI is based on a Mach-Zehnder intensity modulator and is used to load the electrical signal output by the signal generator SG1 to be encoded into the output optical signal of the laser.
[0045] Specifically, the center waveguide of the TWC1 three-waveguide coupled structure is a single-mode waveguide, supporting only the TE0 mode; the center waveguide of the TWC2 three-waveguide coupled structure is a multimode waveguide, supporting TE0 and TE1 modes; the center waveguide of the TWC3 three-waveguide coupled structure is a multimode waveguide, supporting TE0, TE1, and TE2 modes; and the center waveguide of the TWC4 three-waveguide coupled structure is a multimode waveguide, supporting TE0, TE1, TE2, and TE3 modes.
[0046] Specifically, the mode demultiplexer DEMUX has four output terminals, which correspond to the output of the four modes TE0, TE1, TE2, and TE3 after demultiplexing, from top to bottom.
[0047] The principle of the optical forward error correction coding scheme implemented through the mode selection switch in this embodiment is as follows:
[0048] The electro-optic intensity modulator MZI modulates the optical signal with a center wavelength of 1550nm output from the laser based on the electrical signal output from the signal generator SG1. In this embodiment, the electrical signal output corresponding to the "0" code is modulated to the minimum value of the MZI output light intensity, and the electrical signal output corresponding to the "1" code is modulated to the maximum value of the MZI output light intensity. Subsequently, the output signal of the MZI is split into two optical signals of equal intensity after passing through the Y-type 1:2 power divider waveguide Y_branch.
[0049] For the generator matrix in this embodiment Where I is a 4*4 identity matrix, and its encoding process is a copy of the original information code. Q is a 4*4 ordinary matrix, corresponding to the generation of the check code sequence. The encoding process in this embodiment will focus on the calculation of the Q matrix. The Q matrix has four columns, which are assigned to the four modes TE0, TE1, TE2, and TE3 from left to right. A matrix element of "1" represents the use of the mode, and an element of "0" represents the non-use of the mode. For the matrix Q in this embodiment, the four elements in the first row are "1110", which means that when the optical signal with the first information bit is loaded is input, the mode selection switch outputs the mode of TE0, TE1, and TE2. At this time, the phase loading of the four phase shifters is: PS0 = 7π / 12, PS1 = -π / 2, PS2 = -π / 12, PS3 = 0. The mode field output obtained through simulation is as follows. Figure 3a As shown; similarly, the four elements in the second row of matrix Q are "1101", which means that when the optical signal with the second information bit is input, the mode selection switch outputs the modes TE0, TE1, and TE3. At this time, the phase loading of the four phase shifters is: PS0 = 7π / 12, PS1 = -π / 2, PS2 = 11π / 12, PS3 = 0. The mode field output obtained through simulation is as follows. Figure 3b As shown; the four elements in the third row of matrix Q are "1011", which means that when the optical signal with the third information bit is loaded is input, the mode selection switch outputs the modes TE0, TE2, and TE3. At this time, the phase loading of the four phase shifters is: PS0 = 7π / 12, PS1 = π, PS2 = π / 2, PS3 = π / 12. The mode field output obtained through simulation is as follows. Figure 3c As shown; the four elements in the fourth row of matrix Q are "0111", which means that when the optical signal with the third information bit is loaded is input, the mode selection switch outputs the modes TE1, TE2, and TE3. At this time, the phase loading of the four phase shifters is: PS0 = π, PS1 = -7π / 12, PS2 = π / 2, PS3 = π / 9. The mode field output obtained through simulation is as follows. Figure 3d As shown.
[0050] With a period of 4 information bits, the above cycle is repeated with subsequent information bit inputs. That is, the mode selection switch will cyclically output four mode combinations: (TE0, TE1, TE2), (TE0, TE1, TE3), (TE0, TE2, TE3), and (TE1, TE2, TE3). Finally, the mode demultiplexer DEMUX demultiplexes the outputs of the four modes TE0, TE1, TE2, and TE3. After sampling and threshold decision, the output of each mode is modulo-2 summed with 4 bits, and the parallel outputs yield the parity check code. In this example, the maximum number of modes of the mode selection switch corresponds one-to-one with the number of columns in Q, i.e., the number of parity bits. Multiplexing the four modes TE0, TE1, TE2, and TE3 corresponds to the generation of 4 parity bits. To achieve encoding of more parity bits, a phase shifter and a TWC three-waveguide coupling structure can be added to the structure of this example, allowing the mode selection switch to cover more modes.
[0051] According to the optical forward error correction coding method based on a mode selection switch provided by the present invention, the optical forward error correction coding device based on the mode selection switch performs the following steps:
[0052] Step 1: Generation of electrical signals and electro-optical conversion. In this embodiment, the transmission signal to be encoded is s = [10100101]. The transmission signal generator SG1 generates an electrical signal, where "1" corresponds to a high-level electrical signal and "0" corresponds to a low-level electrical signal. Here, we only focus on the encoding of the parity check bits. That is... Subsequently, the electrical signal generated by the signal generator SG1 is output to the electro-optic intensity modulator MZI to achieve electro-optic conversion. The "0" code corresponds to the minimum output light intensity, and the "1" code corresponds to the maximum output light intensity. The electro-optic intensity modulator adopts a push-pull structure and can be implemented based on material platforms such as silicon and lithium niobate.
[0053] Step Two: Phase Control of the Phase Shifter and Optical Signal Transmission. The phase shifter module control signal generator SG2 outputs an electrical signal to control the phase of the PS0 phase shifter as follows: PS0 = [7π / 12, 7π / 12, 7π / 12, π, 7π / 12, 7π / 12, 7π / 12, π], corresponding to the eight-bit signal to be encoded and transmitted; the phase shifter module control signal generator SG3 outputs an electrical signal to control the phase of the PS1 phase shifter as follows: PS1 = [-π / 2, -π / 2, π, -7π / 12, -π / 2, -π / 2, π, -7π / 12]; the phase shifter module control signal generator SG4 outputs an electrical signal to control the phase of the PS2 phase shifter as follows: PS2 = [-π / 12, 11π / 12, π / 2, π / 2, -π / 12, 11π / 12, π / 2, π / 2]; The phase shifter module controls the phase of the PS3 phase shifter by outputting an electrical signal from the signal generator SG5: PS3 = [0, 0, π / 12, π / 9, 0, 0, π / 12, π / 9]; Based on the regulation of the four phase shifters, the mode selection switch outputs different mode combinations according to the input bit stream, that is, the output of the center waveguide of the TWC4 three-waveguide coupled structure will contain optical signals of four modes: TE0, TE1, TE2, and TE3. Due to the difference in weight configuration, the optical signals of different modes have different energies.
[0054] Step 3: Demultiplexing of optical signals and conversion of optical signals of different modes into electrical signals. The signal output from the center waveguide of the TWC4 three-waveguide coupled structure is input into the mode demultiplexer DEMUX. The four outputs of the mode demultiplexer DEMUX, from top to bottom, correspond to the optical signals of the four modes TE0, TE1, TE2, and TE3, respectively. They are then transmitted through single-mode optical fibers to photodetectors PD1, PD2, PD3, and PD4, respectively, and converted into electrical signals. The amplitude of the output electrical signal of the photodetector is proportional to the light intensity of the input optical signal.
[0055] Step 4: Decoding and processing of electrical signals. The output signals of low-pass filters LPF1, LPF2, LPF3, and LPF4 are sampled, amplified, and processed for decision-making. The output signal of low-pass filter LPF1 corresponds to the output of TE0 mode, as shown below. Figure 4a As shown, the processing result of TE0 mode is "01"; the output signal of low-pass filter LPF2 corresponds to the output of TE1 mode, as follows. Figure 4b As shown, the processing result of TE1 mode is "10"; the output signal of low-pass filter LPF3 corresponds to the output of TE2 mode, as follows. Figure 4c As shown, the processing result of TE2 mode is "01"; the output signal of low-pass filter LPF4 corresponds to the output of TE3 mode, as follows. Figure 4dAs shown, the processing result of TE3 mode is "10"; therefore, the parallel output result of this embodiment is "01011010", which is consistent with the theoretical analysis; by performing a 4-bit modulo 2 addition operation on the output result of each mode, the final result of the supervision code can be obtained by parallel output.
[0056] In the description of this application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0057] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.
[0058] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.
Claims
1. An optical forward error correction coding method, characterized in that, include: Step S1: Input the optical signal output from the laser and the electrical signal output from the signal generator into the electro-optic intensity modulator to modulate the electrical signal carrying the vector to be processed onto the optical signal; Step S2: Based on the matrix element values, change the phase of the phase shifter in the mode beam splitter to adjust the power ratio of different mode optical signals for transmission; Step S3: The output optical signals of different modes are demultiplexed into multiple single-mode optical signals by a mode demultiplexer, and then transmitted to a photodetector to be converted into electrical signals. Step S4: Sample, quantize, and sum all electrical signals to obtain the calculation results; Step S1 includes: the signal generator in Send signal within time , Let represent the i-th element of the 1×n input vector x. , The duration of a single bit is determined by the input signal rate of the signal generator. Step S2 includes: the mode beam splitter changes the phase of the phase shifter inside the beam splitter, so that the first... Optical signal power of each mode exist Satisfy within the time limit: in, , These are the element values of an n×k matrix M.
2. The optical forward error correction coding method according to claim 1, characterized in that, Step S3 includes: the mode demultiplexer has N output branches; light from each mode enters a single output branch; each output branch is connected to a photodetector; and the photodetector converts the optical signal into an electrical signal. Within a time period, the first The output electrical signal strength of the photodetector in each branch .
3. The optical forward error correction coding method according to claim 2, characterized in that, Step S4 includes: the sampling interval equals The sampling sequence length is equal to , No. Sampling sequence of each branch ], and then to The elements are summed, and the summation result is quantized and combined into a 1×k vector. ,in , s is the final result of matrix multiplication.
4. The optical forward error correction coding method according to claim 1, characterized in that, The intensity of the output optical signal of the electro-optic intensity modulator is proportional to the intensity of the input electrical signal. The maximum number of modes N obtained by the mode beam splitter is not less than n, and the time for the mode beam splitter to change the mode power ratio is not greater than [missing value]. .
5. The optical forward error correction coding method according to claim 1, characterized in that, Optical forward error correction coding is performed using a low-density parity-check code generation matrix coding method and a Hamming code coding method.
6. An optical forward error correction coding device, characterized in that, The device employing the optical forward error correction coding method according to any one of claims 1 to 5 comprises a laser, a signal generator, an electro-optic intensity modulator, a mode beam splitter, a mode demultiplexer, and a photodetector. The output terminal of the signal generator is electrically connected to the input terminal of the electro-optic intensity modulator, the output terminal of the laser is optically connected to the input terminal of the electro-optic intensity modulator, the output terminal of the electro-optic intensity modulator is optically connected to the input terminal of the mode beam splitter, the output terminal of the mode beam splitter is optically connected to the input terminal of the mode demultiplexer, the output terminal of the mode demultiplexer is optically connected to the input terminal of the photodetector, and the output terminal of the photodetector is the output terminal of the encoding result.
7. The optical forward error correction coding device according to claim 6, characterized in that, The mode beam splitter is composed of alternating cascades of phase shifters and three waveguide coupling structures; The mode beam splitter achieves arbitrary power ratios of different mode optical signals by changing the phase of the corresponding phase shifter; In the mode beam splitter, the first three-waveguide coupling structure is a single-mode optical waveguide, and the other three-waveguide coupling structures are multimode optical waveguides that can accommodate a corresponding number of optical signal modes.
8. The optical forward error correction coding device according to claim 7, characterized in that, The mode beam splitter is fabricated based on a silicon optical waveguide platform, and the phase shifter uses thermo-optic modulation.