Frequency domain equalization method and device based on matrix inversion, equipment and storage medium

Abstract

The invention discloses a frequency domain equalization method, device and equipment based on matrix inversion, and a storage medium, relates to the technical field of signal processing, and discloses a frequency domain equalization method based on matrix inversion, and the method comprises the steps: carrying out the transmitting end signal processing of a pseudo-random binary sequence at a transmitting end based on a constant amplitude zero autocorrelation sequence, and obtaining a transmitting end optical signal; receiving the sending end optical signal transmitted through the space division multiplexing channel at a receiving end to obtain a receiving end optical signal, and performing receiving end signal processing on the receiving end optical signal to obtain a to-be-balanced signal; determining a current channel matrix of a space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized; and performing third-order iteration inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the to-be-balanced signal based on the current inverse matrix to obtain a frequency domain balanced output signal. According to the method, the iteration inversion convergence speed in the frequency domain equalization process can be accelerated, and the channel equalization efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of signal processing technology, and in particular to a frequency domain equalization method, apparatus, device and storage medium based on matrix inversion. Background Art

[0002] Space Division Multiplexing (SDM) technology can significantly increase channel capacity through the parallel transmission capabilities of multi-core or multimode optical fibers. However, as SDM communication systems continue to scale, the complexity of digital signal processing (DSP) at the receiving end also increases significantly. Compared with traditional single-mode systems, crosstalk between different spatial channels (modes or cores) in SDM communication systems is more severe, causing the computational burden of the equalizer to increase exponentially with the multiplexing dimension.

[0003] Currently, mainstream frequency-domain multiple-input multiple-output (FD-MIMO) equalization techniques mostly use zero-forcing (ZF) or minimum mean-square error (MMSE) criteria. These techniques require high-dimensional matrix inversion for each frequency bin, significantly increasing the computational burden on the receiver. In high-dimensional SDM scenarios, as the matrix dimension increases, the traditional Newton-Schulz Iteration (NSI) algorithm requires a higher number of iterations to achieve convergence accuracy, significantly increasing processing latency.

[0004] In the presence of significant mode-dependent loss (MDL), the channel matrix condition number deteriorates, causing the NSI convergence rate to decrease or even fail. Therefore, accelerating the iterative inversion convergence of frequency-domain equalization in high-dimensional SDM communication systems to improve channel equalization efficiency has become an urgent problem.

[0005] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0006] The main purpose of this application is to provide a frequency domain equalization method, device, equipment and storage medium based on matrix inversion, aiming to solve the technical problem of how to accelerate the convergence speed of iterative inversion in the frequency domain equalization process for high-dimensional SDM communication systems to improve the channel equalization efficiency.

[0007] To achieve the above objectives, the present application proposes a frequency domain equalization method based on matrix inversion, wherein the frequency domain equalization method based on matrix inversion is applied to a space division multiplexing communication system, wherein the space division multiplexing communication system includes a transmitting end, a space division multiplexing channel, and a receiving end, wherein the transmitting end is connected to the space division multiplexing channel, and the space division multiplexing channel is connected to the receiving end;

[0008] The frequency domain equalization method based on matrix inversion includes:

[0009] Performing transmitter signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitter to obtain a transmitter optical signal;

[0010] receiving, at the receiving end, the transmitting end optical signal transmitted through the space division multiplexing channel to obtain a receiving end optical signal, and performing receiving end signal processing on the receiving end optical signal to obtain a signal to be equalized;

[0011] Determining a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized;

[0012] Performing a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

[0013] In one embodiment, the step of performing a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal includes:

[0014] Calculating a first norm and a second norm based on the current channel matrix, wherein the first norm is a first norm of the current channel matrix, and the second norm is an infinite norm of the current channel matrix;

[0015] Calculating an initial inverse matrix of the current channel matrix based on the first norm and the second norm;

[0016] Performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step, the identity matrix, and the initial inverse matrix to obtain a current inverse matrix;

[0017] Performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a current equalized signal;

[0018] A frequency domain equalization output signal is obtained based on the number of matrix iterative inversion times and the current equalization signal.

[0019] In one embodiment, the step of performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step size, the identity matrix, and the initial inverse matrix to obtain the current inverse matrix includes:

[0020] Calculate the product of the convergence adjustment step and the unit matrix to obtain the first matrix;

[0021] Calculating the product of the initial inverse matrix and the current channel matrix to obtain a second matrix;

[0022] determining a first-order inversion matrix based on the first matrix and the second matrix;

[0023] Determine a second-order inverse matrix based on the first matrix, the second matrix, and the first-order inverse matrix;

[0024] A current inverse matrix is ​​calculated based on the current channel matrix and the second-order inverse matrix.

[0025] In one embodiment, the step of obtaining a frequency domain equalized output signal based on the number of matrix iterative inversions and the current equalized signal includes:

[0026] When the number of matrix iterative inversions is less than a preset number of iterations, updating the current channel matrix based on the constant amplitude zero autocorrelation sequence and the current equalized signal;

[0027] Determining the current inverse matrix as the updated initial inverse matrix, and returning to the step of performing a third-order iterative inversion of the current channel matrix according to the convergence adjustment step, the identity matrix, and the initial inverse matrix to obtain the current inverse matrix;

[0028] When the number of iterative inversions of the matrix reaches the preset number of iterations, it is determined that the current equalized signal is a frequency domain equalized output signal.

[0029] In one embodiment, the step of performing transmitter signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitter to obtain a transmitter optical signal includes:

[0030] Performing digital signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal;

[0031] A transmitting end optical signal is obtained based on the laser optical signal, the spontaneous emission optical signal and the pre-processed signal.

[0032] In one embodiment, the step of performing transmitting-end digital signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal includes:

[0033] Mapping the pseudo-random binary sequence into quadrature amplitude modulation symbols at the transmitting end;

[0034] Inserting a constant amplitude zero autocorrelation sequence into the orthogonal amplitude modulation symbol to obtain a frame structure sequence;

[0035] Performing interpolation processing on the frame structure sequence to obtain an up-sampled sequence;

[0036] Perform frequency domain shaping on the up-sampled sequence to obtain a preprocessed signal.

[0037] In one embodiment, the step of obtaining the transmitting end optical signal based on the laser optical signal, the spontaneous emission optical signal, and the preprocessed signal includes:

[0038] converting the preprocessed signal into an analog waveform;

[0039] performing dual-polarization modulation on the laser optical signal based on the analog waveform to obtain a modulated optical signal;

[0040] Combining the spontaneous emission light signal and the modulated light signal to obtain a combined light signal;

[0041] The combined optical signals of multiple transmitting ends are coupled to obtain a transmitting end optical signal.

[0042] In addition, to achieve the above-mentioned purpose, the present application also proposes a frequency domain equalization device based on matrix inversion, wherein the frequency domain equalization device based on matrix inversion comprises:

[0043] a transmitting end signal processing module, configured to perform transmitting end signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitting end optical signal;

[0044] a receiving-end signal processing module, configured to receive, at a receiving end, the transmitting-end optical signal transmitted through the space-division multiplexing channel to obtain a receiving-end optical signal, and perform receiving-end signal processing on the receiving-end optical signal to obtain a signal to be equalized;

[0045] A channel estimation module, configured to determine a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized;

[0046] The frequency domain equalization module is used to perform a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and perform linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

[0047] In addition, to achieve the above-mentioned purpose, the present application also proposes a frequency domain equalization device based on matrix inversion, which includes: a memory, a processor, and a computer program stored on the memory and runnable on the processor, wherein the computer program is configured to implement the steps of the frequency domain equalization method based on matrix inversion as described above.

[0048] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the frequency domain equalization method based on matrix inversion as described above are implemented.

[0049] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the frequency domain equalization method based on matrix inversion as described above.

[0050] One or more technical solutions proposed in this application have at least the following technical effects:

[0051] Accurate channel estimation can be achieved through a constant amplitude zero autocorrelation sequence, providing a reliable current channel matrix for frequency domain equalization. Third-order iterative inversion can reduce the number of matrix inversion iterations and computational complexity, allowing the current inverse matrix to be quickly and stably obtained even in high-dimensional spatial division multiplexing scenarios or in the presence of strong pattern-correlated losses. Combined with linear detection, efficient frequency domain equalization can be achieved, effectively eliminating crosstalk and distortion in the spatial division multiplexing channel. The resulting frequency domain equalized output signal is high-quality and can accurately restore the original information. The overall process improves the transmission efficiency and real-time performance of the spatial division multiplexing communication system, enabling it to maintain good stability while supporting high-capacity communication needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0053] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0054] Figure 1 A flowchart of the first embodiment of the frequency domain equalization method based on matrix inversion provided in this application;

[0055] Figure 2This is a diagram of the spatial division multiplexing communication system architecture provided in Example 1 of the frequency domain equalization method based on matrix inversion of the present application;

[0056] Figure 3 A flowchart of the second embodiment of the frequency domain equalization method based on matrix inversion provided by this application;

[0057] Figure 4 This is a flowchart of the application of the M-NSI algorithm of the frequency domain equalization method based on matrix inversion provided in Example 2 of the present application;

[0058] Figure 5 Schematic diagram of the effects of the NSI and M-NSI algorithms of the frequency domain equalization method based on matrix inversion provided in Example 2 of the present application;

[0059] Figure 6 This is a schematic diagram of the module structure of a frequency domain equalization device based on matrix inversion according to an embodiment of the present application;

[0060] Figure 7 Schematic diagram of the device structure of the hardware operating environment involved in the frequency domain equalization method based on matrix inversion in the embodiment of the present application.

[0061] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0063] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0064] The main solution of the embodiment of the present application is: at the transmitting end, a pseudo-random binary sequence is subjected to transmitting end signal processing based on a constant amplitude zero autocorrelation sequence to obtain a transmitting end optical signal; at the receiving end, the transmitting end optical signal transmitted through the spatial division multiplexing channel is received to obtain a receiving end optical signal, and the receiving end optical signal is subjected to receiving end signal processing to obtain a signal to be equalized; the current channel matrix of the spatial division multiplexing channel is determined based on the constant amplitude zero autocorrelation sequence and the signal to be equalized; the current channel matrix is ​​inverted by a third-order iterative method to obtain a current inverse matrix, and the signal to be equalized is linearly detected based on the current inverse matrix to obtain a frequency domain equalized output signal.

[0065] In this embodiment, for ease of description, the following description is made by taking identification of a space division multiplexing communication system as the execution subject.

[0066] The present application provides a solution that can achieve accurate channel estimation through a constant amplitude zero autocorrelation sequence, providing a reliable current channel matrix for frequency domain equalization; the third-order iterative inversion can reduce the number of iterations and computational complexity of matrix inversion, and can quickly and stably obtain the current inverse matrix even in high-dimensional space division multiplexing scenarios or in the presence of strong mode correlation loss; combined with linear detection, efficient frequency domain equalization can be achieved, effectively eliminating the crosstalk and distortion of the space division multiplexing channel, so that the final frequency domain equalization output signal is of high quality and can accurately restore the original information. The overall process improves the transmission efficiency and real-time performance of the space division multiplexing communication system, and can maintain good stability while supporting large-capacity communication needs.

[0067] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device capable of implementing the above functions, a space-division multiplexing communication system, etc. The following uses a space-division multiplexing communication system as an example to illustrate this embodiment and the following embodiments.

[0068] Based on this, the embodiment of the present application provides a frequency domain equalization method based on matrix inversion, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the frequency domain equalization method based on matrix inversion of the present application.

[0069] In this embodiment, the frequency domain equalization method based on matrix inversion is applied to a space division multiplexing communication system, wherein the space division multiplexing communication system includes a transmitting end, a space division multiplexing channel, and a receiving end, wherein the transmitting end is connected to the space division multiplexing channel, and the space division multiplexing channel is connected to the receiving end;

[0070] The frequency domain equalization method based on matrix inversion includes steps S10 to S40:

[0071] Step S10, performing transmitting end signal processing on the pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitting end optical signal;

[0072] It should be understood that a space-division multiplexing communication system is a communication system that improves channel capacity through the parallel transmission capabilities of multi-core or multi-mode optical fibers. It consists of three parts: a transmitter, a space-division multiplexing channel, and a receiver. The transmitter is responsible for signal generation and processing, the space-division multiplexing channel is the medium for signal transmission, and the receiver is responsible for signal reception and further processing.

[0073] Reference Figure 2 , Figure 2 This is a diagram of the spatial division multiplexing communication system architecture provided by the first embodiment of the frequency domain equalization method based on matrix inversion in this application. Figure 2As shown, the transmitting end (TX1, TX2, etc.) includes a laser and a source of amplified spontaneous emission (ASE) for generating optical signals, as well as a transmitting-end digital signal processing module for generating electrical signals. These signals are processed by a 128GSa / s arbitrary waveform generator, a dual-polarization I / Q modulator, an erbium-doped fiber amplifier (EDFA), and a variable optical attenuator (VOA). After processing, they are transmitted through a coupler and fan-in equipment into a multi-core optical fiber, i.e., a spatial division multiplexing channel. The receiving end includes a receiving-end digital signal processing module for processing electrical signals. This module is connected to an 80GSa / s digital storage oscilloscope to obtain the electrical signals after optical-to-electrical conversion of the optical signals received by the fan-out equipment. The optical signals received by the fan-out equipment are processed by a splitter before being transmitted to the 80GSa / s digital storage oscilloscope and received by an integrated coherent receiver. The splitter also receives a reference signal provided by a local oscillator.

[0074] It should be noted that the Constant Amplitude Zero Auto-correlation (CAZAC) sequence is a sequence with constant amplitude and zero autocorrelation. Using the CAZAC sequence as a training sequence for channel estimation during communication can effectively distinguish useful signals from interference, improving the accuracy of channel estimation. A pseudo-random binary sequence (PRBS) is a binary sequence generated by a specific mathematical formula or circuit. It is repeatable and close to true random data, allowing comprehensive testing of system performance when transmitting different data modes. PRBS is used as a test signal for signal modulation in space-division multiplexing communication systems. Inserting a constant amplitude zero autocorrelation sequence into the pseudo-random binary sequence at the transmitter helps to more accurately estimate channel characteristics at the receiver.

[0075] Furthermore, after the pseudo-random binary sequence is generated at the transmitter, a constant amplitude zero autocorrelation sequence is inserted into the pseudo-random binary sequence, facilitating accurate channel estimation at the receiver. After digital signal processing is performed on the pseudo-random binary sequence with the constant amplitude zero autocorrelation sequence inserted, the processed signal can be used to modulate an optical signal to generate the transmitter optical signal.

[0076] In a feasible implementation, step S10 may include steps S11 to S12:

[0077] Step S11, performing digital signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal;

[0078] It should be understood that when digital signal processing is performed on a pseudo-random binary sequence, a constant amplitude zero autocorrelation sequence is inserted into a specific position within the pseudo-random binary sequence during the frame structure construction process to obtain a frame structure sequence. This frame structure sequence is then further digitally pre-processed. This pre-processed signal is a signal in the digital domain that has not yet been converted into an optical signal and serves as the input basis for subsequent electro-optical conversion.

[0079] In a feasible implementation, step S11 may include steps S111 to S114:

[0080] Step S111, mapping the pseudo-random binary sequence into orthogonal amplitude modulation symbols at the transmitting end;

[0081] It should be noted that the quadrature amplitude modulation symbols are complex symbols. Through the Quadrature Amplitude Modulation Mapping (QAM-Mapping) technology, the pseudo-random binary sequence can be mapped into complex symbols to obtain quadrature amplitude modulation symbols. Specifically, the quadrature amplitude modulation symbols can be 16-QAM modulation symbols. 16-QAM contains 16 different symbols, each symbol corresponding to a 4-bit binary number, such as 0000 corresponds to one amplitude and phase, and 0001 corresponds to another. Through 16-QAM modulation, every 4-bit binary number in the pseudo-random binary sequence will be converted into a symbol with a specific amplitude and phase. Quadrature amplitude modulation symbols can transmit more data per unit time and improve spectrum utilization.

[0082] Step S112, inserting a constant amplitude zero autocorrelation sequence into the orthogonal amplitude modulation symbol to obtain a frame structure sequence;

[0083] It should be noted that a frame structure sequence is a sequence formed by combining orthogonal amplitude modulation symbols and a constant amplitude zero autocorrelation sequence in a specific format. Specifically, the constant amplitude zero autocorrelation sequence can be placed at the beginning of the sequence (i.e., the frame header) to help the receiver identify the starting position of the frame and assist in channel estimation. The orthogonal amplitude modulation symbols are placed in the subsequent part of the beginning (i.e., the frame body) to carry the actual information to be transmitted, resulting in a frame structure sequence. This frame structure construction enables the receiver to quickly and accurately parse the signal, improving communication reliability.

[0084] Step S113, performing interpolation processing on the frame structure sequence to obtain an up-sampled sequence;

[0085] It should be noted that the upsampling sequence is a sequence obtained after interpolation, or upsampling. Compared to the frame structure sequence, it has more sampling points and a higher sampling rate. During interpolation, new sampling points are inserted between the sampling points of the frame structure sequence, increasing the signal sampling rate and obtaining an upsampling sequence to better meet subsequent filtering and modulation requirements. The higher sampling rate can more precisely describe the signal waveform, reduce the distortion generated during frequency domain shaping and electro-optical conversion, and lay the foundation for generating high-quality analog waveforms. The sampling frequency of the upsampling sequence should match the requirements of subsequent processing equipment (such as arbitrary waveform generators) or transmission channels.

[0086] Step S114: performing frequency domain shaping on the up-sampled sequence to obtain a pre-processed signal.

[0087] It should be noted that frequency shaping is the process of adjusting the frequency characteristics of a signal through a filter. Specifically, a root raised cosine filter (RRCFilter) can be used to perform frequency shaping on the upsampled sequence, adjusting the frequency components of the signal and confining the signal spectrum within the transmission bandwidth. This produces a preprocessed signal, reducing out-of-band leakage and suppressing inter-symbol interference.

[0088] Step S12: obtaining a transmitting end optical signal based on the laser optical signal, the spontaneous emission optical signal and the pre-processed signal.

[0089] It should be noted that the laser optical signal is a continuous optical carrier signal generated by a laser (Laser), with stable wavelength, frequency and power, and is the carrier of information in optical communications. The characteristics of the laser optical signal (such as wavelength) determine its transmission performance in the optical fiber. For example, optical signals of different wavelengths have different attenuation and dispersion characteristics in the optical fiber. The spontaneous emission optical signal is a wide-spectrum output generated by the amplified spontaneous emission (ASE) light source. It is a random noise optical signal generated by the optical amplifier during the amplification process. It has a wide spectrum range and low power, similar to the background noise in communication transmission. Introducing the spontaneous emission optical signal at the transmitting end can simulate the noise interference existing in the actual transmission environment, making the generated optical signal closer to the actual transmission scenario, thereby ensuring that the processing algorithm at the receiving end can adapt to the noisy environment.

[0090] It should be understood that based on the preprocessing signal, the laser optical signal can be dual-polarization modulated to obtain a modulated optical signal, and the spontaneous emission optical signal and the modulated optical signal can be combined to obtain a combined optical signal. The combined optical signals of multiple transmitting ends are coupled to obtain a transmitting end optical signal.

[0091] In a feasible implementation, step S12 may include steps S121 to S124:

[0092] Step S121, converting the preprocessed signal into an analog waveform;

[0093] It should be noted that an analog waveform is a continuous-time electrical signal obtained by converting a digital preprocessed signal through digital-to-analog conversion. Unlike the discrete nature of digital signals, the amplitude of an analog waveform varies continuously over time, accurately reflecting the information contained in the preprocessed signal. High-quality analog waveforms can reduce signal distortion during the modulation process. The analog waveform can be obtained by performing digital-to-analog conversion on the digital preprocessed signal using a 128GSa / s arbitrary waveform generator (AWG), converting the discrete digital signal into a continuously varying electrical waveform.

[0094] Step S122, performing dual-polarization modulation on the laser optical signal based on the simulated waveform to obtain a modulated optical signal;

[0095] It should be noted that the modulated optical signal is an optical signal obtained through dual-polarization modulation. Its amplitude, phase, or frequency varies with the input analog waveform, thereby loading the information in the analog waveform onto the optical carrier. The modulated optical signal simultaneously carries information in both polarization states and has a high information density. It is the core signal in optical transmission, carrying all the data that the transmitter needs to transmit. The analog waveform is input into the dual-polarization I / Q modulator (Dual Polarization In-phase / Quadrature Modulator, Dual Pol. I / Q Mod.). The modulator simultaneously controls the horizontal and vertical polarization states of the laser optical signal (such as the in-phase component XI and the quadrature component XQ of the X polarization state, and the in-phase component YI and the quadrature component YQ of the Y polarization state), so that the amplitude and phase of the two polarization states vary with the analog waveform, resulting in a modulated optical signal. Through dual-polarization modulation, the two polarization states can carry different information, which is equivalent to achieving parallel transmission on the same optical carrier. This can double the transmission capacity without increasing bandwidth, thereby improving the capacity of optical communication systems.

[0096] Step S123, combining the spontaneous emission light signal and the modulated light signal to obtain a combined light signal;

[0097] It should be noted that the combined optical signal is obtained by mixing the spontaneous emission signal and the modulated optical signal through a coupler. This signal contains both useful modulation information (from the modulated signal) and background noise (from the spontaneous emission signal), and is closer to the characteristics of optical signals in actual fiber transmission. Introducing the combined optical signal allows testing the receiver's signal recovery capability in the presence of noise interference, ensuring system reliability in real-world applications.

[0098] Step S124: couple the combined optical signals of multiple transmitting ends to obtain a transmitting end optical signal.

[0099] It should be noted that fan-in devices (FAN IN) can be used to couple the combined optical signals from multiple transmitters into the same spatial division multiplexing fiber to generate the transmitter optical signal. Spatial division multiplexing fiber has multiple spatial channels (such as the different cores of multi-core fiber or the different modes of few-mode fiber). The coupling operation allows each combined optical signal to enter an independent spatial channel, enabling the parallel transmission of multiple optical signals, significantly increasing the transmission capacity of the entire communication system.

[0100] Step S20, receiving, at the receiving end, the transmitting end optical signal transmitted through the space division multiplexing channel to obtain a receiving end optical signal, and performing receiving end signal processing on the receiving end optical signal to obtain a signal to be equalized;

[0101] It should be understood that a space division multiplexing channel is a channel that uses space division multiplexing technology to achieve signal transmission. The space division multiplexing channel can be one of multi-core fiber (MCF), few-mode fiber (FMF) or multi-core few-mode fiber (MCFMF). By utilizing multiple spatial dimensions of optical fiber (such as different cores or modes) to transmit multiple signals simultaneously, the transmission capacity of the communication system can be significantly improved. During the signal transmission process, due to the physical proximity between different spatial channels, crosstalk will inevitably occur, and the signal will also be affected by channel damage such as attenuation and dispersion.

[0102] Furthermore, the receiving optical signal is the optical signal received by the receiving end after being transmitted through the spatial division multiplexing channel. Due to the effects of crosstalk, attenuation, and dispersion in the spatial division multiplexing channel during transmission, its signal quality is degraded compared to the transmitting optical signal, potentially exhibiting waveform distortion and reduced signal-to-noise ratio. These receiving optical signals carry the information transmitted by the transmitting end, but require processing to restore the original signal.

[0103] It should be noted that multiple optical signals in a spatial division multiplexing channel may be transmitted through different modes or cores in the same optical fiber, resulting in mixed optical signals received at the receiving end. After receiving the transmitting optical signal transmitted through the spatial division multiplexing channel, the receiving end uses an optical demultiplexing device (such as a fan-out device) to separate the multi-channel mixed optical signal into optical signals of each independent channel for separate processing. The separated optical signal is converted into an electrical signal through an optoelectronic conversion device, achieving the form conversion of the optical signal to the electrical signal. The electrical signal is then subjected to receiver signal pre-processing to obtain the signal to be equalized. The receiver signal pre-processing process for the electrical signal includes Gram-Schmidt orthogonalization process (GSOP), matched filtering, and time-frequency synchronization.

[0104] Specifically, during the Gram-Schmidt orthogonalization process, the electrical signal is linearly transformed, converting a set of linearly independent signal vectors into a set of orthogonal signal vectors. For the multiple electrical signal vectors received, one of the signals is first selected as the reference vector, and the other signal vectors are orthogonalized in turn to make them orthogonal to the previous reference vector. The orthogonalization process can be achieved through projection and subtraction operations, that is, for each new signal vector, its projection components on all previous orthogonal vectors are calculated, and these projection components are subtracted from the original vector to obtain a new orthogonal vector. In this way, a set of mutually orthogonal signal vectors are finally obtained.

[0105] Furthermore, after completing the Gram-Schmidt orthogonalization process, the resulting orthogonal signal undergoes matched filtering. During the matched filtering process, a matched filter (MatchFilter) is designed at the receiving end based on the known signal template, so that the filter's impulse response matches the time-reversed version of the signal template to maximize the output signal-to-noise ratio. When the received signal passes through the matched filter, the maximum output amplitude can be achieved at a specific moment (such as the expected moment of signal arrival), improving the signal detection capability. The design of the matched filter needs to be based on the waveform characteristics of the transmitting signal.

[0106] After matched filtering, the signal undergoes time-frequency synchronization to produce the equalized signal. This synchronization involves both time and frequency synchronization, ensuring that the receiver and transmitter are aligned in time and frequency, enabling accurate demodulation and recovery of the signal.

[0107] It should be understood that when pre-processing the received optical signal, the receiver sequentially performs Gram-Schmidt orthogonalization, matched filtering, and time-frequency synchronization to effectively eliminate signal correlation, improve signal detection accuracy, and achieve correct signal demodulation and recovery, thereby providing a high-quality equalized signal for subsequent frequency domain equalization. However, the equalized signal still contains distortion and interference caused by channel impairments, requiring subsequent equalization.

[0108] Step S30, determining a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized;

[0109] It's important to note that the current channel matrix describes the transmission channel characteristics of a spatial division multiplexing channel. Each element in the current channel matrix represents the transmission coefficient from a spatial channel at the transmitter to a spatial channel at the receiver, reflecting the attenuation, phase shift, and crosstalk effects of the signal as it travels between these two channels.

[0110] It should be understood that because the transmitter inserts a constant amplitude zero autocorrelation sequence into the signal, the characteristics of this sequence are known, allowing the receiver to accurately identify the constant amplitude zero autocorrelation sequence portion of the signal to be equalized. The receiver compares and analyzes the constant amplitude zero autocorrelation sequence in the signal to be equalized with the constant amplitude zero autocorrelation sequence known to the transmitter. By calculating the relationship between the two, the transmission coefficients from each transmitting channel to each receiving channel when the signal is transmitted in the spatial division multiplexing channel can be determined, including information such as attenuation, phase change, and crosstalk. Based on these transmission coefficients, a current channel matrix is ​​constructed, describing the transmission characteristics of the current channel.

[0111] Step S40 , performing a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

[0112] It should be noted that the third-order iterative inversion is an iterative algorithm for solving the inverse of a matrix, and has a third-order convergence characteristic, that is, with each iteration, the estimated error of the matrix inverse will decrease at a cubic rate. Compared with the traditional Newton-Schultz iteration, the third-order iterative inversion can achieve higher accuracy in fewer iterations, and is particularly suitable for the inversion scenario of high-dimensional matrices. The current inverse matrix is ​​the inverse matrix of the current channel matrix obtained by the third-order iterative inversion, which can offset the influence of the current channel matrix on the signal, that is, the current channel matrix of the spatial division multiplexing channel is determined based on the constant amplitude zero autocorrelation sequence and the signal to be equalized, and the operation is performed on the signal to be equalized, which can eliminate the signal distortion and crosstalk caused by channel damage, and obtain a frequency domain equalized output signal.

[0113] In addition, linear detection is a method of processing the received signal using linear operations to restore the original transmitted signal. During the linear detection process, the current inverse matrix is ​​multiplied by the signal to be equalized, and the interference and distortion in the signal to be equalized are eliminated through linear transformation to obtain a frequency domain equalized output signal.

[0114] It should be understood that the frequency-domain equalization output signal is the result of linear detection of the current inverse matrix on the signal to be equalized. This eliminates the main effects of crosstalk and dispersion in the spatial division multiplexing channel, significantly improving signal quality and approaching the original signal at the transmitter. After obtaining the frequency-domain equalization output signal, frequency offset estimation (FOE) or carrier phase recovery (CPR) can be performed on the frequency-domain equalization output signal. Key operations such as bit error ratio (BER) analysis can also be performed on the signal after frequency offset estimation or carrier phase recovery.

[0115] This embodiment can achieve accurate channel estimation through a constant amplitude zero autocorrelation sequence, providing a reliable current channel matrix for frequency domain equalization; the third-order iterative inversion can reduce the number of matrix inversion iterations and computational complexity, and can quickly and stably obtain the current inverse matrix even in high-dimensional space division multiplexing scenarios or in the presence of strong mode correlation loss; combined with linear detection, efficient frequency domain equalization can be achieved, effectively eliminating crosstalk and distortion in the space division multiplexing channel, so that the final frequency domain equalization output signal is of high quality and can accurately restore the original information. The overall process improves the transmission efficiency and real-time performance of the space division multiplexing communication system, and can maintain good stability while supporting large-capacity communication needs.

[0116] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 3 , step S40 may include steps S41 to S45:

[0117] Step S41, calculating a first norm and a second norm based on the current channel matrix, wherein the first norm is the first norm of the current channel matrix, and the second norm is the infinite norm of the current channel matrix;

[0118] The first norm is the 1-norm of the current channel matrix, a measure of matrix size. It is calculated by taking the maximum sum of the absolute values ​​of the elements in each column of the current channel matrix. The second norm is the infinity norm of the current channel matrix, a measure of matrix size. It is calculated by taking the maximum sum of the absolute values ​​of the elements in each row of the current channel matrix.

[0119] Step S42: calculating an initial inverse matrix of the current channel matrix based on the first norm and the second norm;

[0120] It should be noted that the initial inverse matrix is ​​the starting value for the iterative matrix inversion process. It is an approximate inverse matrix calculated based on the first norm, infinite norm, and Hermitian transpose of the current channel matrix. The initial inverse matrix does not need to be completely accurate, but it must be reasonable enough to enable subsequent iterations to quickly converge to the true inverse matrix. The Hermitian transpose is a special transposition operation on a complex matrix. This involves first taking the complex conjugate of the matrix elements and then performing the transposition (swapping rows and columns).

[0121] For example, the calculation formula of the initial inverse matrix E0 is as follows:

[0122]

[0123] Where, ‖A‖1 and They represent the 1-norm and infinity-norm of matrix A, respectively. A represents the matrix to be inverted, i.e., the current channel matrix. ()H represents the Hermitian transpose.

[0124] Step S43, performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step size, the identity matrix, and the initial inverse matrix to obtain a current inverse matrix;

[0125] It should be understood that the convergence adjustment step size is a parameter in the third-order iterative inversion formula. Its value range is (0, 1), and it can be set to 0.9 to balance convergence speed and stability. It is used to adjust the convergence rate and numerical stability of the iterative process: when the value of θ is too small, the iteration converges slowly but is more stable; when the value of θ is close to 1, the convergence is fast but may fluctuate under poor channel conditions (such as strong mode correlation loss). By dynamically adjusting θ, the iteration can maintain efficient convergence under different channel conditions. The identity matrix is ​​a special square matrix whose main diagonal elements are all 1 and the remaining elements are all 0. The current inverse matrix is ​​the approximate inverse matrix of the current channel matrix obtained after the third-order iterative inversion.

[0126] It should be noted that third-order iterative inversion is a matrix inversion algorithm with third-order convergence properties. That is, with each iteration, the estimated error of the inverse matrix decreases at a cubic rate. A modified Newton-Schultz iteration (M-NSI) can be used to perform a third-order iterative inversion of the current channel matrix based on the convergence adjustment step size, the identity matrix, and the initial inverse matrix to obtain the current inverse matrix.

[0127] In a feasible embodiment, step S43 may include: calculating the product of the convergence adjustment step and the unit matrix to obtain a first matrix; calculating the product of the initial inverse matrix and the current channel matrix to obtain a second matrix; determining a first-order inverse matrix based on the first matrix and the second matrix; determining a second-order inverse matrix based on the first matrix, the second matrix and the first-order inverse matrix; and calculating the current inverse matrix based on the current channel matrix and the second-order inverse matrix.

[0128] It should be noted that the first matrix, second matrix, first-order inverse matrix, and second-order inverse matrix are intermediate matrices used in the third-order iterative inversion of the current channel matrix. The first matrix is ​​the product of the convergence adjustment step and the identity matrix, multiplied by 3; the second matrix is ​​the product of the initial inverse matrix and the current channel matrix. The first-order inverse matrix is ​​the result of subtracting the second matrix from the first matrix; the second-order inverse matrix is ​​the result of subtracting the second matrix from the first matrix and multiplying it by the first-order inverse matrix; and the current inverse matrix is ​​the product of the initial inverse matrix and the second-order inverse matrix.

[0129] For example, the modified Newton-Schulz iteration (M-NSI) iterative update formula is as follows:

[0130]

[0131] Where, E k represents the inverse matrix estimate output by the k-1th iteration, that is, the current inverse matrix obtained by the k-1th iteration, A represents the matrix to be inverted, that is, the current channel matrix, I represents the identity matrix, It represents the inverse matrix estimate of the k-th iteration output, that is, the current inverse matrix obtained by the k-th iteration, and θ represents the convergence adjustment step size.

[0132] Step S44, performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a current equalized signal;

[0133] It should be noted that the signal to be equalized is the electrical signal obtained after signal processing at the receiving end. It may be distorted and interfered with by impairments such as crosstalk and dispersion in the spatial division multiplexing channel. Linear detection uses linear operations (such as matrix multiplication) to process the signal to be equalized to restore the original signal. The current inverse matrix can be applied to the zero-forcing (ZF) equalization algorithm. This algorithm performs linear detection on the signal to be equalized, eliminates intersymbol interference, and obtains the current equalized signal.

[0134] For example, the ZF balancing algorithm formula is as follows:

[0135] X ZF =(H H H) -1 H HY

[0136] Where H represents the current channel matrix, ()H represents the Hermitian transpose, Y represents the received signal to be equalized, X ZF is the output signal after equalization, that is, the current equalized signal.

[0137] Step S45 : obtaining a frequency domain equalized output signal based on the number of matrix iterative inversion times and the current equalized signal.

[0138] It should be noted that the number of matrix inversion iterations refers to the total number of third-order iterations performed in step S43, which is determined by the preset accuracy requirements or system real-time requirements. Too few iterations may result in insufficient inverse matrix accuracy and poor equalization; too many iterations increase computational latency and reduce real-time performance. By properly setting the number of iterations (e.g., dynamically adjusting based on channel conditions), a balance between accuracy and efficiency can be achieved.

[0139] It should be understood that each time the current channel matrix is ​​iterated three times based on step S43, the number of matrix iterative inversions will be updated. When the number of matrix iterative inversions does not reach the preset value, it returns to step S43 to continue iterating until the number requirement is met, and then the final current equalized signal is used as the frequency domain equalization output signal.

[0140] In a feasible embodiment, step S45 may include: when the number of matrix iterative inversions is less than a preset number of iterations, updating the current channel matrix based on the constant amplitude zero autocorrelation sequence and the current equalized signal; determining the current inverse matrix to be the updated initial inverse matrix, and returning to the step of performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step, the unit matrix and the initial inverse matrix to obtain the current inverse matrix; when the number of matrix iterative inversions reaches the preset number of iterations, determining the current equalized signal to be a frequency domain equalized output signal.

[0141] It should be understood that when the number of matrix iterative inversions is less than the preset number of iterations, the portion corresponding to the constant amplitude zero autocorrelation sequence inserted by the transmitter (i.e., the equalized CAZAC sequence) is extracted from the current equalized signal. By comparing the CAZAC sequence inserted by the transmitter and the corresponding CAZAC sequence in the current equalized signal, the difference between the two can be calculated. Based on this difference, the transmission coefficients of the original current channel matrix are corrected (such as adjusting the amplitude and phase of the matrix elements), and an updated current channel matrix can be obtained. The current inverse matrix is ​​determined to be the updated initial inverse matrix, and the process returns to step S43 to perform a third-order iterative inversion on the updated current channel matrix based on the updated initial inverse matrix to obtain the current inverse matrix for the next iteration, until the number of matrix iterative inversions reaches the preset number of iterations, and the current equalized signal obtained by the last iteration is determined to be the frequency domain equalization output signal.

[0142] Reference Figure 4 , Figure 4 This is a flowchart of the application of the M-NSI algorithm for the frequency domain equalization method based on matrix inversion provided in the second embodiment of this application. Figure 4 As shown, after the current channel matrix is ​​obtained by channel estimation based on the constant amplitude zero autocorrelation sequence, the inverse matrix is ​​estimated based on the current channel matrix in the improved Newton-Schultz iterative module to obtain the current inverse matrix H -1 , and linear detection is performed based on the current inverse matrix to obtain the current equalized signal. In the Successive Interference Cancellation (SIC) module, when the number of matrix iterative inversions is less than the preset number of iterations, that is, when i<2N, the current channel matrix H is updated based on the current equalized signal, and the updated current channel matrix H is estimated based on the updated initial inverse matrix (that is, the current inverse matrix obtained in the previous round of iterations), and linear detection is performed until the number of matrix iterative inversions reaches the preset number of iterations, and the current equalized signal obtained in the last iteration is determined to be the frequency domain equalized output signal. The improved Newton-Schultz iteration module and the continuous interference cancellation module cooperate with each other, and by iteratively optimizing the inverse matrix and cyclically eliminating interference, signal processing based on constant amplitude zero autocorrelation sequence channel estimation can be achieved, thereby improving the performance of the communication system.

[0143] Reference Figure 5 , Figure 5 Schematic diagram of the NSI and M-NSI algorithm effects of the frequency domain equalization method based on matrix inversion provided in Example 2 of the present application. Figure 5 The error of the Newton-Schultz iteration (NSI) algorithm and the improved Newton-Schultz iteration (M-NSI) algorithm in different dimensions (Dim) is shown as the number of iterations changes. The horizontal axis represents the number of iterations (Iteration), and the vertical axis represents the size of the error (Error). The 2-norm (2-norm) can be used to measure the distance between the result of each iterative inversion and the unit matrix to obtain the error. Figure 5 It can be found that compared with the NSI algorithm, the M-NS algorithm usually has a faster convergence speed and lower final error under the same dimension.

[0144] This embodiment calculates the initial inverse matrix of the current channel matrix based on the first norm and the second norm, performs third-order iterative inversion using the convergence adjustment step, the unit matrix and the initial inverse matrix, optimizes the calculation of the inverse matrix, and makes the inverse matrix closer to the ideal value, thereby improving the accuracy of signal processing. Based on the current inverse matrix, linear detection is performed on the equalized signal, effectively reducing the influence of signal interference and noise, improving signal quality, and combining the number of matrix iterative inversions and the current equalized signal to obtain a frequency domain equalized output signal, which can optimize the frequency domain characteristics of the signal and improve the stability and reliability of the signal.

[0145] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the frequency domain equalization method based on matrix inversion in the present application. More forms of simple transformations based on this technical concept are all within the scope of protection of the present application.

[0146] This application also provides a frequency domain equalization device based on matrix inversion, please refer to Figure 6 , the frequency domain equalization device based on matrix inversion includes:

[0147] A transmitting end signal processing module 10 is configured to perform transmitting end signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitting end optical signal;

[0148] a receiving-end signal processing module 20 configured to receive, at a receiving end, the transmitting-end optical signal transmitted through the space-division multiplexing channel, to obtain a receiving-end optical signal, and perform receiving-end signal processing on the receiving-end optical signal to obtain a signal to be equalized;

[0149] A channel estimation module 30 is configured to determine a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized;

[0150] The frequency domain equalization module 40 is configured to perform a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and perform linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

[0151] In one embodiment, the frequency domain equalization module 40 is further used to calculate a first norm and a second norm based on the current channel matrix, wherein the first norm is the first norm of the current channel matrix and the second norm is the infinite norm of the current channel matrix; calculate the initial inverse matrix of the current channel matrix based on the first norm and the second norm; perform a third-order iterative inversion on the current channel matrix according to the convergence adjustment step, the unit matrix and the initial inverse matrix to obtain the current inverse matrix; perform linear detection on the signal to be equalized based on the current inverse matrix to obtain the current equalized signal; and obtain a frequency domain equalized output signal based on the number of matrix iterative inversions and the current equalized signal.

[0152] In one embodiment, the frequency domain equalization module 40 is further used to calculate the product of the convergence adjustment step and the unit matrix to obtain a first matrix; calculate the product of the initial inverse matrix and the current channel matrix to obtain a second matrix; determine a first-order inverse matrix based on the first matrix and the second matrix; determine a second-order inverse matrix based on the first matrix, the second matrix and the first-order inverse matrix; and calculate the current inverse matrix based on the current channel matrix and the second-order inverse matrix.

[0153] In one embodiment, the frequency domain equalization module 40 is further used to update the current channel matrix based on the constant amplitude zero autocorrelation sequence and the current equalized signal when the number of matrix iterative inversions is less than a preset number of iterations; determine the current inverse matrix as the updated initial inverse matrix, and return to the step of performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step, the unit matrix and the initial inverse matrix to obtain the current inverse matrix; when the number of matrix iterative inversions reaches the preset number of iterations, determine the current equalized signal as the frequency domain equalization output signal.

[0154] In one embodiment, the transmitting end signal processing module 10 is further used to perform digital signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal; and obtain a transmitting end optical signal based on the laser optical signal, the spontaneous radiation optical signal and the preprocessed signal.

[0155] In one embodiment, the transmitting end signal processing module 10 is further used to map a pseudo-random binary sequence into an orthogonal amplitude modulation symbol at the transmitting end; insert a constant amplitude zero autocorrelation sequence into the orthogonal amplitude modulation symbol to obtain a frame structure sequence; perform interpolation processing on the frame structure sequence to obtain an up-sampled sequence; and perform frequency domain shaping on the up-sampled sequence to obtain a preprocessed signal.

[0156] In one embodiment, the transmitting end signal processing module 10 is further used to convert the preprocessed signal into an analog waveform; perform dual-polarization modulation on the laser optical signal based on the analog waveform to obtain a modulated optical signal; combine the spontaneous emission optical signal and the modulated optical signal to obtain a combined optical signal; and couple the combined optical signals of multiple transmitting ends to obtain a transmitting end optical signal.

[0157] The matrix inversion-based frequency domain equalization device provided in this application, which employs the matrix inversion-based frequency domain equalization method of the above-mentioned embodiment, can solve the technical problem of how to accelerate the convergence of iterative inversion in the frequency domain equalization process for high-dimensional SDM communication systems to improve channel equalization efficiency. Compared with the prior art, the beneficial effects of the matrix inversion-based frequency domain equalization device provided in this application are the same as the beneficial effects of the matrix inversion-based frequency domain equalization method provided in the above-mentioned embodiment, and the other technical features of the matrix inversion-based frequency domain equalization device are the same as those disclosed in the above-mentioned embodiment method, and are not further described here.

[0158] The present application provides a frequency domain equalization device based on matrix inversion, and the frequency domain equalization device based on matrix inversion includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the frequency domain equalization method based on matrix inversion in the above-mentioned embodiment 1.

[0159] Reference below Figure 7 , which shows a schematic structural diagram of a frequency domain equalization device based on matrix inversion suitable for implementing the embodiments of the present application. The frequency domain equalization device based on matrix inversion in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 7 The frequency domain equalization device based on matrix inversion shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0160] like Figure 7As shown, the frequency domain equalization device based on matrix inversion may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a ROM (Read Only Memory) 1002 or a program loaded from a storage device 1003 into a RAM (Random Access Memory) 1004. Various programs and data required for the operation of the frequency domain equalization device based on matrix inversion are also stored in the RAM 1004. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input device 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output device 1008 including, for example, an LCD (Liquid Crystal Display), speaker, vibrator, etc.; storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and communication device 1009. Communication device 1009 may allow the matrix inversion-based frequency domain equalization device to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a matrix inversion-based frequency domain equalization device with various systems, it should be understood that implementation or presence of all the illustrated systems is not required. More or fewer systems may alternatively be implemented or present.

[0161] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0162] The matrix inversion-based frequency domain equalization device provided in this application, which employs the matrix inversion-based frequency domain equalization method of the above-described embodiment, can solve the technical problem of how to accelerate the convergence of iterative inversion in the frequency domain equalization process for high-dimensional SDM communication systems, thereby improving channel equalization efficiency. Compared with the prior art, the beneficial effects of the matrix inversion-based frequency domain equalization device provided in this application are the same as those of the matrix inversion-based frequency domain equalization method provided in the above-described embodiment. Other technical features of the matrix inversion-based frequency domain equalization device are the same as those disclosed in the above-described embodiment and are not further described here.

[0163] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0164] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0165] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, a computer program) stored thereon, wherein the computer-readable program instructions are used to execute the frequency domain equalization method based on matrix inversion in the above embodiment.

[0166] The computer-readable storage medium provided in this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory) or flash memory, optical fiber, CD-ROM (CD-Read Only Memory, portable compact disk read-only memory), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0167] The computer-readable storage medium may be included in a frequency domain equalization device based on matrix inversion; or may exist independently without being assembled into the frequency domain equalization device based on matrix inversion.

[0168] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the frequency domain equalization device based on matrix inversion, the frequency domain equalization device based on matrix inversion: performs transmitter signal processing on the pseudo-random binary sequence based on the constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitter optical signal; receives the transmitter optical signal transmitted through the space division multiplexing channel at the receiving end to obtain a receiver optical signal, and performs receiver signal processing on the receiver optical signal to obtain a signal to be equalized; determines the current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized; performs third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performs linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalization output signal

[0169] The computer program code for performing the operations of the present application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a LAN (Local Area Network) or a WAN (Wide Area Network), or can be connected to an external computer (e.g., using an Internet service provider to connect via the Internet).

[0170] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0171] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0172] The computer-readable storage medium provided in this application stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned matrix inversion-based frequency domain equalization method. This computer-readable storage medium addresses the technical problem of accelerating the convergence of iterative inversion during frequency domain equalization for high-dimensional SDM communication systems, thereby improving channel equalization efficiency. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the matrix inversion-based frequency domain equalization method provided in the aforementioned embodiment, and are not further elaborated here.

[0173] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned frequency domain equalization method based on matrix inversion when executed by a processor.

[0174] The computer program product provided in this application can address the technical problem of accelerating the convergence of iterative inversion during frequency domain equalization for high-dimensional SDM communication systems, thereby improving channel equalization efficiency. Compared to the prior art, the beneficial effects of the computer program product provided in this application are similar to those of the matrix inversion-based frequency domain equalization method provided in the aforementioned embodiments, and are not further elaborated here.

[0175] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A frequency domain equalization method based on matrix inversion, characterized in that: The frequency domain equalization method based on matrix inversion is applied to a space division multiplexing communication system, wherein the space division multiplexing communication system includes a transmitting end, a space division multiplexing channel, and a receiving end, wherein the transmitting end is connected to the space division multiplexing channel, and the space division multiplexing channel is connected to the receiving end; The frequency domain equalization method based on matrix inversion includes: Performing transmitter signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitter to obtain a transmitter optical signal; receiving, at the receiving end, the transmitting end optical signal transmitted through the space division multiplexing channel to obtain a receiving end optical signal, and performing receiving end signal processing on the receiving end optical signal to obtain a signal to be equalized; Determining a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized; Performing a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

2. The method according to claim 1, wherein The step of performing a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal includes: Calculating a first norm and a second norm based on the current channel matrix, wherein the first norm is a first norm of the current channel matrix, and the second norm is an infinite norm of the current channel matrix; Calculating an initial inverse matrix of the current channel matrix based on the first norm and the second norm; Performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step, the identity matrix, and the initial inverse matrix to obtain a current inverse matrix; Performing linear detection on the signal to be equalized based on the current inverse matrix to obtain a current equalized signal; A frequency domain equalization output signal is obtained based on the number of matrix iterative inversion times and the current equalization signal.

3. The method according to claim 2, wherein The step of performing a third-order iterative inversion on the current channel matrix according to the convergence adjustment step size, the identity matrix, and the initial inverse matrix to obtain a current inverse matrix includes: Calculate the product of the convergence adjustment step and the unit matrix to obtain the first matrix; Calculating the product of the initial inverse matrix and the current channel matrix to obtain a second matrix; determining a first-order inversion matrix based on the first matrix and the second matrix; Determine a second-order inverse matrix based on the first matrix, the second matrix, and the first-order inverse matrix; A current inverse matrix is ​​calculated based on the current channel matrix and the second-order inverse matrix.

4. The method according to claim 2, wherein The step of obtaining a frequency domain equalized output signal based on the number of matrix iterative inversion times and the current equalized signal comprises: When the number of matrix iterative inversions is less than a preset number of iterations, updating the current channel matrix based on the constant amplitude zero autocorrelation sequence and the current equalized signal; Determining the current inverse matrix as the updated initial inverse matrix, and returning to the step of performing a third-order iterative inversion of the current channel matrix according to the convergence adjustment step, the identity matrix, and the initial inverse matrix to obtain the current inverse matrix; When the number of iterative inversions of the matrix reaches the preset number of iterations, it is determined that the current equalized signal is a frequency domain equalized output signal.

5. The method according to claim 1, wherein The step of performing transmitting end signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitting end optical signal comprises: Performing digital signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal; A transmitting end optical signal is obtained based on the laser optical signal, the spontaneous emission optical signal and the pre-processed signal.

6. The method according to claim 5, wherein The step of performing transmitting-end digital signal processing on the pseudo-random binary sequence based on the constant amplitude zero autocorrelation sequence at the transmitting end to obtain a preprocessed signal includes: Mapping the pseudo-random binary sequence into quadrature amplitude modulation symbols at the transmitting end; Inserting a constant amplitude zero autocorrelation sequence into the orthogonal amplitude modulation symbol to obtain a frame structure sequence; Performing interpolation processing on the frame structure sequence to obtain an up-sampled sequence; Perform frequency domain shaping on the up-sampled sequence to obtain a preprocessed signal.

7. The method according to claim 5, wherein The step of obtaining a transmitting end optical signal based on the laser optical signal, the spontaneous emission optical signal and the preprocessing signal comprises: converting the preprocessed signal into an analog waveform; performing dual-polarization modulation on the laser optical signal based on the analog waveform to obtain a modulated optical signal; Combining the spontaneous emission light signal and the modulated light signal to obtain a combined light signal; The combined optical signals of multiple transmitting ends are coupled to obtain a transmitting end optical signal.

8. A frequency domain equalization device based on matrix inversion, characterized in that: The device comprises: a transmitting end signal processing module, configured to perform transmitting end signal processing on a pseudo-random binary sequence based on a constant amplitude zero autocorrelation sequence at the transmitting end to obtain a transmitting end optical signal; a receiving-end signal processing module, configured to receive, at a receiving end, the transmitting-end optical signal transmitted through the space-division multiplexing channel to obtain a receiving-end optical signal, and perform receiving-end signal processing on the receiving-end optical signal to obtain a signal to be equalized; A channel estimation module, configured to determine a current channel matrix of the space division multiplexing channel based on the constant amplitude zero autocorrelation sequence and the signal to be equalized; The frequency domain equalization module is used to perform a third-order iterative inversion on the current channel matrix to obtain a current inverse matrix, and perform linear detection on the signal to be equalized based on the current inverse matrix to obtain a frequency domain equalized output signal.

9. A frequency domain equalization device based on matrix inversion, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the frequency domain equalization method based on matrix inversion according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the frequency domain equalization method based on matrix inversion according to any one of claims 1 to 7 are implemented.

Images