A method and apparatus for processing a signal, a storage medium, and an electronic device

Abstract

An embodiment of the present invention provides a signal processing method, apparatus, storage medium, and electronic device. The method includes performing a non-linear effect compensation process on a digital signal after converting an optical analog signal into a digital signal. Therefore, the problem that non-linear impairments in an optical communication process cannot be well processed in the related art can be solved, and the technical effect of improving the accuracy of signal processing can be achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of communications, and more particularly, to a method and apparatus for processing signals, a storage medium, and an electronic device. Background Art

[0002] Optical fiber communication systems are important infrastructures for carrying information transmission tasks. Compared with wireless communication, optical fiber communication has the advantages of large bandwidth and low loss, and can support optical transmission networks of dozens or even hundreds or thousands of kilometers. Whether it is the optical interconnection or optical access network between data centers between 2 km and 20 km, or the metropolitan area network or backbone network of hundreds or thousands of kilometers, optical fiber analog signals will be damaged by various noises in the entire transmission network, resulting in a decrease in the signal-to-noise ratio. Among them, there are two most influential factors. One is the inter-symbol interference (ISI) introduced by various dispersions in the optical fiber and the limited bandwidth of optoelectronic devices, and the other is the influence of numerous non-linear effects in the transmission network. Therefore, in order to correctly send the received signal to the end user, it is necessary to cooperate with digital signal processing to compensate and recover the signal on the receiving side.

[0003] In related technologies, digital signal processing mainly compensates for the inter-symbol interference introduced by dispersion in the optical fiber link and the limited bandwidth of devices, but cannot well handle the signal damage caused by non-linear effects in the link.

[0004] In view of the problem that non-linear damage in the process of optical communication cannot be well handled in related technologies, this patent proposes a new type and there is no existing solution yet. Summary of the Invention

[0005] Embodiments of the present invention provide a method and apparatus for processing signals, a storage medium, and an electronic device, so as to at least solve the problem that non-linear damage in the process of optical communication cannot be well handled in related technologies.

[0006] According to an embodiment of the present invention, there is provided a method for processing signals, including: after receiving an optical analog signal, converting the optical analog signal into a digital signal; performing non-linear effect compensation processing on the digital signal.

[0007] According to another embodiment of the present invention, there is provided a signal processing apparatus, including:

[0008] a conversion module, configured to convert the optical analog signal into a digital signal after receiving the optical analog signal;

[0009] a processing module, configured to perform non-linear effect compensation processing on the digital signal.

[0010] According to another embodiment of the present invention, there is also provided a computer-readable storage medium, in which a computer program is stored, and the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0011] According to another embodiment of the present invention, there is also provided an electronic device, including a memory and a processor, a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0012] Through the present invention, since the digital signal is subjected to non-linear effect compensation processing after the optical analog signal is converted into a digital signal, the problem that the non-linear damage in the optical communication process in the related art cannot be well processed can be solved, and the technical effect of improving the signal processing accuracy can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a hardware structure block diagram of an optical communication receiving end of a signal processing method according to an embodiment of the present invention;

[0014] Figure 2 It is a flowchart of a signal processing method according to an embodiment of the present invention;

[0015] Figure 3 It is a structure block diagram of a signal processing device according to an embodiment of the present invention;

[0016] Figure 4 It is a schematic diagram of the process of digital signal processing at the receiving end according to an exemplary embodiment of the present invention;

[0017] Figure 5 It is a schematic diagram of the architecture of an equalizer according to an exemplary embodiment of the present invention;

[0018] Figure 6 It is a schematic diagram of the architecture of a feed-forward equalizer according to an exemplary embodiment of the present invention;

[0019] Figure 7 It is a schematic diagram of the comparison of the equalization effects between the equalization architecture according to an exemplary embodiment of the present invention and the equalizer architecture in the related art;

[0020] Figure 8 It is a schematic diagram of the architecture of a feedback equalizer according to an exemplary embodiment of the present invention;

[0021] Figure 9 It is a schematic diagram of the architecture of a feed-forward plus feedback equalizer according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] Embodiments of the present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence.

[0024] The method embodiments provided in the embodiments of the present application can be executed in an optical communication receiving end or a similar receiving device. Taking the operation on the optical communication receiving end as an example, Figure 1 is a hardware structure block diagram of an optical communication receiving end for a signal processing method according to an embodiment of the present invention. As Figure 1 shown, the optical communication receiving end may include one or more ( Figure 1 only one is shown in the figure) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Among them, the above-mentioned optical communication receiving end may further include a transmission device 106 for communication functions and an input / output device 108. Those of ordinary skill in the art can understand that Figure 1 the structure shown is only schematic and does not limit the structure of the above-mentioned optical communication receiving end. For example, the optical communication receiving end may further include more or fewer components than Figure 1 shown in the figure, or have a different configuration from Figure 1 shown in the figure.

[0025] The memory 104 can be used to store computer programs. For example, software programs and modules of application software, such as the computer program corresponding to the signal processing method in the embodiments of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above-mentioned method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor 102, and these remote memories can be connected to the optical communication receiving end through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and their combinations.

[0026] The transmission device 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by a communication provider at the optical communication receiving end. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, abbreviated as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0027] It should be noted that the optical communication receiving end may further include an analog-to-digital conversion module for converting an optical analog signal into a digital signal, and then using the processor 102 to execute the computer program stored in the memory 104 to further process the digital signal.

[0028] In this embodiment, a method for processing signals running on the above-mentioned optical communication receiving end is provided. Figure 2 It is a flowchart of the signal processing method according to the embodiment of the present invention, as Figure 2 shown, the process includes the following steps:

[0029] Step S202, after receiving the optical analog signal, convert the optical analog signal into a digital signal;

[0030] Step S204, perform non-linear effect compensation processing on the digital signal.

[0031] Through the above steps, since the non-linear effect compensation processing is performed on the digital signal after converting the optical analog signal into a digital signal, the problem that the non-linear damage in the optical communication process in the related art cannot be well processed can be solved, and the technical effect of improving the signal processing accuracy can be achieved.

[0032] In an exemplary embodiment, after converting the optical analog signal into the digital signal, perform linear filtering processing on the digital signal;

[0033] The non-linear effect compensation processing on the digital signal includes: performing the non-linear effect compensation processing on the digital signal after the linear filtering processing.

[0034] It should be noted that the linear filtering processing can compensate for the inter-symbol interference introduced by the dispersion in the optical fiber link and the limited bandwidth of the device to better correct the signal. After the linear filtering processing, performing the non-linear effect compensation processing on the digital signal can further compensate for the non-linear damage in the optical communication process and better correct the signal.

[0035] In an exemplary embodiment, the method further includes: after performing the non-linear effect compensation processing on the digital signal, determining a processing error according to the digital signal after the non-linear effect compensation processing, where the processing error is used to indicate the error between the processed digital signal and a preset processing target; and adjusting a tap coefficient according to the processing error, where the tap coefficient is used to perform the linear filtering processing on the digital signal.

[0036] It should be noted that adjusting the tap coefficient by using the error between the processed digital signal and the processing target can better optimize the processing effect of the linear filtering processing to continuously approach the processing target. Among them, the processing target can be the target of some preset processing parameters, such as thresholds, preset ranges, etc.

[0037] In an exemplary embodiment, performing the non-linear effect compensation processing on the digital signal includes: using an activation function to perform the non-linear effect compensation processing on the digital signal.

[0038] In an exemplary embodiment, the activation function satisfies the following conditions: the response curve of the activation function is a non-linear function; and, the activation function is a monotonically increasing or monotonically decreasing function; and, the activation function is smoothly differentiable within the data range of the digital signal.

[0039] In an exemplary embodiment, the activation function further satisfies at least one of the following conditions: the output value of the activation function is between 0 and 1 or between -1 and 1; the derivative of the activation function is available. It should be noted that the output value of the activation function being between 0 and 1 or between -1 and 1 can make the activation function consistent with the logical value after the symbol demapping of the communication system, which is helpful for the processing of digital signals.

[0040] In an exemplary embodiment, the activation function includes one of the following: Sigmoid function, Tanh function, or Leaky ReLU function.

[0041] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation manner. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

[0042] In this embodiment, a signal processing device is further provided. This device is used to implement the above-mentioned embodiment and preferred implementation manners, and those that have been described will not be elaborated again. As used hereinafter, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0043] Figure 3 is a structural block diagram of a signal processing device according to an embodiment of the present invention. As Figure 3 shown, the device includes:

[0044] A conversion module 31, configured to convert the optical analog signal into a digital signal after receiving the optical analog signal;

[0045] A processing module 33, configured to perform non-linear effect compensation processing on the digital signal.

[0046] Through the above steps, since the non-linear effect compensation processing is performed on the digital signal after converting the optical analog signal into a digital signal, the problem that the non-linear damage in the optical communication process in the related art cannot be well processed can be solved, and the technical effect of improving the signal processing accuracy can be achieved.

[0047] In an exemplary implementation manner, the device further includes: a linear filtering module, configured to perform linear filtering processing on the digital signal after converting the optical analog signal into the digital signal; the processing module is further configured to perform the non-linear effect compensation processing on the digital signal after the linear filtering processing.

[0048] In an exemplary implementation manner, the device further includes: a determination module, configured to determine a processing error according to the digital signal after the non-linear effect compensation processing after performing the non-linear effect compensation processing on the digital signal, where the processing error is used to indicate the error between the processed digital signal and a preset processing target; the linear filtering module is further configured to adjust the tap coefficient according to the processing error, where the tap coefficient is used to perform the linear filtering processing on the digital signal.

[0049] In an exemplary implementation manner, the processing module is further configured to perform the non-linear effect compensation processing on the digital signal by using an activation function.

[0050] In an exemplary implementation manner, the activation function satisfies the following conditions: the response curve of the activation function is a non-linear function; and, the activation function is a monotonically increasing or monotonically decreasing function; and, the activation function is smoothly differentiable within the data range of the digital signal.

[0051] In an exemplary embodiment, the activation function further satisfies at least one of the following conditions: the output value of the activation function is between 0 and 1 or between -1 and 1; the derivative of the activation function is obtainable. In an exemplary embodiment, the activation function includes one of the following: Sigmoid function, Tanh function, or Leaky ReLU function.

[0052] It should be noted that the above-mentioned respective modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited thereto: all the above-mentioned modules are located in the same processor; or, the above-mentioned respective modules are located in different processors in any combined form.

[0053] Exemplary embodiment

[0054] In this embodiment, a non-linear activation function is added to the linear equalizer architecture, so that the linear equalizer architecture has a certain non-linear response, thereby compensating for non-linear effects. In view of the situation in the related art where non-linear factors cannot be well compensated, this embodiment proposes a new type of equalization compensation architecture, enabling the equalizer to not only eliminate the influence of inter-symbol interference but also perform a certain degree of non-linear compensation.

[0055] This embodiment mainly aims at compensating for the non-linear effects introduced by optical fibers and devices in the optical fiber transmission link in the digital domain for the received signal sent into the equalizer, and this compensation can be realized through computer software algorithms.

[0056] Figure 4 is a schematic diagram of the process of receiving-end digital signal processing according to the exemplary embodiment of the present invention, as Figure 4 shown, and includes the following steps:

[0057] The analog signal in the received optical fiber is first converted into a digital signal by an analog-to-digital conversion module and then sent into a digital signal processing chip for data recovery and compensation. Generally, the digital signal processing chip includes the following functions: resampling: making the sampling rate become an integer multiple sampling rate required for signal recovery; clock / phase recovery: enabling the receiving-end system to recover the clock signal of the data and find the optimal sampling point and phase point of the data; frame synchronization: synchronously finding the frame header; equalization processing: compensating for inter-symbol interference introduced by dispersion and other factors in network transmission; the equalization processing further includes using an activation function to perform non-linear effect compensation processing on the signal.

[0058] In this embodiment, a non-linear activation function module is added on the basis of the equalization compensation architecture, aiming to enable the new equalizer to have the ability to compensate for non-linear signals. The non-linear compensation module described in the present invention can be a non-linear activation function.

[0059] In an exemplary embodiment, for the equalization technology module, structures such as Feed Forward Equalization (FFE), Decision Feedback Equalization (DFE), or FFE+DFE can be adopted on the equalizer architecture.

[0060] In an exemplary embodiment, the non-linear activation function should at least satisfy the following three conditions simultaneously:

[0061] The response curve of the activation function module needs to be a non-linear function; the activation function should satisfy the property of being monotonically increasing / monotonically decreasing, and its output value should be between 0 and 1 or between -1 and 1, consistent with the logical value after symbol demapping in the communication system; the activation function should satisfy the property of being smooth and differentiable within the data range.

[0062] In an exemplary embodiment, since the function derivative will be reflected in the system algorithm, the function derivative should have the property of being easily obtained.

[0063] In an exemplary embodiment, the equalizer provided in this embodiment, in terms of the selection of the linear filter, is applicable to various architectures including a Feed Forward Equalizer (FFE), a Decision Feedback Equalizer (DFE), and a Feed Forward plus Decision Feedback Equalizer (FFE+DFE).

[0064] In an exemplary embodiment, the equalizer provided in this embodiment, in terms of the selection of the equalizer training method, can include training sequence-based and blind equalization methods, as well as other training methods such as the Constant Modulus Algorithm, etc.

[0065] In an exemplary embodiment, the equalizer provided in this embodiment, in terms of the selection of the convergence algorithm, can include common algorithms such as the Least Mean Square (LMS) algorithm and the Recursive Least Square (RLS) algorithm.

[0066] The digital processing solution for compensating the link non-linearity algorithm provided in this embodiment can compensate for certain non-linear distortions in the link. Based on the traditional common equalizer architecture, only a non-linear activation function is added, which is easy to perform technology iteration and upgrade in the traditional technology reserve, and the change in the hardware system architecture is small, facilitating implementation.

[0067] Figure 5 is a schematic diagram of the architecture of the equalizer according to an exemplary embodiment of the present invention, as Figure 5As shown, the equalizer architecture can be equivalent to a linear filter for compensating inter-symbol interference in the signal, but it cannot compensate for the non-linearity in the signal. A non-linear activation function can be added to the equalizer architecture to compensate for the non-linear part of the signal. The signal processing flow includes the following steps:

[0068] Step 1: The digital signal X(n) fed into the new equalizer first passes through a linear filter module. The linear filter architecture includes FFE, DFE, and FFE+DFE, etc. This step can adopt the linear filter processing flow or method, which is not limited in the embodiments of the present invention.

[0069] Step 2: The digital signal preprocessed by the linear filter is then fed into the non-linear activation function module to generate a non-linear response. The purpose of this step is to compensate for the non-linear problems in the system link. The specific processing flow of this step can refer to other exemplary implementation manners.

[0070] It should be emphasized that in the related art, there is no such step of non-linear activation function in signal processing. The entire equalizer architecture is a linear system, which can only compensate for inter-symbol interference in the result and cannot eliminate non-linear factors. The role of the activation function here is to increase the non-linear response of the entire equalizer to the data.

[0071] In an exemplary implementation manner, the selection of the activation function needs to meet the following several restrictive requirements:

[0072] 1. The input-output response curve of the activation function module should be a non-linear function.

[0073] 2. The activation function should satisfy the property of monotonic increase / monotonic decrease, and its output value should be between 0 and 1 or between -1 and 1, which is consistent with the logical value after symbol demapping in the communication system.

[0074] 3. The activation function should satisfy the property of being smooth and differentiable within the data range.

[0075] In an exemplary implementation manner, the derivative of the function will be reflected in the system algorithm, so the derivative of the function should have the property of being easy to obtain.

[0076] In an exemplary implementation manner, several non-linear activation functions are recommended here, such as Sigmoid: y = 1 / (1 + e -x ) function, Tanh: y = (1 - e -x ) / (1 + e -x ) function or Leaky ReLU function, etc.

[0077] Such as Figure 5As shown, in an exemplary embodiment, a part of the output signal Y(n) after non-linear processing is sent back to the linear filter, and according to certain algorithms and error calculations, the filter tap coefficients of the linear filter are modified in reverse. The purpose of this step is to make the frequency response curve of the linear filter better adapt to the performance of the transmission system. The specific processing flow of this step can refer to other exemplary embodiments.

[0078] In addition, there are no mandatory requirements for the training target method and convergence algorithm of the feedforward feedback equalizer, and various training methods can be applied. The above is only an example, and this embodiment is not limited thereto.

[0079] Figure 6 is a schematic diagram of the architecture of a feedforward feedback equalizer according to an exemplary embodiment of the present invention, as Figure 6 shown, the digital bit signal to be equalized is delayed by T / n seconds through a fractional delay element, where T is the symbol period corresponding to one bit of data, and n is the resampling upsampling multiple. The number of delay elements can be optimized according to actual requirements. After the delayed data, each data is multiplied by a multiplier and then added together. w 0 to w n are the tap coefficients multiplied when passing through the multiplier before each data. The tap coefficients are initially set to an initial value, and subsequent values will be adaptively updated based on the training target with certain convergence conditions. The result after the tap coefficient accumulation will pass through a non-linear activation function to output a non-linear value Z k , and at the same time, the output data also needs to be fed back backward to update the forward tap coefficients in reverse based on certain convergence conditions.

[0080] Figure 7 is a schematic diagram comparing the equalization effects of the equalization architecture according to an exemplary embodiment of the present invention and the equalizer architecture in related technologies, showing the comparison results of the transmission performance of the equalizer architecture according to an exemplary embodiment of the present invention and the traditional FFE equalizer architecture algorithm used to simplify the coherent experimental system. In the simplified coherent system, the non-linear sources of the entire link include non-linearity in the transmitting end laser, non-linear detection of the receiving end photodetector, and non-linear effects of the envelope detection method in the backend analog signal processing, etc. The traditional linear equalizer architecture mainly compensates for inter-symbol interference in the signal. When focusing on the bit error condition of 1e-2, its received optical power is -23 dBm. Correspondingly, using the new equalizer architecture of Embodiment 1 of this patent, while compensating for inter-symbol interference, it also performs certain non-linear compensation. When also focusing on the bit error condition of 1e-2, its received optical power is about -25 dBm, with a performance improvement of about 2 dB.

[0081] Figure 8It is a schematic diagram of the architecture of a backward feedback equalizer according to an exemplary embodiment of the present invention. It is an embodiment in which a non-linear activation function is added on the basis of the current backward feedback equalizer (DFE) architecture. As Figure 8 shown, the difference from the forward feedback equalizer is that the input data bit signal first passes through a non-linear activation function. After the non-linear data Zk is output, one path is used as the output, and the other path is sent to each delay element for fractional delay of T / n seconds, where T is the symbol period corresponding to one bit of data, and n is the resampling upsampling multiple. The number of delay elements can be optimized according to actual needs. After the delayed data, each data is multiplied by a multiplier and then added together. w 0 to w n is the tap coefficient that needs to be multiplied before each data. The tap coefficient is initially set to an initial value, and the subsequent values will adaptively update the tap coefficient based on the training objective with certain convergence conditions, including LMS (Least Mean Square), RLS (Recursive Least Square), CMA (Constant Modulus Algorithm), etc.

[0082] Figure 9 It is a schematic diagram of the architecture of a forward plus backward feedback equalizer according to an exemplary embodiment of the present invention. It is an embodiment in which a non-linear activation function is added on the basis of the equalizer architecture that combines the forward equalization architecture (FFE) and the backward equalization architecture (DFE). As Figure 9 shown, the received data bit stream first passes through a forward equalizer architecture and is summed as the input of the subsequent backward equalizer architecture. At the same time, the non-linear data Z output by the backward equalizer through the non-linear activation function k adaptively updates the tap coefficients of the FFE and DFE with a certain convergence algorithm.

[0083] An embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored. Among them, the computer program is set to execute the steps in any one of the above method embodiments when running.

[0084] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as USB flash drives, read-only memories (ROM for short), random access memories (RAM for short), mobile hard disks, magnetic disks, or optical discs that can store computer programs.

[0085] An embodiment of the present invention further provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0086] In an exemplary embodiment, the above electronic device may further include a transmission device and an input / output device. The transmission device is connected to the processor, and the input / output device is connected to the processor.

[0087] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary embodiments, and details are not repeated herein.

[0088] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device, so that they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. Thus, the present invention is not limited to any specific combination of hardware and software.

[0089] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing a signal, characterized in that, applied to an optical communication receiving end, comprising: after receiving an optical analog signal, converting the optical analog signal into a digital signal; performing nonlinear effect compensation processing on the digital signal; wherein, the performing nonlinear effect compensation processing on the digital signal comprises: using an activation function to perform the nonlinear effect compensation processing on the digital signal; wherein, the using an activation function to perform the nonlinear effect compensation processing on the digital signal comprises: inputting the digital signal into a backward equalizer, and performing nonlinear effect compensation on the digital signal through a nonlinear activation function in the backward equalizer.

2. The method according to claim 1, characterized in that, further comprising: after converting the optical analog signal into the digital signal, performing linear filtering processing on the digital signal; the performing nonlinear effect compensation processing on the digital signal comprises: performing the nonlinear effect compensation processing on the digital signal after the linear filtering processing.

3. The method according to claim 2, characterized in that, further comprising: after performing the nonlinear effect compensation processing on the digital signal, determining a processing error according to the digital signal after the nonlinear effect compensation processing, wherein the processing error is used to indicate an error between the processed digital signal and a preset processing target; adjusting a tap coefficient according to the processing error, wherein the tap coefficient is used to perform the linear filtering processing on the digital signal.

4. The method according to claim 1, characterized in that, the activation function satisfies the following conditions: the response curve of the activation function is a nonlinear function; and, the activation function is a monotonically increasing or monotonically decreasing function; and, the activation function is smoothly differentiable within the data range of the digital signal.

5. The method according to claim 1, characterized in that, the activation function further satisfies at least one of the following conditions: the output value of the activation function is between 0 and 1 or between -1 and 1; the derivative of the activation function is obtainable.

6. The method according to claim 1, characterized in that, the activation function includes one of the following: Sigmoid function, Tanh function or Leaky ReLU function.

7. A signal processing device, characterized in that, comprising: a conversion module, configured to convert the optical analog signal into a digital signal after receiving the optical analog signal; a processing module, configured to perform nonlinear effect compensation processing on the digital signal; wherein, the device is further configured to input the digital signal into a backward equalizer, and perform nonlinear effect compensation on the digital signal through a nonlinear activation function in the backward equalizer.

8. A computer-readable storage medium, characterized in that, a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute the method according to any one of claims 1 to 6 when running.

9. An electronic device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is configured to run the computer program to execute the method described in any one of claims 1 to 6.

Images