Signal generation method and device and nonlinear system noise measurement method and system

Abstract

The invention discloses a signal generation method and device and a nonlinear system noise measurement method and system, and the method comprises the steps: carrying out the diffusion processing of a standard PAM (Pulse Amplitude Modulation) reference signal, and obtaining a standard PAM diffusion reference signal; according to the standard PAM diffusion reference signal, carrying out primary notch processing and probability reconstruction processing on the one-dimensional random input signal; and re-quantizing the signal after the trapped wave processing and the probability reconstruction processing, and outputting the generated PAM trapped wave approximate signal, by adopting the technical scheme of the invention, the signal conforming to the specific probability distribution and the specific spectrum distribution can be stably generated, and the complexity is reduced at the same time.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a signal generation method and device, and a nonlinear system noise measurement method and system. Background Art

[0002] With the rapid development of communication technology, the problem of "capacity crunch" has become increasingly prominent and needs to be solved urgently. In order to meet this challenge in optical fiber communication systems, in addition to adopting emerging space division multiplexing technology to significantly improve communication capacity, another key research direction is to overcome the nonlinear problem in traditional optical fiber communication systems. In recent years, this field has received extensive attention and research. The reason why the study of nonlinear problems is difficult is that it is not only related to the characteristics of the system itself, but also closely related to the properties of the input signal. The actual optical signal input is different from the traditional signal used to test nonlinearity, which makes it complicated to obtain accurate nonlinear estimation. The classic nonlinear monitoring method, noise power ratio (NPR), can obtain the correct in-band response of a specific input or a specific system by simply creating a narrowband notch in the power spectrum, which has great limitations.

[0003] At present, existing methods can generate signals that meet the requirements of specific probability distribution and specific spectrum distribution in a complex way. The above methods are very complex and cannot generate target signals stably. At the same time, the above methods detect nonlinearity through direct response of signals. For different target sequences, adaptive modulation is required in the process to generate detection signals that meet the detection range requirements. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a signal generation method and device, a nonlinear system noise measurement method and system, which can stably generate signals that conform to a specific probability distribution and a specific spectrum distribution while reducing complexity.

[0005] To achieve the above object, the present invention adopts the following technical solution:

[0006] A signal generation method, comprising:

[0007] Performing diffusion processing on a standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal;

[0008] According to the standard PAM diffusion reference signal, a notch processing and probability reconstruction processing are performed on the one-dimensional random input signal;

[0009] The signal after notch processing and probability reconstruction processing is re-quantized, and the generated PAM notch isoprobable signal is output.

[0010] Preferably, the diffusion processing performed on the standard PAM reference signal is adding Gaussian noise to the standard PAM reference signal.

[0011] Preferably, the notch processing is to set a specific frequency band of the signal to 0, and the probability reconstruction is to replace the notched signal one by one in order of size according to the distribution of the reference signal.

[0012] Preferably, the requantization is to restore the signal to which Gaussian noise is added to a standard level distribution through decision making.

[0013] The present invention also provides a signal generating device, comprising:

[0014] The first processing module is used to perform diffusion processing on the standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal;

[0015] The second processing module is used to perform a notch processing and a probability reconstruction processing on the one-dimensional random input signal according to the standard PAM diffusion reference signal;

[0016] The second processing module is used to re-quantize the signal after the notch processing and the probability reconstruction processing, and output the generated PAM notch isoprobability signal.

[0017] Preferably, the diffusion processing performed on the standard PAM reference signal is adding Gaussian noise to the standard PAM reference signal.

[0018] Preferably, the notch processing is to set a specific frequency band of the signal to 0, and the probability reconstruction is to replace the notched signal one by one in order of size according to the distribution of the reference signal.

[0019] Preferably, the requantization is to restore the signal to which Gaussian noise is added to a standard level distribution through decision making.

[0020] The present invention also provides a method for measuring noise in a nonlinear system, comprising:

[0021] Generate PAM notch and other isoprobable signals;

[0022] Measuring the response of a nonlinear system using the PAM notch isoprobable signal;

[0023] Based on the original response of the test signal passing through the system, the nonlinear response at the notch frequency point is obtained without BER decoding.

[0024] The present invention also provides a nonlinear system noise measurement system, comprising:

[0025] A signal generating device, used for generating a PAM notch isoprobable signal;

[0026] A test device for measuring the response of a nonlinear system using the PAM notch isoprobable signal;

[0027] The post-processing device is used to obtain the nonlinear response at the notch frequency point based on the original response of the test signal passing through the system without obtaining the BER of the communication system.

[0028] The present invention has the following beneficial effects:

[0029] According to the PAM diffusion signal based on the standard PAM reference signal, a notch processing and a probability reconstruction processing are performed on the one-dimensional arbitrary input signal, so that the one-dimensional input signal has the same probability distribution as the reference signal. At the same time, except for the notch position in the frequency domain, the spectrum distribution is consistent with the standard signal. In addition, this method can be applied to various high-order amplitude modulation scenarios.

[0030] In addition, the PAM notch isoprobable signal obtained by the present invention can obtain the same accuracy as the signal obtained by the existing method through subsequent signal reconstruction. The nonlinear noise of the nonlinear system can be measured more accurately. At the same time, the signal generation is simplified and the stability of the process is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0032] Figure 1 This is a flow chart of a signal generation method according to an embodiment of the present invention;

[0033] Figure 2 Schematic diagram of the signal generation process of Comparative Example 1;

[0034] FIG3( a ) is a comparison of the probability distribution of the signal generated by the present invention and the signal of equal spectrum and equal probability in comparative example 2;

[0035] FIG3( b ) is a comparison of the power spectral density performance of the signal generated by the present invention and the signal of equal spectrum and equal probability in comparative example 2;

[0036] FIG4( a ) is a memoryless nonlinear channel response result of a signal generated by the present invention;

[0037] FIG4( b ) is a memoryless nonlinear channel response result of generating equal-spectrum and equal-probability signals in Comparative Example 2;

[0038] Figure 5 Schematic diagram of the process of the nonlinear monitoring post-processing method in Embodiment 2 of the present invention;

[0039] FIG6( a ) is a comparison of the response implemented by the present invention after reconstruction processing and the result of orthogonal decomposition of the traditional nonlinear monitoring method;

[0040] FIG6( b ) is a comparison of the response of Example 2 after reconstruction and the orthogonal decomposition result of the traditional nonlinear monitoring method. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] Embodiment 1:

[0044] like Figure 1 As shown, the present invention provides a signal generating method, comprising:

[0045] Step 1: Perform diffusion processing on a standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal;

[0046] Step 2: Perform a notch process and a probability reconstruction process on the one-dimensional random input signal according to the standard PAM diffusion reference signal;

[0047] Step 3: re-quantize the signal after the notch processing and probability reconstruction processing, and output the generated PAM notch iso-probability signal.

[0048] Further, in step 1, the diffusion processing performed on the standard PAM reference signal is to add Gaussian noise to the standard PAM reference signal.

[0049] Further, in step 2, the random one-dimensional input signal can be the same length as the target reference signal, with arbitrarily distributed noise such as Gaussian noise, and uniformly distributed random numbers; notch processing refers to setting a specific frequency band of the signal to 0, and the specific operation is: performing Fourier transform (FFT) on the random input signal to obtain the real and imaginary information of the original input frequency domain, directly setting the real and imaginary parts of the desired frequency band to 0 according to monitoring requirements, and performing inverse Fourier transform on the result with 0 point in the frequency domain to obtain a signal that has completed notch processing; probability reconstruction refers to replacing the signal after notching one by one in order of size according to the distribution of the reference signal; the specific operation is: sorting the diffuse reference signal and the signal after notching respectively, and replacing them in a one-to-one correspondence, assuming that the signal length is 10000; the maximum value of the reference signal replaces the maximum value after notching, and the reference signal ranked 2000 replaces the reference signal ranked 2000 after notching, regardless of the data size, only the size relationship.

[0050] Furthermore, requantization is to restore the signal to which Gaussian noise is added to a standard level distribution through decision making.

[0051] In the implementation of the present invention, diffusion processing is a method of adding noise to a standard PAM signal, the purpose of which is to record the order of each symbol without changing the signal distribution. This step lays the foundation for the subsequent probability reconstruction operation. Notch processing refers to setting the value of the input one-dimensional signal at a specific frequency point to zero, while keeping the values ​​at other positions unchanged. Through this processing, the power within the notch bandwidth is significantly reduced.

[0052] The main difference from the existing scheme is that after the implementation of the present invention, the quantization output can be directly performed without additional determination steps after the completion of the probability reconstruction processing, and the needs of subsequent nonlinear monitoring of the present invention can be met. The key to the probability reconstruction processing is that it replaces the amplitude of the signal after the notch according to the amplitude sequence of the reference signal after diffusion, ensuring that the replaced positions have the same size. Since noise is introduced into the reference signal, re-quantization is required before outputting the PAM test signal with standard level distribution.

[0053] Comparative Example 1 provides a method for generating a two-dimensional isospectral isoprobable signal (Reference 1: Li Jingnan. Signal generation device and method: 202210502752.2[P].2022-05-10.), such as Figure 2The schematic diagram of the embodiment closest to the process of the present invention is shown, and the steps are as follows: Step 1: Perform diffusion processing on the two-dimensional reference signal to obtain a two-dimensional diffusion signal; Step 2: Perform isoprobability processing on the two-dimensional input signal according to the two-dimensional diffusion signal to obtain a two-dimensional isoprobability signal; Step 3: Perform perturbation processing on the two-dimensional isoprobability signal to obtain a two-dimensional perturbation signal; Perform isospectral processing on the two-dimensional perturbation signal to obtain a two-dimensional isospectral signal; Return to the second step, perform isoprobability processing on the two-dimensional isospectral signal again to obtain a two-dimensional isoprobability signal; Determine whether the two-dimensional isoprobability signal meets the preset conditions, and when the judgment result is "yes", output the two-dimensional isoprobability signal; When the judgment result is "no", continue to execute the third step; Finally, output the two-dimensional isoprobability signal, because the isospectral processing has been performed before, the two-dimensional isoprobability signal is a two-dimensional isospectral isoprobability signal. However, the invention still retains isospectral and perturbation processing, and it is necessary to perform judgment iteration, which is not stable generation.

[0054] Comparative Example 2 is a notch-likelihood signal generated by the prior art (Reference 2: Ye, T., Su, X., Zhang, K. et al. Nonlinear noise spectrum measurement using a probability-maintained noise power ratio method. Commun Eng 1, 49 (2022).), which is the application of Comparative Example 1 in a one-dimensional signal. In order to demonstrate the innovation of the present invention in one-dimensional signal processing, the PAM8 signal is selected as an example. For the convenience of distinction, the product of the comparative example is a probability-maintained notch signal (PMNS), and the product of the present invention is a simplified probability-maintained notch signal (PMNS). As shown in Figure 3, we compared and analyzed the technology implemented by the present invention with the prior art from two dimensions: power spectral density and probability distribution. The results show that there are significant differences between the two in terms of power spectral density, but they are almost completely consistent in probability distribution.

[0055] In order to further distinguish the difference between the signals generated by the implementation of the present invention and the comparative example 2, a memoryless nonlinear channel is used to simulate the two. The implementation of the present invention mainly targets the third-order nonlinearity in the optical communication system, so the channel is set as:

[0056] y=Ax+Bx 3

[0057] Wherein, A is the linear term coefficient, B is the third-order nonlinear term coefficient, x is the specific test signal generated by the implementation of the present invention and comparative example 2, and y is the simulated receiving signal.

[0058] FIG4 is a graph showing the response of the signal generated by comparative example 2 and the signal generated by the embodiment of the present invention to the same preset memoryless nonlinear channel; the nonlinear coefficient B is kept at -0.015.

[0059] FIG4(a) shows the direct reception response of the signal implemented in the present invention, while FIG4(b) shows the direct response of the signal generated in Comparative Example 2. Through comparative analysis, we can clearly observe the significant difference between the two responses. This difference shows that the signal implemented in the present invention is not equivalent to the prior art. In addition, the signal implemented in the present invention fails to correctly reflect the nonlinear characteristics of the system in the direct response.

[0060] Embodiment 2:

[0061] The present invention also provides a signal generating device, comprising:

[0062] The first processing module is used to perform diffusion processing on the standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal;

[0063] The second processing module is used to perform a notch processing and a probability reconstruction processing on the one-dimensional random input signal according to the standard PAM diffusion reference signal;

[0064] The second processing module is used to re-quantize the signal after the notch processing and the probability reconstruction processing, and output the generated PAM notch isoprobability signal.

[0065] As an implementation manner of the embodiment of the present invention, performing diffusion processing on the standard PAM reference signal is adding Gaussian noise to the standard PAM reference signal.

[0066] As an implementation method of the embodiment of the present invention, the notch processing is to set a specific frequency band of the signal to 0, and the probability reconstruction is to replace the notched signal one by one in the order of size according to the distribution of the reference signal.

[0067] As an implementation manner of the embodiment of the present invention, re-quantization is to return the signal to which Gaussian noise is added to a standard level distribution through decision.

[0068] Embodiment 3:

[0069] The present invention also provides a method for measuring noise in a nonlinear system, comprising:

[0070] Generate PAM notch and other isoprobable signals;

[0071] Measuring the response of a nonlinear system using the PAM notch isoprobable signal;

[0072] Based on the original response of the test signal passing through the system, the nonlinear response at the notch frequency can be obtained without obtaining the BER of the communication system.

[0073] In the embodiments of the present invention, the notch signal generated by the present invention is used as the test signal. The prior art method can directly detect and obtain nonlinear monitoring results through a spectrometer or an electric spectrometer, but the signal used in the implementation of the present invention needs to be post-processed, so the receiving unit is limited to a receiver; the prior art method is received by the receiver; generally, equalization, decoding and other processing are performed to obtain the nonlinear response of the system. The signal generated by the implementation of the present invention can directly obtain the correct nonlinear response after being received by the receiver without the need for traditional decoding processing. The post-processing operation of the present invention to obtain the nonlinear response after the signal is received is as follows: Figure 5 As shown, it includes three parts: synchronization, power matching and linear stripping processing. The whole is the post-processing unit of this method. The specific operations are: synchronization: the received test signal is cross-correlatedly aligned with the transmitted signal; power matching: the signal will experience a variety of linear and nonlinear effects such as loss, gain, dispersion, Kerr effect, etc. during the transmission process. In order to utilize the characteristics of the signal implemented by the present invention and weaken the linear part of the received signal as much as possible, the implementation of the present invention uses a repeatable and effective process: adding a coefficient β to the transmitted signal so that its total power is consistent with the received signal; taking the memoryless nonlinear channel simulation as an example, when the power is matched, the following formula is satisfied:

[0074]

[0075] The purpose of this step is to scale the transmitted signal and pave the way for stripping the linear part as much as possible;

[0076] Linear stripping: The received signal is subtracted from the transmitted signal that is synchronized and scaled by the coefficient β, and the linear part of the received signal is removed; the expression of the signal to be processed after linear stripping is:

[0077] y n =(A-β)x+Bx 3

[0078] At this time, the ratio of the linear part to the nonlinear part is:

[0079]

[0080] At this time, the ratio tends to 0, and as the nonlinear effect increases, β increases. The smaller the ratio of linearity to nonlinearity is, the more accurate the monitoring result is, and the distortion power spectrum within the notch frequency band can be obtained.

[0081] Although there are significant differences between the responses of the implementation signal of the present invention and the comparative signal, the research results show that the nonlinear response trends of the two are consistent. To further verify this, this matching stripping method is also applicable to the existing technical solution, which improves the detection range of the existing solution signal. As shown in Figure 6 (a) and Figure 6 (b), the experimental results show that the implementation of the present invention can achieve an accuracy similar to that of the existing solution, and the error is controlled within 0.5dB.

[0082] Embodiment 4:

[0083] The present invention also provides a nonlinear system noise measurement system, comprising:

[0084] A signal generating device, used for generating a PAM notch isoprobable signal;

[0085] A test device for measuring the response of a nonlinear system using the PAM notch isoprobable signal;

[0086] The post-processing device is used to obtain the nonlinear response at the notch frequency point based on the original response of the test signal passing through the system without using BER decoding.

[0087] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A signal generation method, characterized in that: include: Performing diffusion processing on a standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal; According to the standard PAM diffusion reference signal, a notch processing and probability reconstruction processing are performed on the one-dimensional random input signal; The signal after notch processing and probability reconstruction processing is re-quantized, and the generated PAM notch isoprobable signal is output.

2. The signal generating method according to claim 1, characterized in that: The diffusion processing performed on the standard PAM reference signal is to add Gaussian noise to the standard PAM reference signal.

3. The signal generating method according to claim 2, characterized in that: The notch processing is to set a specific frequency band of the signal to 0, and the probability reconstruction is to replace the notched signal one by one in the order of size according to the distribution of the reference signal.

4. The signal generating method according to claim 3, characterized in that: Requantization is to restore the signal with Gaussian noise added to the standard level distribution through decision making.

5. A signal generating device, characterized in that: include: The first processing module is used to perform diffusion processing on the standard pulse amplitude modulation (PAM) reference signal to obtain a standard PAM diffusion reference signal; The second processing module is used to perform a notch processing and a probability reconstruction processing on the one-dimensional random input signal according to the standard PAM diffusion reference signal; The second processing module is used to re-quantize the signal after the notch processing and the probability reconstruction processing, and output the generated PAM notch isoprobability signal.

6. The signal generating device according to claim 5, characterized in that: The diffusion processing performed on the standard PAM reference signal is to add Gaussian noise to the standard PAM reference signal.

7. The signal generating device according to claim 6, characterized in that: The notch processing is to set a specific frequency band of the signal to 0, and the probability reconstruction is to replace the notched signal one by one in the order of size according to the distribution of the reference signal.

8. The signal generating device according to claim 7, characterized in that: Requantization is to restore the signal with Gaussian noise added to the standard level distribution through decision making.

9. A method for measuring noise in a nonlinear system, characterized in that: include: Generate PAM notch and other isoprobable signals; Use PAM notch isoprobable signals to measure the response of nonlinear systems; Based on the original response of the test signal passing through the system, the nonlinear response at the notch frequency can be obtained without obtaining the BER of the communication system.

10. A nonlinear system noise measurement system, characterized in that: include: A signal generating device, used for generating a PAM notch isoprobable signal; A test device for measuring the response of a nonlinear system using a PAM notch isoprobable signal; The post-processing device is used to obtain the nonlinear response at the notch frequency point based on the original response of the test signal passing through the system without using BER decoding.

Images