A phase noise insensitive high-order cumulant modulation identification method

Abstract

This invention relates to the field of optical communication technology, specifically to a phase noise-insensitive high-order cumulant modulation identification method. The key technical points include the following steps: S1, signal preprocessing; S2, calculating the phase noise-insensitive high-order cumulant of the signal, and constructing two feature parameters based on the phase noise-insensitive high-order cumulant; S3, constructing a modulation format classifier using a decision tree method; S4, inputting the feature values ​​from step S2 into the modulation format classifier constructed in step S3 to determine the modulation format of the signal under test. By addressing the issue that traditional high-order cumulant algorithms are affected by phase noise caused by laser linewidth in optical fiber communication systems, this invention improves the high-order cumulant algorithm, calculates the phase noise-insensitive high-order cumulant value of the signal, constructs two feature parameters based on the phase noise-insensitive high-order cumulant, and finally uses the feature parameters as classification features, inputting them into the decision tree modulation format classifier for classification.

Description

Technical Field

[0001] This invention relates to the field of optical communication technology, and more specifically to a high-order cumulative modulation identification method that is insensitive to phase noise. Background Technology

[0002] With the rapid development of emerging technologies such as massive data analysis, distributed computing platforms, and intelligent algorithms, the demand for high data transmission rates and high capacity is constantly growing. High-speed optical fiber communication has become a key solution to improve transmission efficiency and meet the explosive growth of data traffic. However, modern transmission systems are becoming increasingly complex, posing challenges to signal demodulation and adaptive processing. Modulation format recognition can distinguish different modulation formats without prior information, which is crucial for signal reception and demodulation in high-speed optical communication systems.

[0003] Based on this, this invention proposes a phase noise-insensitive high-order cumulant modulation identification method. This method achieves accurate identification of modulation formats in high-speed optical fiber communication. First, addressing the issue that traditional high-order cumulants are affected by phase noise caused by the laser linewidth in coherent optical communication systems, the high-order cumulant algorithm is improved by proposing a phase noise-insensitive high-order cumulant algorithm. The algorithm calculates the value of the phase noise-insensitive high-order cumulant of the signal and constructs two feature parameters F1 and F2 based on the phase noise-insensitive high-order cumulant. Finally, the feature parameters are used as classification features and input into the modulation format classifier of the decision tree for classification, thus realizing the identification of the modulation format of signals in high-speed optical fiber communication systems. The actual value and theoretical value of the signal feature values ​​obtained by this method have a small error, achieving accurate identification of signals in high-speed optical fiber communication systems. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a high-order cumulative modulation identification method that is insensitive to phase noise, thus solving the problems mentioned in the background technology.

[0005] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0006] A method for identifying high-order cumulant modulation that is insensitive to phase noise includes the following steps:

[0007] S1. Signal preprocessing: Acquiring signals in four modulation formats from a high-speed fiber optic communication system;

[0008] S2. To address the issue that traditional high-order cumulants are affected by phase noise caused by the laser linewidth in fiber optic communication systems, an improved high-order cumulant algorithm is proposed. This algorithm calculates the phase noise-insensitive high-order cumulant of the signal and constructs two characteristic parameters F1 and F2 based on this algorithm. The newly constructed characteristic parameters are then used to classify modulation formats.

[0009] S3. Construct a modulation format classifier by combining decision tree methods;

[0010] S4. Input the feature values ​​from step S2 into the modulation format classifier constructed in step S3 to determine the modulation format of the signal to be tested.

[0011] Preferably, the signal higher-order cumulant algorithm in step S2 is as follows:

[0012] Assuming s = {X(t)} is a zero-mean, independent and identically distributed complex random process, then the p-th order mixture moment is:

[0013] (1)

[0014] Where E[] represents the expectation, * represents taking the conjugate, and q represents the number of conjugate sequences. hour, It can be represented as:

[0015] (2)

[0016] Laser phase noise can be considered as a Wiener process, that is:

[0017] (3)

[0018] It is an independently distributed Gaussian random process with a mean of 0 and a variance of:

[0019] (4)

[0020] Indicates the laser linewidth. For the signal period, when hour, It can be represented as:

[0021] (5)

[0022] From equation (2), it can be seen that when At that time, phase noise component Increase, making The value of is affected by phase noise, which increases as the laser linewidth increases, leading to... As the error between the theoretical value and the actual value increases, in order to reduce the error, when When the expression for the algorithm of this invention is:

[0023] (6)

[0024] Even with increased laser phase noise, the error between the simulated and theoretical values ​​of this algorithm remains small.

[0025] From equation (5), it can be seen that when When the signal is multiplied by its conjugate, the phase noise component can be observed. The phase noise has been eliminated; therefore, multiplying the signal by its complex conjugate reduces the effect of phase noise. Thus, although phase noise increases with increasing laser linewidth, it is less effective against phase noise. If the value has little effect, then when At that time, the expression for the algorithm in this paper is:

[0026] (7)

[0027] Then, the values ​​of phase noise-insensitive higher-order cumulants can be calculated using the mixed moments. The calculation method for phase noise-insensitive higher-order cumulants is as follows:

[0028] Second-order cumulant:

[0029] (8)

[0030] (9)

[0031] Fourth-order cumulative quantity:

[0032] (10)

[0033] (11)

[0034] Sixth-order cumulative quantity:

[0035] (12)

[0036] This allows us to obtain the higher-order cumulative quantities of the four signals;

[0037] This invention constructs characteristic parameters using higher-order cumulants that are insensitive to signal phase noise, primarily based on two criteria: first, to eliminate the influence of the power factor E, the characteristic parameters must be expressed in ratio form; second, the absolute value of the higher-order cumulants that are insensitive to signal phase noise is taken to eliminate the influence of phase jitter. Therefore, the higher-order cumulants that are insensitive to phase noise for each type of signal are combined to form two types of characteristic parameters, namely:

[0038] (13)

[0039] (14)

[0040] These two feature parameters completed the feature classification of the four signals.

[0041] Preferably, step S3, based on two features F1 and F2 and a selected threshold, where the threshold is the midpoint of the theoretical values ​​of the two feature parameters at each classification node, utilizes the algorithm design of a decision tree classifier to accurately identify four signals in a high-speed optical fiber communication system.

[0042] Preferably, the simulation system in step S1 acquires constellation diagrams of four optical signals with OSNR of 10dB~30dB and a step size of 2.5dB, and acquires a set of 100 constellation diagrams for each modulation format.

[0043] In summary, the present invention has the following main beneficial effects:

[0044] This invention achieves accurate signal identification in high-speed optical fiber communication systems. First, addressing the issue that traditional high-order cumulant algorithms are affected by phase noise caused by laser linewidth in optical fiber communication systems, this invention improves upon the high-order cumulant algorithm by proposing a phase noise-insensitive high-order cumulant algorithm. The algorithm calculates the value of the phase noise-insensitive high-order cumulant of the signal and constructs two feature parameters based on the phase noise-insensitive high-order cumulant. Finally, the feature parameters are used as classification features and input into the modulation format classifier of the decision tree for classification. Attached Figure Description

[0045] Figure 1 A schematic flowchart of a phase noise-insensitive high-order cumulant modulation identification method provided in an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of a coherent optical communication system provided in an embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram of a phase noise-insensitive high-order cumulant algorithm provided in an embodiment of the present invention.

[0048] Figure 4 This is a schematic diagram of the decision tree modulation format classifier provided in an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] The following embodiments are used to illustrate the present invention, but should not be used to limit the scope of protection of the present invention. The conditions in the embodiments can be further adjusted according to specific conditions, and simple improvements to the method of the present invention under the premise of the concept of the present invention are all within the scope of protection claimed by the present invention.

[0051] refer to Figure 1 As shown, the present invention provides a high-order cumulant modulation identification method that is insensitive to phase noise, comprising:

[0052] I. Pretreatment

[0053] Schematic diagram of coherent optical communication system (reference) Figure 2 As shown, the simulation system collected constellation diagrams for four modulation formats with an OSNR range of 10dB to 30dB and a step size of 2.5dB. For each modulation format, a set of 100 constellation diagrams was collected.

[0054] II. Feature Extraction Algorithm for Hybrid Signals Based on Higher-Order Cumulatives

[0055] High-order cumulant algorithms that are insensitive to phase noise, such as Figure 3 As shown, assuming s = {X(t)} is a zero-mean, independent and identically distributed complex random process, then the p-th order mixture moment is:

[0056] (1)

[0057] Where E[] represents the expectation, * represents taking the conjugate, and q represents the number of conjugate sequences. hour, It can be represented as:

[0058] (2)

[0059] Laser phase noise can be considered as a Wiener process, that is:

[0060] (3)

[0061] It is an independently distributed Gaussian random process with a mean of 0 and a variance of:

[0062] (4)

[0063] Indicates the laser linewidth. For the signal period, when hour, It can be represented as:

[0064] (5)

[0065] From equation (2), it can be seen that when At that time, phase noise component Increase, making The value of is affected by phase noise, which increases as the laser linewidth increases, leading to... As the error between the theoretical value and the actual value increases, in order to reduce the error, when When the expression for the algorithm of this invention is:

[0066] (6)

[0067] Even with increased laser phase noise, the error between the simulated and theoretical values ​​of this algorithm remains small.

[0068] From equation (5), it can be seen that when When the signal is multiplied by its conjugate, the phase noise component can be observed. The phase noise has been eliminated; therefore, multiplying the signal by its complex conjugate reduces the effect of phase noise. Thus, although phase noise increases with increasing laser linewidth, it is less effective against phase noise. If the value has little effect, then when At that time, the expression for the algorithm in this paper is:

[0069] (7)

[0070] Then, the values ​​of phase noise-insensitive higher-order cumulants can be calculated using the mixed moments. The calculation method for phase noise-insensitive higher-order cumulants is as follows:

[0071] Second-order cumulant:

[0072] (8)

[0073] (9)

[0074] Fourth-order cumulative quantity:

[0075] (10)

[0076] (11)

[0077] Sixth-order cumulative quantity:

[0078] (12)

[0079] This allows us to obtain the higher-order cumulative quantities of the four signals;

[0080] This invention constructs characteristic parameters using higher-order cumulants that are insensitive to signal phase noise, primarily based on two criteria: first, to eliminate the influence of the power factor E, the characteristic parameters must be expressed in ratio form; second, the absolute value of the higher-order cumulants that are insensitive to signal phase noise is taken to eliminate the influence of phase jitter. Therefore, the higher-order cumulants that are insensitive to phase noise for each type of signal are combined to form two types of characteristic parameters, namely:

[0081] (13)

[0082] (14)

[0083] These two feature parameters completed the feature classification of the four signals.

[0084] III. Constructing a modulation format classifier using decision tree methods for classification and recognition.

[0085] like Figure 4 As shown, the eigenvalues ​​F1 and F2 of the higher-order cumulants, which are insensitive to phase noise, are first simulated. Since the F1 values ​​of three modulation formats are relatively close, they cannot completely distinguish the four modulation formats. Therefore, F1 can first distinguish the 32QAM signal in the entire OSNR range, and the eigenvalue F2 can then distinguish the remaining three modulation formats, namely QPSK, 16QAM and 64QAM. Then, the threshold is set according to the theoretical value of the eigenvalues ​​to identify the signal. By using the algorithm design of the decision tree classifier, the accurate identification of the four modulation formats is completed.

[0086] This invention establishes a coherent optical communication simulation system with an optical fiber length of 100km and an amplifier gain of 16dB, with a 40GBaud output. The optical signal-to-noise ratio ranges from 10 to 30dB with a step size of 2.5dB. The MFI scheme determines four different modulation formats in the high-speed optical fiber communication system: QPSK, 16QAM, 32QAM, and 64QAM. The results show that the proposed method has low algorithm complexity, high recognition rate, and can overcome the limitations of traditional high-order cumulant algorithms affected by phase noise. It also considers the actual situation of the signal passing through the optical fiber channel and achieves accurate identification of the modulation format in the high-speed optical fiber communication system.

[0087] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that, unless otherwise defined, the technical or scientific terms used in this invention should be understood in the ordinary sense by those skilled in the art to which this invention pertains, and the terms "comprising" or "including" or similar terms used in this invention mean that the element or object preceding the word covers the element or object listed after the word and its equivalents.

[0088] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-order cumulant modulation identification method insensitive to phase noise, characterized in that, Includes the following steps: S1. Signal preprocessing: Acquiring signals in four modulation formats from a high-speed fiber optic communication system; S2. To address the issue that traditional high-order cumulants are affected by phase noise caused by the laser linewidth in fiber optic communication systems, an improved high-order cumulant algorithm is proposed. This algorithm calculates the phase noise-insensitive high-order cumulant of the signal and constructs two characteristic parameters F1 and F2 based on this algorithm. The newly constructed characteristic parameters are then used to classify modulation formats. S3. Construct a modulation format classifier by combining decision tree methods; S4. Input the feature values ​​from step S2 into the modulation format classifier constructed in step S3 to determine the modulation format of the signal to be tested.

2. The phase noise-insensitive high-order cumulant modulation identification method according to claim 1, characterized in that, The higher-order cumulant algorithm for the signal in step S2 is as follows: Assuming s = {X(t)} is a zero-mean, independent and identically distributed complex random process, then the p-th order mixture moment is: (1) Where E[] represents the expectation, * represents taking the conjugate, and q represents the number of conjugate sequences. hour, It can be represented as: (2) Laser phase noise can be considered as a Wiener process, that is: (3) It is an independently distributed Gaussian random process with a mean of 0 and a variance of: (4) Indicates the laser linewidth. For the signal period, when hour, It can be represented as: (5) From equation (2), it can be seen that when At that time, phase noise component Increase, making The value of is affected by phase noise, which increases as the laser linewidth increases, leading to... As the error between the theoretical value and the actual value increases, in order to reduce the error, when When the expression for the algorithm of this invention is: (6) Even with increased laser phase noise, the error between the simulated and theoretical values ​​of this algorithm remains small. From equation (5), it can be seen that when When the signal is multiplied by its conjugate, the phase noise component can be observed. The phase noise has been eliminated; therefore, multiplying the signal by its complex conjugate reduces the effect of phase noise. Thus, although phase noise increases with increasing laser linewidth, it is less effective against phase noise. If the value has little effect, then when At that time, the expression for the algorithm in this paper is: (7) Then, the values ​​of phase noise-insensitive higher-order cumulants can be calculated using the mixed moments. The calculation method for phase noise-insensitive higher-order cumulants is as follows: Second-order cumulant: (8) (9) Fourth-order cumulative quantity: (10) (11) Sixth-order cumulative quantity: (12) This allows us to obtain the higher-order cumulative quantities of the four signals; This invention constructs characteristic parameters using higher-order cumulants that are insensitive to signal phase noise, primarily based on two criteria: first, to eliminate the influence of the power factor E, the characteristic parameters must be expressed in ratio form; second, the absolute value of the higher-order cumulants that are insensitive to signal phase noise is taken to eliminate the influence of phase jitter. Therefore, the higher-order cumulants that are insensitive to phase noise for each type of signal are combined to form two types of characteristic parameters, namely: (13) (14) These two feature parameters completed the feature classification of the four signals.

3. According to claim 1, the high-order cumulative modulation identification method that is insensitive to phase noise, step S3, based on two features F1 and F2 and a selected threshold, wherein the threshold is selected as the midpoint of the theoretical values ​​of the two feature parameters at each classification node, utilizes the algorithm design of a decision tree classifier to accurately identify four signals in a high-speed optical fiber communication system.

4. The phase noise-insensitive high-order cumulant modulation identification method according to claim 1, characterized in that, The simulation system described in step S1 acquired constellation diagrams for four optical signals with OSNR ranging from 10dB to 30dB and a step size of 2.5dB. For each modulation format, a set of 100 constellation diagrams was acquired.

Images