A chaotic optical communication system time delay characteristic estimation method, system and application

Abstract

This invention belongs to the field of chaotic optical communication technology and discloses a method, system, and application for estimating the time delay characteristics of a chaotic optical communication system. The chaotic signal generated by the chaotic optical communication system is transmitted from the transmitter to the receiver, and the chaotic signal is captured from the transmission link. After data acquisition, the delay characteristics of the chaotic signal in the chaotic system are directly estimated. A reservoir network is used to establish a model between the chaotic time series and the reconstructed delay series. The functional relationship between the two sequences at the correct delay characteristic parameters is analyzed, and the delay characteristics are estimated by traversing the reconstructed delay, thereby realizing the estimation of the delay characteristics of the chaotic optical communication system. This invention adopts a reservoir network modeling method, which has certain advantages and effectiveness. It can directly estimate the delay characteristic parameters from the chaotic signal of the chaotic optical communication system, further improving computational efficiency and providing a new approach for delay extraction in chaotic optical communication systems.

Description

Technical Field

[0001] This invention belongs to the field of chaotic optical communication technology, and particularly relates to a method, system and application for estimating time delay characteristics of chaotic optical communication systems. Background Technology

[0002] Currently, with the continuous integration and development of communication and computer technologies, the world is gradually entering the information network era. In the field of communication, although fiber optic transmission has a shorter development history compared to other modern communication technologies, it has already become one of the important means of information transmission in modern communication networks. With the development of fiber optic communication technology, the security performance of optical communication systems has also attracted much attention. Traditional communication systems are now relatively mature, and their security mainly relies on mature algorithms such as encryption algorithms and authentication protocols to encrypt information at the application layer. However, the pursuit of security in communication systems is endless. To further improve the security of information transmission, chaotic communication systems have gradually attracted the attention of researchers. Chaotic communication has advantages such as strong anti-interference ability, high security, and good compatibility, and is currently one of the physical layer encryption technologies widely used in the field of secure communication. Typically, when the parameters of the transmitting and receiving ends of a chaotic communication system match, chaotic communication can be achieved through chaotic synchronization. In other words, chaotic communication relies on the operating parameters of hardware devices to achieve encryption and decryption at the physical layer. Chaotic optical communication is an emerging information security transmission technology that can be fully compatible with traditional communication systems and enhances the security of application layer encryption algorithms through the advantages of the physical layer.

[0003] Pooled recurrent neural networks (RNNs) are a novel type of neural network based on recurrent neural networks. Their core design consists of a large-scale, randomly generated, sparsely connected dynamic feedback pool. Pooled recurrent neural networks predetermine their internal connection parameters and employ linear regression for computation, exhibiting unique advantages in network stability and computational efficiency. Therefore, they have applications in mechanical control and computer science, and play a crucial role in nonlinear system identification and chaotic time series prediction.

[0004] Currently, time delay characteristic parameters are generally estimated using mathematical statistics methods, such as autocorrelation functions and delay mutual information methods. These methods estimate time delay characteristics by studying the relationship between a continuous chaotic sequence and the delayed sequence. However, the effectiveness of these methods decreases when the nonlinearity of the chaotic system reaches a certain level, and they also require a large amount of data. In addition, machine learning-based time delay characteristic estimation methods can also effectively estimate delay parameters. Currently, convolutional neural networks have successfully estimated the accurate time delay characteristic parameters of intensity chaotic optoelectronic systems, phase chaotic optoelectronic systems, and semiconductor laser chaotic systems with optical feedback. This approach improves recognition accuracy and range, but it requires prior knowledge of some aspects of the chaotic system. Furthermore, due to the use of convolutional networks and the transformation of data into a two-dimensional graph, the algorithm structure and training process are relatively complex (see Opt. Express, Vol. 28, No. 10, P.15221, 2020). A time delay parameter estimation method for optical chaotic communication systems based on a reservoir network can further reduce training costs, requiring only a small amount of time series data from the chaotic system. Therefore, it is essential to apply reservoir neural network technology to delay parameter estimation in optical chaotic communication systems.

[0005] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0006] (1) Existing mathematical statistics methods are not applicable to highly nonlinear scenarios. As the nonlinearity of the chaotic system increases, the effectiveness of the method will decrease significantly, and the amount of data required by the method is relatively large.

[0007] (2) Existing machine learning-based time delay feature estimation methods require prior knowledge of the chaotic system, and the algorithm structure is complex and the computation time is relatively long. Summary of the Invention

[0008] To address the problems existing in the prior art, this invention provides a method, system, and application for estimating the time delay characteristics of a chaotic optical communication system, particularly relating to a method, system, medium, device, and terminal for estimating the time delay characteristics of a chaotic optical communication system based on a reservoir.

[0009] This invention is implemented as follows: a method for estimating the time delay characteristics of a chaotic optical communication system, the method comprising:

[0010] The chaotic signal generated by the chaotic optical communication system is transmitted from the transmitter to the receiver, and the chaotic signal is captured from the transmission link. After acquiring the data, the delay characteristics of the chaotic signals in the chaotic system are directly estimated; the functional relationship between the chaotic signal sequence and the delayed signal is analyzed. Using nonlinear functions With delay parameters Establish a functional relationship and arbitrarily select two subsequences. and . subsequence As input to the reserve pool network, Set the desired output, then train the reservoir network to perform sequence-to-sequence regression. The expression for the reservoir network training result is: From this, we can obtain The approximate model, when hour Approaching ;when At that point, the fixed functional relationship will no longer exist. Subsequently, the functional relationship is obtained multiple times by traversing and reconstructing the relationship after a delay of s. The process of traversal is the search The process involves selecting a sequence for testing and using mean square error to describe the prediction performance of the corresponding delay s. When the mean square error is at its minimum, the correct delay is obtained, thereby realizing the estimation of the delay characteristics of the chaotic optical communication system.

[0011] Furthermore, the time delay characteristic estimation method for the chaotic optical communication system includes the following steps:

[0012] Step 1: Generate chaotic signals using an optical chaotic communication system;

[0013] Step 2: Receive and collect chaotic signals;

[0014] Step 3: Generate a delayed sequence of chaotic signals within a certain delay range (usually tens to hundreds of nanoseconds);

[0015] Step four: Divide the sample set according to the original sequence and the delayed sequence;

[0016] Step 5: Train the reservoir network to perform prediction regression from the original sequence to the delayed sequence;

[0017] Step 6: Use mean squared error to evaluate the model performance and plot the results.

[0018] Step 7: Estimate the time delay characteristic parameters of the optical chaotic communication system.

[0019] Furthermore, in steps one and two, in the optical chaotic communication system, a time-delayed chaotic signal of a certain length (hundreds to thousands of nanoseconds in duration) is collected in the transmission link, the received data is collected, and the chaotic data is recorded as... .

[0020] In step three, the received data obtained in step one is... Decompose the sequence and select a sequence length of... Time-delay chaotic signal Select a data segment of length from the sequence. for a subsequence of As the initial training signal, a segment with the same length as the initial training signal and a delay of [length missing] is selected. Delay sequence As a training target signal.

[0021] Furthermore, in step four, while performing step three, the dataset is formatted, and the training set format is as follows: The test prediction set format is The training data pairs and the test prediction data pairs are respectively positioned on the time axis as follows: Step and Move at the speed of a step, where For the length of the training data, To test the predicted data length.

[0022] Furthermore, in step five, the parameters of the reserve pool network are set, and the training set and test prediction set are divided in step four. After the reserve pool training is completed, a corresponding output weight is obtained. The weights represent the correspondence between the two sequences; with the selected delay... The changes are used to obtain the corresponding relationship of the delayed sequences; after the output weights are determined, the overall network structure of the reservoir network is completely determined, and only the test prediction set needs to be input to obtain the prediction data; another subsequence of the chaotic signal is selected. Input to corresponding output weights Predicted sequences are generated in the reservoir network. .

[0023] Furthermore, in step six, the mean squared error is used as the error function to calculate the predicted sequence obtained in step five. With the actual sequence The differences in traversal time are used to evaluate the predictive performance of the reservoir network. A plot is created with traversal delay on the x-axis and mean squared error on the y-axis. Result image.

[0024] In step seven, when the traversal delay and the correct delay are equal, the prediction result is... It has relatively high accuracy, so it is drawn according to step six. When displaying the results, in The minimum value appears at a certain time, and a downward peak appears in the graph, thus enabling the estimation of the time delay characteristic parameters.

[0025] Another object of the present invention is to provide a time delay feature estimation system for a chaotic optical communication system that applies the aforementioned time delay feature estimation method, the time delay feature estimation system for a chaotic optical communication system comprising:

[0026] The chaotic signal acquisition module is used to generate chaotic signals using a chaotic optical communication system and transmit them from the transmitter to the receiver, capturing the chaotic signals from the transmission link.

[0027] The sample set partitioning module is used to generate a delayed sequence of chaotic signals within a certain delay range, and to partition the sample set according to the original sequence and the delayed sequence;

[0028] The model building module is used to establish a model between the chaotic time series and the reconstructed delayed series using a reservoir network, and to analyze the functional relationship between the two series at the correct time delay characteristic parameters.

[0029] The model performance evaluation module is used to evaluate the model performance and draw the results graph after the original sequence is predicted and regressed to the delayed sequence through the training reservoir network.

[0030] The delay feature estimation module is used to estimate the delay features by traversing and reconstructing the delay, thereby realizing the estimation of the delay features of the chaotic optical communication system.

[0031] Another object of the present invention is to provide a computer device comprising a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the following steps:

[0032] The chaotic signal generated by the chaotic optical communication system is transmitted from the transmitter to the receiver, and the chaotic signal is captured from the transmission link. After the data is collected, the delay characteristics of the chaotic signal in the chaotic system are directly estimated. A model between the chaotic time series and the reconstructed delay series is established using a reservoir network. The functional relationship between the two series at the correct delay characteristic parameters is analyzed, and the delay characteristics are estimated by traversing the reconstructed delay, thereby realizing the estimation of the delay characteristics of the chaotic optical communication system.

[0033] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0034] The chaotic signal generated by the chaotic optical communication system is transmitted from the transmitter to the receiver, and the chaotic signal is captured from the transmission link. After the data is collected, the delay characteristics of the chaotic signal in the chaotic system are directly estimated. A model between the chaotic time series and the reconstructed delay series is established using a reservoir network. The functional relationship between the two series at the correct delay characteristic parameters is analyzed, and the delay characteristics are estimated by traversing the reconstructed delay, thereby realizing the estimation of the delay characteristics of the chaotic optical communication system.

[0035] Another objective of this invention is to provide an information data processing terminal, which is used to implement the time delay characteristic estimation system of the chaotic optical communication system.

[0036] Based on the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solution to be protected by this invention from the following aspects:

[0037] This invention proposes a model-free delay feature estimation method based on a reservoir network. The delay parameters are directly estimated from the time series of an optical chaotic communication system. A reservoir network is used to establish a model between the sequence and its delay sequence to estimate the correlation between the two sequences. The mean square error is used as the evaluation index to explore the relationship between delay and delay parameters, thereby successfully estimating the delay parameters of the chaotic communication system.

[0038] The time delay characteristic parameter estimation method for optical chaotic communication systems in this invention directly extracts the correlation function relationship from the time series of the optical chaotic communication system, utilizes a reservoir network to perform predictive regression between the original sequence and the delayed sequence, and evaluates the prediction effect using the mean square error. This method remains effective even in extreme environments and has advantages in terms of training algorithm and training efficiency.

[0039] Compared with traditional mathematical and statistical estimation methods, the method of this invention has certain advantages in highly nonlinear, high-noise environments and experimental environments, and requires a shorter length of the original chaotic sequence.

[0040] Compared with estimation methods based on convolutional networks, the method of this invention provides a time delay estimation method in optical chaotic communication systems that does not require prior knowledge, and further simplifies the algorithm structure and improves training efficiency.

[0041] The time delay feature estimation method for chaotic optical communication systems provided by this invention adopts the method of reservoir network modeling, which can directly estimate the time delay feature parameters from the chaotic signals of chaotic optical communication systems, improve computational efficiency, and provide a new approach for time delay extraction in chaotic optical communication systems. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart of the time delay feature estimation method for a chaotic optical communication system provided in an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the time delay feature estimation method for a chaotic optical communication system provided in an embodiment of the present invention;

[0045] Figure 3 This is a structural diagram of the time delay feature estimation system for the chaotic optical communication system provided in an embodiment of the present invention;

[0046] In the diagram: 1. Chaotic signal acquisition module; 2. Sample set partitioning module; 3. Model building module; 4. Model performance evaluation module; 5. Delay feature estimation module.

[0047] Figure 4 This is a structural diagram of a chaotic communication system provided in an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of the reservoir network training process provided in an embodiment of the present invention;

[0049] Figure 6 This is a schematic diagram of the delay feature estimation process provided in an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of the estimation results of the delay feature estimation provided in the embodiment of the present invention;

[0051] In the figure: (a) estimation results using the autocorrelation function method; (b) estimation results using the delayed mutual information method; (c) estimation results using the method of this invention.

[0052] Figure 8 This is a schematic diagram of the estimation results of the conventional method under the same data length as in Embodiment 2, provided by an embodiment of the present invention;

[0053] In the figure: (a) Estimation results using the autocorrelation function method; (b) Estimation results using the delayed mutual information method. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0055] To address the problems existing in the prior art, this invention provides a method, system, and application for estimating time delay characteristics in a chaotic optical communication system. The invention will be described in detail below with reference to the accompanying drawings.

[0056] To enable those skilled in the art to fully understand how the present invention is specifically implemented, this section provides an explanatory description of the embodiments that expand upon the technical solutions of the claims.

[0057] like Figure 1 As shown, the time delay feature estimation method for a chaotic optical communication system provided in this embodiment of the invention includes the following steps:

[0058] S101, uses an optical chaotic communication system to generate chaotic signals;

[0059] S102, receives and collects chaotic signals;

[0060] S103 generates a delayed sequence of chaotic signals within a certain delay range;

[0061] S104, the sample set is divided according to the original sequence and the delayed sequence;

[0062] S105, the prediction regression from the original sequence to the delayed sequence is completed by training the reservoir network;

[0063] S106, use mean squared error to evaluate the model performance and plot the results;

[0064] S107 estimates the time delay characteristic parameters of the optical chaotic communication system.

[0065] The principle diagram of the time delay feature estimation method for the chaotic optical communication system provided in this embodiment of the invention is shown below. Figure 2 .

[0066] like Figure 3 As shown, the time delay feature estimation system for a chaotic optical communication system provided in this embodiment of the invention includes:

[0067] Chaotic signal acquisition module 1 is used to generate chaotic signals using a chaotic optical communication system and transmit them from the transmitter to the receiver, capturing the chaotic signals from the transmission link;

[0068] Sample set partitioning module 2 is used to generate a delayed sequence of chaotic signals within a certain delay range, and partition the sample set according to the original sequence and the delayed sequence;

[0069] Model building module 3 is used to establish a model between the chaotic time series and the reconstructed delayed series using a reservoir network, and to analyze the functional relationship between the two series at the correct time delay characteristic parameters;

[0070] Model performance evaluation module 4 is used to evaluate the model performance and draw the result graph after the original sequence is predicted and regressed to the delayed sequence through the training reservoir network.

[0071] The delay feature estimation module 5 is used to estimate the delay features by traversing and reconstructing the delay, thereby realizing the estimation of the delay features of the chaotic optical communication system.

[0072] Example 1

[0073] To address the shortcomings or improvement needs of existing technologies, this invention proposes a model-free delay feature estimation method based on a reservoir network. The delay parameters are directly estimated from the time series of an optical chaotic communication system. A reservoir network is used to establish a model between the sequence and its delay sequence to estimate the correlation between the two sequences. The mean square error is used as an evaluation index to explore the relationship between delay and delay parameters, thereby successfully estimating the delay parameters of the chaotic communication system.

[0074] To achieve the above objectives, this invention provides a method for estimating the delay characteristics of a chaotic optical communication system based on a reservoir network, comprising the following steps:

[0075] (1) In an optical chaotic communication system, a time-delayed chaotic signal of a certain length is collected in the transmission link, the received data is collected, and the chaotic data is denoted as... ;

[0076] (2) The received data obtained in step (1) Decompose the sequence and select a sequence length of... Time-delay chaotic signal Select a data segment of length from the sequence. for a subsequence of As the initial training signal, a segment with the same length as the initial training signal and a delay of [length missing] is selected. Delay sequence As a training target signal;

[0077] (3) While implementing step (2), in order to obtain a sufficient training set from the time series with a finite length, the dataset is formatted, and the training set format is as follows: The test prediction set format is The training data pairs and the test prediction data pairs are respectively on the time axis. Step and Move at the speed of a step, where For the length of the training data, To test the length of the predicted data;

[0078] (4) Set the parameters of the reservoir network. In step (3), divide the training set and the test prediction set. After the reservoir training is completed, a corresponding output weight can be obtained. The weight represents the correspondence between the two sequences. With the selected delay... By observing the changes in the output weights, the corresponding relationships between the delayed sequences can be obtained. Once the output weights of the reservoir network are determined, the overall network structure is completely defined. Only the test prediction set needs to be input to obtain the prediction data, i.e., another subsequence of the chaotic signal is selected. Input to corresponding output weights Predicted sequences are generated in the reservoir network. ;

[0079] (5) The mean squared error is used as the error function to calculate the predicted sequence obtained in step (4). With the actual sequence The differences are used to evaluate the performance of the prediction model, plotting the traversal delay on the x-axis and the mean squared error on the y-axis. Resulting image;

[0080] (6) In principle, the prediction result is only valid when the traversal delay and the correct delay are equal. It has relatively high accuracy, therefore, it is drawn according to step (5). When displaying the results, in The time delay will reach a minimum value, and a downward peak will appear in the graph, thus enabling the estimation of the time delay characteristic parameters.

[0081] The time delay characteristic parameter estimation method for optical chaotic communication systems in this invention directly extracts the correlation function relationship from the time series of the optical chaotic communication system, uses a reservoir network to perform predictive regression between the original sequence and the delayed sequence, and evaluates the prediction effect through mean square error. Compared with traditional mathematical statistical estimation methods, this method has certain advantages in highly nonlinear, high-noise environments and experimental conditions. Compared with other neural network methods, this method provides a time delay estimation method for optical chaotic communication systems that does not require prior knowledge and further improves computational efficiency. Therefore, it provides a new approach for ensuring the security of optical communication.

[0082] Example 2

[0083] For ease of understanding, the embodiments of the present invention use an intensity modulation experiment as an example, such as... Figure 4As shown, continuous light emitted from a TLG-200 laser diode is polarized by a polarization controller, and then electro-optically modulated by an EOSPACE AX-OMSS-20-PFA-PFA-LV MZM modulator. An optical fiber is then introduced as the optical storage medium. Due to the time delay introduced by the transmitted signal, the power of the optical signal is adjusted by a variable optical attenuator. A 10G photodetector converts the optical signal into an electrical signal. The electrical signal is split into two parts by an electrical divider. One part is amplified and then fed back into the MZM modulator. The other part is sampled and received by a Tektronix DSA 72504D digital storage oscilloscope at a rate of 50 GSa / s.

[0084] From PD acquisition to experimental waveform (50GS / s), a 3920-point (78.4ns) time series was selected for TDS parameter estimation. The parameters in the training and test prediction set data formats were set, and the training data length was... 1000, test data length 1000, training stride length 1. Test step size 1. Maximum training length 1920, testing maximum length 1920. Implement dataset formatting within the interval [0.02ns, 38.4ns] with a fixed step size of 0.02ns.

[0085] The ridge regression method was used to solve the linear regression problem for the reservoir network. The primary task of the reservoir network in this scheme is prediction. The number of reservoir nodes was set to 100, the update rate to 0.3, and the regularization coefficient to 10. -8 .like Figure 5 As shown, multiple prediction regressions are performed within the interval [0.02ns, 38.4ns] to obtain the correspondence between the time-delayed chaotic signal and its delay sequence, and the output weights for the corresponding delays are obtained. Figure 6 As shown, sequences from the test prediction set are input into a reservoir network with corresponding output weights to generate predicted sequences. Mean squared error is used to evaluate the difference between the predicted and actual sequences and to assess the performance of the prediction model.

[0086] Plot the E(s) result graph with traversal delay as the x-axis and mean square error as the y-axis. When the delay... A downward spike appears in the result graph at a time delay of 19.12 ns, thus the estimated time delay characteristic parameter can be determined to be 19.12 ns. Furthermore, the mathematical statistical estimation method ACF also yields a time delay characteristic parameter of 19.12 ns, which fully verifies that the reservoir network successfully estimates the time delay characteristic parameter of the intensity-modulated chaotic signal.

[0087] To demonstrate the inventiveness and technical value of the technical solution of this invention, this section provides application examples of the technical solution of the claims on specific products or related technologies.

[0088] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0089] In the experimental process of Embodiment 2 of the present invention, some experimental conclusions were also obtained, such as... Figure 7 As shown, the time delay characteristic parameter determined by the autocorrelation function method and the delay mutual information method is 19.12 ns, and the time delay characteristic parameter of the intensity-modulated chaotic signal estimated by the reservoir network is also 19.12 ns. In the intensity modulation experiment, the reservoir model-free estimation method requires a total of 3920 data points; therefore, data of the same length are selected for estimation using both the autocorrelation function and delay mutual information methods. The estimation results of the two methods are shown below. Figure 8 As shown. Figure 8 As shown, the autocorrelation function method and the delay mutual information method cannot estimate the time delay characteristic parameters from a time series of length 3920. Through... Figure 7 and Figure 8 This fully demonstrates that, compared with traditional mathematical statistical estimation methods, the method of this invention requires a shorter original chaotic sequence.

[0090] In implementing Example 2, a chaotic signal sequence of the intensity modulation system was obtained through simulation to analyze the influence of nonlinear factors (i.e., feedback gain). The minimum required data length was selected as the metric, demonstrating that the method of this invention has certain advantages in highly nonlinear experimental environments compared to traditional mathematical statistical estimation methods. As shown in Table 1, when the system nonlinearity is large, the autocorrelation function method cannot estimate the time delay characteristics; the delay mutual information method also requires a relatively large amount of data for estimation. However, the method of this invention can estimate the time delay parameters under highly nonlinear conditions using a smaller amount of data.

[0091] Table 1. Minimum data size required for various time delay characteristic estimation methods under different nonlinear coefficients.

[0092]

[0093] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for estimating the time delay characteristics of a chaotic optical communication system, characterized in that, The time delay characteristic estimation method for the chaotic optical communication system includes: The chaotic signal generated by the chaotic optical communication system is transmitted from the transmitter to the receiver, and the chaotic signal is captured from the transmission link. After the data is collected, the delay characteristics of the chaotic signal in the chaotic system are directly estimated. A model between the chaotic time series and the reconstructed delay series is established using a reservoir network. The functional relationship between the two series at the correct delay characteristic parameters is analyzed. The delay characteristics are estimated by traversing the reconstructed delay, thereby realizing the estimation of the delay characteristics of the chaotic optical communication system. The method for estimating the time delay characteristics of the chaotic optical communication system includes the following steps: Step 1: Generate chaotic signals using an optical chaotic communication system; Step 2: Receive and collect chaotic signals; Step 3: Generate a delayed sequence of chaotic signals within a certain delay range; Step four: Divide the sample set according to the original sequence and the delayed sequence; Step 5: Train the reservoir network to perform prediction regression from the original sequence to the delayed sequence; Step 6: Use mean squared error to evaluate the model performance and plot the results. Step 7: Estimate the time delay characteristic parameters of the optical chaotic communication system; In step five, the parameters of the reservoir network are set. In step four, the training set and the test prediction set are divided. After the reservoir training is completed, a corresponding output weight is obtained. The weights represent the correspondence between the two sequences; with the selected delay... The changes are used to obtain the corresponding relationship of the delayed sequences; after the output weights are determined, the overall network structure of the reservoir network is completely determined, and only the test prediction set needs to be input to obtain the prediction data; another subsequence of the chaotic signal is selected. Input to corresponding output weights Predicted sequences are generated in the reservoir network. ; In step six, the mean squared error is used as the error function to calculate the predicted sequence obtained in step five. With the actual sequence The differences are used to evaluate the performance of the prediction model, plotting the traversal delay on the x-axis and the mean squared error on the y-axis. Resulting image; In step seven, when the traversal delay and the correct delay are equal, the prediction result is... It has relatively high accuracy, so it is drawn according to step six. When displaying the results, in The minimum value appears at a certain time, and a downward peak appears in the graph, thus enabling the estimation of the time delay characteristic parameters.

2. The method for estimating the time delay characteristics of a chaotic optical communication system as described in claim 1, characterized in that, In steps one and two, in the optical chaotic communication system, a time-delayed chaotic signal of a certain length is collected in the transmission link, the received data is collected, and the data is recorded as... ; In step three, the received data obtained in step one is... Decompose the sequence and select a sequence length of... Time-delay chaotic signal Select a data segment of length from the sequence. for a subsequence of As the initial training signal, a segment with the same length as the initial training signal and a delay of [length missing] is selected. Delay sequence As a training target signal.

3. The time delay characteristic estimation method for a chaotic optical communication system as described in claim 2, characterized in that, In step four, while performing step three, the dataset is formatted, and the training set format is as follows: The test prediction set format is The training data pairs and the test prediction data pairs are respectively positioned on the time axis as follows: Step and Move at the speed of a step, where For the length of the training data, To test the predicted data length.

4. A time delay feature estimation system for a chaotic optical communication system applying the time delay feature estimation method for chaotic optical communication systems as described in any one of claims 1 to 3, characterized in that, The time delay characteristic estimation system for the chaotic optical communication system includes: The chaotic signal acquisition module is used to generate chaotic signals using a chaotic optical communication system and transmit them from the transmitter to the receiver, capturing the chaotic signals from the transmission link. The sample set partitioning module is used to generate a delayed sequence of chaotic signals within a certain delay range, and to partition the sample set according to the original sequence and the delayed sequence; The model building module is used to establish a model between the chaotic time series and the reconstructed delayed series using a reservoir network, and to analyze the functional relationship between the two series at the correct time delay characteristic parameters. The model performance evaluation module is used to evaluate the model performance and draw the results graph after the original sequence is predicted and regressed to the delayed sequence through the training reservoir network. The delay feature estimation module is used to estimate the delay features by traversing and reconstructing the delay, thereby realizing the estimation of the delay features of the chaotic optical communication system.

5. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the steps of the time delay feature estimation method for the chaotic optical communication system as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the time delay characteristic estimation method for a chaotic optical communication system as described in any one of claims 1 to 3.

Images