Adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network
Through the adaptive judgment threshold selection method based on long and short-term memory neural network, the problem of high bit error rate in spatial optical communication under turbulent atmosphere is solved, and the adaptive change and accuracy of judgment threshold are achieved to adapt to light in complex environments Strong ups and downs.
Patent Information
- Application Number
- CN202411322258.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-12-13
AI Technical Summary
The traditional fixed judgment threshold method cannot adaptively change the judgment threshold value and cannot effectively deal with complex scenarios of spatial optical communication under turbulent atmosphere, resulting in an increase in bit error rate.
Adaptive judgment threshold selection method based on long and short-term memory neural network is adopted, and the adaptive judgment threshold selection method is achieved by obtaining optical received signal data, sampling and segmenting, deep learning model is built, and backpropagation algorithm training is used to achieve adaptive changes in judgment thresholds.
It improves the accuracy of judgment, reduces the bit error rate, enhances the adaptability of the spatial optical communication system, and adapts to the fluctuations of light intensity in complex environments.
Smart Images
Figure CN119135262A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wireless optical communication, and in particular to an adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network. Background Art
[0002] In the past few decades, free space optical communications (FSO) have begun to receive more and more attention. Similar to radio frequency links, typical free space optical communication links include signal sources, modulators, transmitters, propagation channels and receivers. There are three main types of common free space optical communications: infrared (IR) communication, visible light communication and ultraviolet (UV) communication. According to the working mode, free space optical communications can also be divided into two categories, line of sight (LOS) link communication and non-line of sight (NLOS) link communication. Compared with traditional radio frequency (RF) communication, FSO has the advantages of wide bandwidth, no spectrum license required, and good security. However, as the communication distance increases, atmospheric turbulence will cause radiometric fluctuations (flicker), which will affect the performance of the communication system.
[0003] During the transmission of optical signals in the atmospheric channel, the absorption and scattering effects caused by the interaction of photons with gas molecules and aerosol particles in the atmosphere cause the optical signal to attenuate, resulting in a decrease in the optical power at the receiving end; after long-distance transmission, the aperture of the light receiving is smaller than the size of the light beam, resulting in geometric power loss. While being absorbed and scattered by the atmosphere, it is also affected by atmospheric turbulence. The turbulence effect causes the movement speed, temperature, and refractive index of the atmosphere to fluctuate randomly in both time and space dimensions, among which the fluctuation of the refractive index will directly affect the propagation characteristics of the optical signal. The beam jitter caused by the turbulence effect will cause the light wave to deviate from the receiving aperture and reduce the signal strength. The beam flicker caused by turbulence will cause background noise to appear in the optical signal, thereby reducing the receiving signal-to-noise ratio, thereby affecting the correct judgment and increasing the bit error rate.
[0004] The traditional signal decision method with a fixed decision threshold cannot adaptively change the decision threshold value to cope with complex scenarios such as space optical communication in a turbulent atmosphere. In view of the above problems, it is very important to study an effective adaptive signal decision method. Accurate signal decision is of great significance to space optical communication systems. In recent years, deep learning, as one of the most popular technologies in the field of machine learning, has been widely used in image recognition, time series prediction, and decision selection. Introducing deep learning to solve the signal decision problem of space optical communication in a turbulent atmosphere can use its powerful feature learning ability and excellent adaptability to help the space optical communication system adaptively adjust the decision threshold, reduce the adverse effects of atmospheric turbulence on the communication system to a certain extent, judge the signal more accurately, and effectively reduce the bit error rate. Summary of the invention
[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network. The method utilizes the neural network model and optimization algorithm in deep learning to solve the problems and shortcomings existing in the traditional method, including large fluctuations in light intensity, poor adaptability of the decision threshold to complex environments, and the need for a large amount of prior information. The method has the characteristics of adaptive changes of the decision threshold with the influence of turbulence and high decision accuracy.
[0006] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions:
[0007] On the one hand, the present invention provides an adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network, comprising:
[0008] Obtain optical receiving signal data through space optical communication experiments in turbulent atmosphere;
[0009] Sampling is performed according to the optical reception signal data, dividing the sampled optical reception signal data into signal segments of equal length, and comparing the sampled optical reception signal data with the original reception signal data to obtain an optimal decision threshold for each signal segment;
[0010] Building a long short-term memory neural network learning model, and using the signal segment and the optimal decision threshold of the signal segment to train the long short-term memory neural network learning model to obtain a long short-term memory neural network learning model for decision threshold selection;
[0011] The data that need to be selected for decision threshold are sampled and segmented in sequence, input into the long short-term memory neural network learning model used for decision threshold selection, and the decision threshold result is output.
[0012] Optionally, the space optical communication experiment in a turbulent atmosphere is implemented by a constructed space optical communication system; the space optical communication system adopts an on-off keying modulation method, and relies on a combination arrangement of unipolar non-return-to-zero codes to realize modulation of the transmission signal.
[0013] Optionally, the light receiving signal data includes light intensity fluctuations and Gaussian white noise caused by atmospheric turbulence, and its expression is as follows:
[0014]
[0015] in, Indicates the optical receiving signal data, Represents the effective photoelectric conversion ratio of the photoelectric receiver, represents the light intensity fluctuations caused by atmospheric turbulence, Indicates the transmitted signal using the on-off keying modulation method, represents Gaussian white noise.
[0016] Optionally, during sampling, the middle point of each symbol time in the light receiving signal data is used as the sampling time.
[0017] Optionally, the best decision threshold of each signal segment and the next adjacent signal segment are grouped together, and the long short-term memory neural network learning model is trained using a back propagation algorithm.
[0018] Optionally, the long short-term memory neural network learning model is a 5-layer network structure, including a convolutional layer, a pooling layer, an LSTM layer, a fully connected layer and a softmax layer.
[0019] Optionally, the long short-term memory neural network learning model uses a cross entropy function as a loss function.
[0020] In a second aspect, the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps of the method for adaptive decision threshold selection for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the first aspects are implemented.
[0021] In a third aspect, the present invention provides a computer device / equipment / system, comprising:
[0022] Memory, for storing computer programs / instructions;
[0023] A processor is used to execute the computer program / instructions to implement the steps of the adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the first aspects.
[0024] In a fourth aspect, the present invention provides a computer program product, comprising a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the steps of the adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the first aspects are implemented.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] 1. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network provided by the present invention introduces deep learning into the signal decision of space optical communication in turbulent atmosphere, and accurately decides the received signal by learning the changing characteristics of the signal in time sequence;
[0027] 2. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network provided by the present invention solves the problems and shortcomings of the traditional method by using the neural network model and optimization algorithm in deep learning, and has the characteristics of adaptive change of decision threshold with the influence of turbulence and high decision accuracy;
[0028] 3. The computer-readable storage medium, computer device / equipment / system and computer program product provided by the present invention can execute the steps of the adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network provided by the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] Figure 1 A flowchart of an adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network according to an embodiment of the present invention;
[0030] Figure 2 A structural diagram of a long short-term memory neural network learning model provided according to an embodiment of the present invention. DETAILED DESCRIPTION
[0031] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.
[0032] It should be noted that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0033] Embodiment 1:
[0034] The embodiment of the present invention discloses an adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network, which specifically includes the following steps:
[0035] S1, obtain optical receiving signal data through space optical communication experiments in turbulent atmosphere;
[0036] S2, sampling according to the optical reception signal data, dividing the sampled optical reception signal data into signal segments of equal length, and comparing with the original reception signal data to obtain the optimal decision threshold for each signal segment;
[0037] S3, building a long short-term memory neural network learning model, and using the signal segment and the optimal decision threshold of the signal segment to train the long short-term memory neural network learning model to obtain a long short-term memory neural network learning model for decision threshold selection;
[0038] S4, sampling and segmenting the data required for decision threshold selection in sequence, inputting them into the long short-term memory neural network learning model used for decision threshold selection, and outputting the decision threshold result.
[0039] Specifically,
[0040] In step S1, during the transmission of the laser beam in the atmospheric channel, the absorption and scattering effects caused by the interaction of photons with gas molecules and aerosol particles in the atmosphere cause the optical signal to attenuate, resulting in a decrease in the optical power at the receiving end; after long-distance transmission, the aperture of the light receiving is smaller than the size of the light beam, resulting in geometric power loss; the atmospheric turbulence effect interferes with the received optical signal, causing the bit error rate to increase, which seriously affects the stability and reliability of free-space optical communication; atmospheric turbulence can cause beam drift, beam expansion, and light intensity flicker, and light intensity flicker can cause random fluctuations in the received light intensity, thereby seriously degrading the communication performance of the space optical communication system.
[0041] Therefore, the space optical communication experiment under turbulent atmosphere is realized by a constructed space optical communication system; the space optical communication system adopts an on-off keying (OOK) modulation method, and relies on the combination arrangement of unipolar non-return-to-zero codes to realize the modulation of the transmission signal. When the signal "1" is to be sent, the optical pulse occupying the entire bit duration responds to a high level, and when the signal "0" is to be sent, the response continues to be a low level.
[0042] When the modulated laser beam passes through the channel affected by turbulence, the optical receiving signal data includes light intensity fluctuations and Gaussian white noise caused by atmospheric turbulence, and its expression is as follows:
[0043]
[0044] in, Indicates the optical receiving signal data, Represents the effective photoelectric conversion ratio of the photoelectric receiver, represents the light intensity fluctuations caused by atmospheric turbulence, Indicates the transmitted signal using the on-off keying modulation method, represents Gaussian white noise; the effect of turbulence on the signal is multiplicative, which essentially makes the received signal Sending a signal A low-frequency random fluctuation is generated on the basis of
[0045] In step S2, the midpoint of each symbol time in the optical reception signal data is used as the sampling time during sampling, the optical reception signal data is sampled to obtain sampled optical reception signal data, the sampled optical reception signal data is divided into signal segments of equal length, and the optimal decision threshold of each signal segment is obtained by comparing with the original signal; the optimal decision threshold of each signal segment and the next adjacent signal segment are grouped as a group, which are used as training data for offline training in step S3.
[0046] In this embodiment, the sampled light receiving signal data is divided into signal segments of equal length, the space optical communication system has The optimal decision threshold value is selected, and the divided signal segment is compared with the original signal to obtain the optimal decision threshold for each signal segment; the optimal decision threshold combination of each signal segment and the next signal segment , as the training data in step 3, where Indicates signal segment, Indicates Signal segment The best decision threshold value; and They are all positive integers and can be adjusted according to actual conditions.
[0047] A Recurrent Neural Network (RNN) is a type of recursive neural network that takes sequence data as input, recursively in the direction of sequence evolution, and all nodes (recurrent units) are connected in a chain. Recurrent neural networks are mainly used to process sequence data (such as natural language, audio, stock market price trends, etc.). They have certain advantages in learning the nonlinear characteristics of sequences and are widely used in time series prediction, audio processing, and natural language processing. A Long Short-Term Memory (LSTM) network is a type of RNN that can solve the shortcoming of RNN's short-term memory. When a sequence is long enough, it will be difficult for RNN to transmit information from an earlier time step to a later time step, while LSTM can learn long-term dependent information and remember information from earlier time steps.
[0048] In step S3, a decision threshold selection model based on the long short-term memory network algorithm is designed to make decisions on space optical communication signals in a turbulent atmosphere. The network structure and hyperparameters are preliminarily set, and the neural network model is continuously optimized using the training data obtained in step S2. The light intensity fluctuation law caused by atmospheric turbulence is obtained through the feature learning ability of the neural network, and the decision threshold is adaptively changed to achieve accurate signal decision.
[0049] The neural network model selected in this embodiment is an LSTM neural network. The LSTM network is a neural network model commonly used in sequence data processing tasks such as natural language processing. It can effectively handle long-term dependency problems in long sequence data. Its input is the signal segment obtained in step S2, and its output is the optimal decision threshold of the next signal segment in step S2. In this method, the value of the decision threshold is finite and discrete, and the LSTM neural network used is used for the decision threshold selection task.
[0050] refer to Figure 2 As shown, the LSTM model constructed in this embodiment is a 5-layer network structure, including a convolution layer, a pooling layer, an LSTM layer, a fully connected layer, and a softmax layer; the softmax layer maps the output features to the output results, which are divided into The output result is Different decision threshold values; the loss function selects the cross entropy function.
[0051] This embodiment uses a back propagation algorithm to train the above neural network. After the training is completed, the trained LSTM decision threshold selection model is used to assist the space optical communication system in signal decision, reduce the bit error rate, and improve the performance of the space optical communication system.
[0052] In summary, the adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network provided in this embodiment obtains optical receiving signal data of space optical communication in a turbulent atmosphere through experiments, and utilizes the powerful feature learning ability and excellent adaptability of deep learning. The neural network model obtained through training can help the receiving end to adaptively change the signal decision threshold, accurately complete the signal decision task, reduce the bit error rate, and improve the performance of the space optical communication system; it solves the problems and shortcomings existing in the traditional method, including large fluctuations in light intensity, the poor adaptability of the decision threshold to complex environments, and the need for a large amount of prior information; it has the characteristics of adaptive changes of the decision threshold with the influence of turbulence and high decision accuracy.
[0053] Embodiment 2:
[0054] This embodiment provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps of the method for selecting an adaptive decision threshold for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the first embodiments are implemented.
[0055] Embodiment three:
[0056] This embodiment provides a computer device / equipment / system, including:
[0057] Memory, for storing computer programs / instructions;
[0058] A processor is used to execute the computer program / instructions to implement the steps of the adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the embodiments in Example 1.
[0059] Embodiment 4:
[0060] The present embodiment provides a computer program product, including a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the steps of the method for selecting an adaptive decision threshold for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of the first embodiments are implemented.
[0061] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0062] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0063] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0064] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0065] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.
Claims
1. An adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network, It is characterized in that include: Obtain optical receiving signal data through space optical communication experiments in turbulent atmosphere; Sampling is performed according to the optical reception signal data, dividing the sampled optical reception signal data into signal segments of equal length, and comparing the sampled optical reception signal data with the original reception signal data to obtain an optimal decision threshold for each signal segment; Building a long short-term memory neural network learning model, and using the signal segment and the optimal decision threshold of the signal segment to train the long short-term memory neural network learning model to obtain a long short-term memory neural network learning model for decision threshold selection; The data that need to be selected for decision threshold are sampled and segmented in sequence, input into the long short-term memory neural network learning model used for decision threshold selection, and the decision threshold result is output.
2. According to claim 1, the adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network, It is characterized in that The space optical communication experiment under the turbulent atmosphere is realized by a constructed space optical communication system; the space optical communication system adopts an on-off keying modulation method and relies on a combination arrangement of unipolar non-return-to-zero codes to realize modulation of the transmission signal.
3. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network according to claim 2, It is characterized in that The light receiving signal data includes light intensity fluctuations caused by atmospheric turbulence and Gaussian white noise, and its expression is as follows: in, Indicates the optical receiving signal data, Represents the effective photoelectric conversion ratio of the photoelectric receiver, represents the light intensity fluctuations caused by atmospheric turbulence, Indicates the transmitted signal using the on-off keying modulation method, represents Gaussian white noise.
4. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network according to claim 1, It is characterized in that During sampling, the middle point of each symbol time in the light receiving signal data is used as the sampling time.
5. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network according to claim 1, It is characterized in that The best decision thresholds of each signal segment and the next adjacent signal segment are grouped together, and the long short-term memory neural network learning model is trained by using a back propagation algorithm.
6. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network according to claim 1, It is characterized in that The long short-term memory neural network learning model is a 5-layer network structure, including a convolutional layer, a pooling layer, an LSTM layer, a fully connected layer and a softmax layer.
7. The adaptive decision threshold selection method for space optical communication in turbulent atmosphere based on long short-term memory neural network according to claim 1, It is characterized in that The long short-term memory neural network learning model adopts a cross entropy function as a loss function.
8. A computer-readable storage medium having a computer program / instructions stored thereon, It is characterized in that When the computer program / instruction is executed by a processor, the steps of the method for selecting an adaptive decision threshold for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of claims 1 to 7 are implemented.
9. A computer device / equipment / system, It is characterized in that include: Memory, for storing computer programs / instructions; A processor, used to execute the computer program / instructions to implement the steps of the adaptive decision threshold selection method for space optical communication in a turbulent atmosphere based on a long short-term memory neural network according to any one of claims 1 to 7.
10. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instruction is executed by a processor, the steps of the method for selecting an adaptive decision threshold for space optical communication in a turbulent atmosphere based on a long short-term memory neural network as described in any one of claims 1-7 are implemented.