Wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization

Abstract

The invention belongs to the technical field of wireless optical communication, and particularly relates to a wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization. The method comprises the following steps: firstly, acquiring multi-dimensional interference signals such as light intensity flicker and polarization state offset by using a sensor array, extracting related characteristics, and then establishing an interference mapping model through a ResNet neural network; multi-wavelength allocation is triggered according to the interference level, an optimization function is constructed, an optimal wavelength allocation matrix is generated with the help of an intelligent algorithm, and the parameter weight can be adjusted along with the interference level; the transmitting end converts the electric signals into optical signals and transmits multi-wavelength diversity signals according to a matrix, and the receiving end receives and converts the multi-wavelength diversity signals into electric signals; and finally, integrating the electric signals by improving maximum ratio combination, including the steps of time sequence alignment, channel stability calculation, weight updating, signal combination and the like. According to the method, various interference problems are effectively solved, and the reliability and the anti-interference capability of wireless optical communication are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless optical communication, and particularly relates to an anti-interference method for wireless optical communication based on diversity technology and multi-wavelength optimization. Background Art

[0002] Wireless optical communication exhibits important application value in scenarios such as satellite-ground communication and short-distance high-speed interconnection due to its advantages of high bandwidth, anti-electromagnetic interference, and flexible deployment. However, signal transmission is vulnerable to various interferences, severely restricting the communication quality. For example, light intensity scintillation in the atmospheric channel can cause fluctuations in the received optical power, leading to signal bit errors; polarization state offset can change the transmission characteristics of optical signals, affecting the effective reception of signals; co-frequency interference signals can confuse useful signals, reducing signal recognition; and aiming error angle can also cause signal transmission loss. The existence of these interference factors makes the existing wireless optical communication have significant defects in terms of stability and reliability, and it is difficult to meet the growing demand for high-speed and stable communication. Summary of the Invention

[0003] The present invention addresses the technical problems existing in the above background art and proposes an anti-interference method for wireless optical communication based on diversity technology and multi-wavelength optimization.

[0004] To achieve the above objective, the technical solution adopted by the present invention includes the following steps:

[0005] S1. Interference feature perception: Collect multi-dimensional interference signals including light intensity scintillation, polarization state offset, co-frequency interference signals, and aiming error angle through a sensor array, extract multi-dimensional features, and establish an interference mapping model;

[0006] S2. Multi-wavelength dynamic allocation: Trigger multi-wavelength allocation according to the interference level of the interference model, construct an optimization function including wavelength orthogonality, channel gain, and aiming error, and generate an optimal wavelength allocation matrix through an intelligent algorithm;

[0007] The calculation method of the optimization function is as follows: Where Orth(λ) represents the normalized value of the wavelength interval, measuring the orthogonality between different wavelengths, μ(G) represents the mean value of the channel gain, σ(G) represents the standard deviation of the channel gain, ε is a non-zero extremely small constant, G T is the gain coefficient of the transmitting antenna, θ is the error angle, and α1, α2, α3 are parameter weights;

[0008] Among them, Among them, K is the number of wavelengths, λ max is the maximum value of the wavelength tuning range;

[0009] S3. Signal Transmission: The transmitting end first converts the electrical signal into an optical signal and generates a multi-wavelength signal according to the wavelength allocation matrix, and then transmits it through the space diversity antenna to obtain a multi-wavelength diversity signal.

[0010] S4. Signal Reception: The receiving end acquires the multi-wavelength diversity signal and converts the signal into an electrical signal.

[0011] S5. Signal Integration: Perform an electrical signal integration operation, which is implemented based on improved maximum ratio combining.

[0012] Preferably, the sensor array includes a high-speed optical intensity sensor for collecting the time-domain fluctuation data of the optical intensity scintillation signal; a polarization state sensor for measuring the polarization state rotation angle and extinction ratio parameter of the optical signal; and a broadband optical spectrum analyzer for obtaining the gain distribution of each wavelength channel and the spectral characteristics of the interference signal.

[0013] Preferably, the establishment of the interference mapping model includes feature extraction and model establishment.

[0014] The feature extraction is to extract the optical intensity scintillation index where represents the variance of the optical intensity I, is the average value of the optical intensity, the polarization rotation matrix R(Δθ), where Δθ represents the angle of the polarization state, the spectral distribution of the interference signal P(f), where f represents the power, and the aiming error angle θ.

[0015] The model establishment is to establish an interference mapping model through a ResNet neural network. The input layer is the features extracted above, and the output layer is the interference level.

[0016] Preferably, the parameter weights α1, α2, and α3 in step S2 are corrected according to different interference levels, and the calculation method of the correction is

[0017] Preferably, the space diversity antenna array has a uniform circular array structure, and the distance between adjacent antennas satisfies: where λ avg is the average wavelength, θ max is the maximum arrival angle spread, C is the atmospheric refractive index structure constant, and k is the turbulence influence coefficient.

[0018] Preferably, the specific steps of performing the electrical signal integration operation based on improved maximum ratio combining in step S5 are as follows:

[0019] Timing Alignment: Perform a timing alignment operation on the electrical signals of each wavelength converted by the receiving end.

[0020] Calculate Channel Stability: Calculate the channel stability C stability in the current time window, and its calculation method Among them represents the sum of the changes in the channel gain within this time window, represents the total sum of the channel gain within this time window;

[0021] Determine the initial weights: According to the channel states of the signals of each wavelength at the initial moment;

[0022] Iteratively update the weights: Starting from the second moment, according to the weight improvement calculation method: iteratively update the weights of the signals of each wavelength, where ω l (t) represents the weight of the l-th wavelength signal at the t-th moment, 0 ≤ l ≤ K, μ is the step factor, and e(t) is the error of the combined signal at the t-th moment, represents the conjugate of the l-th wavelength signal at the t-th moment;

[0023] Signal combination: Use the updated weights of the signals of each wavelength to perform weighted combination on the electrical signals of each wavelength after time alignment to obtain the final signal.

[0024] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0025] 1. Precise multi-dimensional interference perception: Collect multi-dimensional signals such as optical intensity scintillation and polarization offset through a sensor array, extract four-dimensional features and use a ResNet neural network for modeling, break through the limitation of single-feature detection, achieve precise classification in complex interference environments, and provide a data basis for anti-interference strategies.

[0026] 2. Dynamically optimize wavelength allocation: Construct an optimization function, dynamically adjust the weights according to the interference level, generate an optimal wavelength matrix through an intelligent algorithm, solve the problem of poor adaptability of fixed allocation to time-varying interference, and improve the utilization rate of wavelength resources and anti-interference ability.

[0027] 3. Enhancement of diversity technology: Use a uniform circular array diversity antenna to suppress turbulence fading, combine multi-wavelength transmission to form a diversity gain; at the receiving end, improve the maximum ratio combining algorithm, dynamically adjust the weights based on channel stability, improve the signal integration quality, significantly reduce the bit error rate, and achieve stable communication in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0029] Figure 1It is a structural flowchart of an anti-interference method for wireless optical communication based on diversity technology and multi-wavelength optimization. Specific implementation manners

[0030] In order to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described below with reference to the accompanying drawings and embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

[0031] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the present invention is not limited by the specific embodiments disclosed in the following specification.

[0032] Embodiment. In the field of wireless optical communication, problems such as optical intensity scintillation, polarization state offset, co-frequency interference and aiming error in the atmospheric channel seriously affect the signal transmission quality, resulting in an increase in the bit error rate and a decrease in communication stability. In order to effectively solve these interference problems and improve the reliability and anti-interference ability of wireless optical communication, the present invention proposes an anti-interference method for wireless optical communication based on diversity technology and multi-wavelength optimization. The specific implementation process is as Figure 1 shown.

[0033] First, interference feature perception is performed. In the prior art, single interference feature detection cannot comprehensively reflect a complex interference environment, resulting in insufficient pertinence of anti-interference strategies. The interference feature perception of the present invention is to collect multi-dimensional interference signals including optical intensity scintillation, polarization state offset, co-frequency interference signals and aiming error angles through a sensor array, and extract multi-dimensional features and establish an interference mapping model. The sensor array includes a high-speed optical intensity sensor for collecting time-domain fluctuation data of optical intensity scintillation signals; a polarization state sensor for measuring the polarization state rotation angle and extinction ratio parameters of optical signals; and a broadband spectrum analyzer for obtaining the gain distribution of each wavelength channel and the spectrum characteristics of interference signals.

[0034] Then, an interference mapping model is established according to the above-extracted features. The interference mapping model includes feature extraction and model establishment.

[0035] The feature extraction is to extract the optical intensity scintillation index where represents the variance of the optical intensity I, is the average value of the optical intensity, the polarization rotation matrix R(Δθ), where Δθ represents the angle of the polarization state, the interference signal spectrum distribution P(f), where f represents the power, and the aiming error angle θ.

[0036] The model is established by using the ResNet neural network to establish an interference mapping model. The input layer is the above-extracted features, and the output layer is the interference level. Specifically, after the input layer of the ResNet neural network receives the feature vector, it is transformed into a shape suitable for processing by the residual blocks through the fully connected layer. The main body of the network adopts a deep residual structure, including multiple residual blocks. Each residual block extracts features through two layers of convolution, batch normalization, and ReLU activation function, and directly maps the input to the output through skip connections, avoiding the problem of gradient disappearance in the training of deep networks. As the number of network layers deepens, the feature dimension is gradually halved and the number of channels is doubled through stride convolution to achieve layer-by-layer abstraction of features. After compressing the features through the global average pooling layer, the output layer is mapped to 5 interference levels through the fully connected layer and the Softmax function.

[0037] Considering that the traditional fixed wavelength allocation method cannot adapt to the time-varying interference environment, resulting in low wavelength resource utilization and limited anti-interference effect. The present invention performs multi-wavelength dynamic allocation, triggers multi-wavelength allocation according to the interference level of the interference model, constructs an optimization function including wavelength orthogonality, channel gain, and aiming error, and generates an optimal wavelength allocation matrix through an intelligent algorithm. Specifically, the calculation method of the optimization function is as follows: where Orth(λ) represents the normalized value of the wavelength interval, measuring the orthogonality between different wavelengths, μ(G) represents the mean value of the channel gain, σ(G) represents the standard deviation of the channel gain, ε is a non-zero extremely small constant, G T is the gain coefficient of the transmitting antenna, θ is the error angle, and α1, α2, α3 are parameter weights. Among them, where K is the number of wavelengths, λ max is the maximum value of the wavelength tuning range, and the parameter weights of the optimization function are calculated according to the current interference level. The calculation method is

[0038] When using the particle swarm optimization algorithm to solve the optimal wavelength allocation matrix, first randomly generate N initial particles. Each particle is represented in matrix form and strictly satisfies the single-wavelength allocation constraint, that is, only one element in each row of the matrix is 1, and the rest are 0. Then, calculate the fitness of each particle. The fitness function is designed based on the optimization goal that fuses wavelength orthogonality, channel gain, and aiming error. During the iteration process, each particle updates its own position according to its individual historical best position and the global best position, and always maintains the single-wavelength allocation constraint during the update to ensure that only one wavelength is selected in each row. By continuously comparing and optimizing, the particle swarm gradually searches for a better solution. The algorithm sets the maximum number of iterations, and finally outputs the matrix with the highest fitness as the optimal wavelength allocation matrix after reaching the termination condition.

[0039] Next, considering that single transmission is vulnerable to fading and has low signal reliability. In this step, a uniform circular array space diversity antenna and multi-wavelength signal transmission are used to solve the signal attenuation problem in a fading channel. At the transmitting end, the electrical signal is first converted into an optical signal and a multi-wavelength signal is generated according to the wavelength allocation matrix, and then transmitted through the space diversity antenna to obtain a multi-wavelength diversity signal. The space diversity antenna array has a uniform circular array structure, and the distance between adjacent antennas satisfies: where λ avg is the average wavelength, θ max is the maximum arrival angle spread, C is the atmospheric refractive index structure constant, and k is the turbulence influence coefficient.

[0040] For signal reception, the receiving end obtains the multi-wavelength diversity signal and converts the signal into an electrical signal. Specifically, the received optical signal contains multiple wavelength components, and interference and noise may be superimposed after transmission through the atmospheric channel. Subsequently, the signal enters a high-speed photodetector, and the optical power change of the optical signal is converted into an electrical signal using the photoelectric effect. The converted electrical signal first passes through a low-noise amplifier to compensate for the loss during transmission and enhance the signal, and then the out-of-band noise is filtered by a band-pass filter, and finally a relatively pure electrical signal is output, providing a basis for subsequent signal integration and processing to ensure the effectiveness of anti-interference communication.

[0041] Finally, for signal integration, the electrical signal integration operation is implemented based on improved maximum ratio combining. The specific implementation steps are as follows: timing alignment: perform a timing alignment operation on the electrical signals of each wavelength converted at the receiving end; calculate the channel stability: within the current time window, calculate the channel stability C stability , and its calculation method is where represents the sum of the changes in the channel gain within this time window, represents the total of the channel gain within this time window; determine the initial weight: according to the channel state of each wavelength signal at the initial moment; iteratively update the weight: starting from the second moment, according to the weight improvement calculation method: iteratively update the weights of the signals of each wavelength, where ω l (t) represents the weight of the l-th wavelength signal at the t-th moment, 0 ≤ l ≤ K, μ is the step factor, e(t) is the error of the combined signal at the t-th moment, represents the conjugate of the l-th wavelength signal at the t-th moment; signal combination: use the updated weights of the signals of each wavelength to perform weighted combination on the electrical signals of each wavelength after timing alignment to obtain the final signal.

[0042] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization, characterized in that: The following steps are involved: S1. Interference feature perception: The sensor array is used to collect multi-dimensional interference signals including light intensity flicker, polarization state offset, co-frequency interference signals and aiming error angles, and the multi-dimensional features are extracted to establish an interference mapping model. S2, multi-wavelength dynamic allocation: trigger multi-wavelength allocation according to the interference level of the interference model, build an optimization function including wavelength orthogonality, channel gain and aiming error, and generate the optimal wavelength allocation matrix through intelligent algorithm; The optimization function is calculated as follows: Where Orth(λ) represents the normalized value of wavelength interval, which measures the degree of orthogonality between different wavelengths, μ(G) represents the mean channel gain, σ(G) represents the standard deviation of channel gain, ε is a very small constant not equal to 0, G T is the transmitting antenna gain coefficient, θ is the error angle, α1, α2, α3 are parameter weights; in, Where K is the number of wavelengths, λ max is the maximum value of the wavelength tuning range; S3, signal transmission: The transmitting end first converts the electrical signal into an optical signal and generates a multi-wavelength signal according to the wavelength allocation matrix, and transmits it through the spatial diversity antenna to obtain a multi-wavelength diversity signal; S4, signal reception: The receiving end obtains the multi-wavelength diversity signal and converts the signal into an electrical signal; S5. Signal integration: performing an electrical signal integration operation, wherein the electrical signal integration operation is implemented based on improved maximum ratio combining.

2. According to claim 1, a wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization is characterized in that: The sensor array includes a high-speed light intensity sensor for collecting time domain fluctuation data of light intensity flicker signals; a polarization state sensor for measuring the polarization state rotation angle and extinction ratio parameters of optical signals; and a broadband spectrum analyzer for obtaining the gain distribution of each wavelength channel and the spectrum characteristics of interference signals.

3. According to claim 1, a wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization is characterized in that: The establishment of the interference mapping model includes feature extraction and model establishment; The feature extraction is to extract the light intensity flicker index in represents the variance of the light intensity I, is the average value of the light intensity, the polarization rotation matrix R(Δθ), where Δθ represents the angle of the polarization state, the interference signal spectrum distribution P(f), where f represents the power, and the aiming error angle θ; The model is established by using a ResNet neural network to establish an interference mapping model, the input layer is the features extracted above, and the output layer is the interference level.

4. According to claim 1, a wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization is characterized in that: The parameter weights α1, α2, and α3 in step S2 are modified according to the different interference levels. The modified calculation method is:

5. According to claim 1, a wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization is characterized in that: The spatial diversity antenna array is a uniform circular array structure, and the distance between adjacent antennas meets the following requirements: where λ avg is the average wavelength, θ max is the maximum arrival angle diffusion, C is the atmospheric refractive index structure constant, and k is the turbulence influence coefficient.

6. The wireless optical communication anti-interference method based on diversity technology and multi-wavelength optimization according to claim 1 is characterized in that: The specific steps of the step S5 to perform the electrical signal integration operation based on the improved maximum ratio combining are: Timing alignment: Perform timing alignment on the electrical signals of each wavelength converted by the receiving end; Calculate channel stability: Calculate channel stability C within the current time window stability , which is calculated as in represents the sum of the changes in channel gain within the time window, represents the sum of the channel gains in this time window; Determine the initial weight: according to the channel state of each wavelength signal at the initial moment; Iteratively update weights: From the second moment on, the calculation method is improved according to the weights: The weight of each wavelength signal is iteratively updated, where: ω l (t) represents the weight of the lth wavelength signal at the tth time, 0≤l≤K, μ is the step size factor, e(t) is the error of the combined signal at the tth time, represents the conjugate of the lth wavelength signal at the tth time; Signal merging: Use the updated signal weights of each wavelength to perform weighted merging on the time-aligned electrical signals of each wavelength to obtain the final signal.

Images