Infinitely long time sequence laser atmospheric transmission simulation method

Abstract

The invention discloses an infinite time sequence laser atmospheric transmission simulation method, and belongs to the technical field of laser communication and atmospheric optical simulation. The method comprises the following steps: firstly, extracting ERA5 meteorological data, and constructing a non-uniform atmospheric turbulence profile; determining a phase screen spacing according to system parameters, and pre-generating each layer of initial phase screen and a prediction and update matrix for recursion; in a transmission link, Fresnel diffraction is calculated layer by layer by using a GPU accelerated step-by-step Fourier transmission method; in a time sequence evolution link, a traditional freezing turbulence ultra-large phase screen mode is abandoned, integer pixel updating and sub-pixel interpolation correction are carried out on phase screen displacement according to the transverse wind speed of each space layer, new phase data are generated in a recursive mode on the wind inlet side, external phase data of a grid are deleted, and continuous time sequence evolution of the phase screen is achieved on the grid with the fixed size. According to the invention, high-precision long-time-sequence laser transmission dynamic simulation at any place is realized under relatively low hardware overhead.

Description

Technical Field

[0001] This invention relates to the fields of laser communication and atmospheric optics simulation technology, and in particular to a method for simulating infinitely long time-series laser atmospheric transmission. Background Technology

[0002] Numerical simulation of laser turbulent transport, as the foundation of laser atmospheric transport, has been widely applied in fields such as space-to-ground laser communication, high-energy laser equipment, and adaptive optics technology.

[0003] Existing numerical simulations of laser turbulent transport typically rely on the "frozen turbulence assumption," employing a method of pre-generating a large phase screen to simulate time-varying turbulence. However, this method is limited by hardware processing bottlenecks, making it difficult to generate large-scale phase screens that can support long-term transport. Furthermore, existing simulations largely depend on general empirical models, making it difficult to integrate real meteorological data under specific spatiotemporal conditions. This results in simulation results lacking environmental adaptability and exhibiting significant deviations from actual atmospheric transport characteristics.

[0004] Chinese patent documents with publication numbers CN118278070A and CN118233327A respectively adopted an improved static calculation method for power spectrum and a dynamic simulation method based on phase screen interpolation clipping. However, the former lacks evolution details in the time dimension, while the latter has the problem of complex calculation logic and difficulty in ensuring strict spatial continuity of phase, making it difficult to balance dynamic realism and calculation efficiency.

[0005] Chinese patent document CN118518211A discloses a hardware-in-the-loop simulation device based on a spatial light modulator (SLM), which simulates the atmospheric environment by loading grayscale images. However, it is highly dependent on expensive hardware performance and is difficult to flexibly access global meteorological data for large-scale statistical analysis, thus limiting its scalability.

[0006] Therefore, there is an urgent need for a simulation method that can combine real multi-dimensional meteorological data and no longer rely on pre-generated large screens, so as to realize laser transmission simulation at any time, any location, and for an unlimited duration. Summary of the Invention

[0007] This invention provides an infinitely long time-series laser atmospheric transmission simulation method that can comprehensively consider global real meteorological data driving, non-uniform turbulent path layering, autoregressive phase recursive evolution and GPU parallel acceleration, supports continuous simulation of any geographical location and any duration, and achieves high-precision laser atmospheric transmission performance evaluation with low hardware resource consumption.

[0008] A simulation method for infinitely long time-series laser atmospheric transport includes: (1) Input the latitude, longitude and time information of the target transmission path, extract the meteorological parameters of the transmission path, and construct the refractive index structure constant that varies with distance. Profile; (2) Configure the wavelength of the laser system Waist radius Sampling grid To construct the initial light field And determine the total number of phase screens. ; (3) Utilizing the refractive index structure constant Total number of profiles and phase screens Determine the axial spacing of each phase screen along the transmission path. and physical dimensions Simultaneously, initial phase screens for each spatial layer are pre-generated. and the prediction matrix used for phase evolution and update matrix , Indicates the first Layered phase screen; (4) Utilize GPU acceleration to execute the step-by-step Fourier transmission method, perform layer-by-layer Fresnel diffraction transmission, and transmit the initial light field sequentially with Each phase screen acts to obtain Light field at the end of time ; (5) The first one extracted according to step (1) Spatial meteorological data, calculate the preset time step Translation of the inner phase screen Based on translation amount Generate new phase data This is combined with the existing phase screen and trimmed to form a new phase screen. ; (6) Using a preset time step Repeat steps (4) to (5) until the preset total time is reached. The simulation sequence yielded a series of light fields transmitted through atmospheric turbulence. , The light intensity distribution map was obtained and statistical analysis was performed.

[0009] In step (1), the refractive index structure constant that varies with distance is constructed. The specific process of contouring is as follows: Based on the latitude, longitude, and time of the target transmission path, multi-layer meteorological parameters corresponding to the grid points of the transmission path are retrieved from the ERA5 database. The meteorological parameters include at least: air temperature, air pressure, and meridional and zonal wind speeds. At altitudes of 1 km and below, the refractive index structure constants of each space layer were calculated using the temperature fluctuation method and the Monin-Obukhov similarity theory. Calculations based on the Tatarskii formula at altitudes above 1km .

[0010] In step (3), the axial spacing of each phase screen along the transmission path is determined. and physical dimensions The specific process is as follows: Calculate the weighted turbulence intensity along the entire transmission path Using the principle of equal-weighted integration, the turbulence intensity of each phase screen is determined as follows: ,calculate This ensures that the turbulence intensity is equal on each phase screen; subsequently, based on the initial light field... Transmit to the The physical dimensions of the phase screen are dynamically set, and the physical width of the phase screen is dynamically adjusted. .

[0011] In step (3), the initial phase screen The generation method is as follows: Set atmospheric internal scale and external dimensions For each phase screen layer, for The atmospheric coherence length can be obtained by integrating the distance. ,based on Generate the initial phase screen .

[0012] In step (3), the prediction matrix and update matrix The generation method is as follows: A template point set consisting of boundary band sampling points and sparse additional points is constructed. The joint covariance between the template points and the new pixels to be generated is calculated based on the von Kármán phase covariance model. Based on this, the prediction matrix for recursively generating new phase data is pre-calculated. and update matrix .

[0013] In step (5), the preset time step is calculated. Translation of the inner phase screen The formula is: ; in, The wind speed at each space level is projected perpendicular to the direction of light propagation. (This will be discussed later.) It is decomposed into integer pixel displacement and subpixel displacement.

[0014] In step (5), for integer pixel displacement, new phase data is generated on the air inlet side. The formula is: ; in, Given the phase vector of the known template point at the edge of the current phase screen. It is a random noise vector that follows a standard normal distribution; For subpixel displacement, bilinear interpolation is used to fine-tune the updated grid, and then phase data outside the grid is removed.

[0015] In step (6), the final time-series light field is obtained. Then, the intensity distribution map is obtained by squaring the complex amplitude of the light field. , The simulation results were analyzed to determine the deviation from the theoretical values, including the long-exposure spot radius. Light intensity scintillation index .

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention surpasses the traditional method of pre-generating ultra-large phase screens. By introducing ERA5 reanalysis data and an autoregressive recursive generation strategy, it enables laser atmospheric transmission simulations with unlimited duration and arbitrary geographical environments on ordinary workstations. This method solves the problem of high hardware overhead in long-term simulations and significantly improves the physical realism of the simulation environment. Combined with GPU acceleration and HDF5 streaming storage, it features high computational efficiency, large data throughput, and strong scalability, making it suitable for large-scale laser communication bit error rate analysis and tracking algorithm verification. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0018] Figure 1 This is a flowchart of a simulation method for infinitely long time-series laser atmospheric transmission according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the laser transmission in a turbulent field, the size and spacing of each phase screen, and the present invention.

[0020] Figure 3 This is a schematic diagram of a single iteration process of the turbulent phase screen of the present invention.

[0021] Figure 4 This is a comparison and verification diagram between the simulation results and theoretical values ​​in the embodiments of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be noted that, unless otherwise specified, the features in the following embodiments and implementation methods can be combined with each other.

[0024] like Figure 1 As shown, a simulation method for infinitely long time-series laser atmospheric transport includes the following steps: Step 1: Construct a data-driven model of a non-uniform atmospheric environment.

[0025] A simulation scenario was set up, for example: from 39.9°N, 116.4°E, altitude 0km to 40.0°N, 116.4°E, altitude 10km, with a time range of 12:00-12:01 on July 15, 2025. Temperature, air pressure, and meridional and zonal wind speed data corresponding to the spatiotemporal points were extracted from the ERA5 database, and the data were fitted and interpolated. The pressure layer was converted into geometric height using the geopotential height formula. For areas with an altitude of 1km or less, the temperature fluctuation method and Monin-Obukhov similarity theory were used to estimate... For areas above 1 km in altitude, the Tatarskii formula is used to calculate the refractive index structure constant based on wind shear and temperature gradients along the potential height gradient. .

[0026] Step 2, System parameter settings.

[0027] Set laser wavelength Launch beam waist radius Set the simulation mesh size to (For example Based on the above parameters, an initial light field is constructed. And set the number of phase screens .

[0028] Step 3: Generation of initial phase screen, prediction matrix and update matrix.

[0029] (1) Determine the phase screen spacing. Calculate the weighted turbulence intensity. The turbulence intensity is then piecewise integrated from the starting point until a suitable phase screen spacing is found such that the integral value for each turbulence intensity interval is equal and all are... ,Sure .

[0030] (2) Determine the size of the phase screen. Calculate the initial light field propagation to the [missing information]. The physical dimensions of the phase screens in layers determine the physical width of each phase screen. Slightly larger than the light field The effective size (e.g., 1.1 times).

[0031] (3) Initial phase screen and update matrix generation. Atmospheric scale settings are established. and external dimensions For each phase screen layer, by... Integrating the integral yields the corresponding atmospheric coherence length. ,based on Generate the initial phase screen Subsequently, an update matrix is ​​generated. Template points are defined by the edge-band pixels of the phase screen and sparse auxiliary pixels. Based on the von Kármán phase covariance model, a joint covariance matrix of the template points and the new pixels to be generated is constructed and divided into blocks. Using a conditional Gaussian model, the prediction matrix is ​​first calculated. Then, the innovation covariance is calculated and subjected to SVD decomposition to obtain the update matrix. .

[0032] Step 4: GPU-accelerated stepwise optical field transport calculation (e.g.) Figure 2 (As shown).

[0033] A GPU parallel computing architecture is used to execute the step-by-step Fourier transport method. The complex amplitude distribution of the initial light field is assumed to be time-invariant. The light field continuously undergoes a cyclic process of vacuum diffraction transmission and phase screen modulation, such as... Figure 2 As shown. The Fresnel diffraction during light propagation is solved using the Fast Fourier Transform (FFT):

[0034] in, and These represent the Fourier transform and inverse Fourier transform, respectively. After the light field reaches the receiving plane, it is resampled and integrated according to the detector pixel size to obtain... .

[0035] Step 5, autoregressive recursive update of the phase screen (e.g.) Figure 3 (As shown).

[0036] Let the time step for each calculation be... For the first The phase screen converts the wind speed in each spatial layer into a lateral wind speed perpendicular to the direction of light propagation. Calculate its in Lateral displacement within and will It is decomposed into two parts: integer pixel update and subpixel update.

[0037] (1) Integer pixel update. For example... Figure 3 As shown, an autoregressive prediction model is used to generate new phase data with spatial correlation on the wind inlet side. : ; in, Given the phase vector of the known template point at the edge of the current phase screen. It is a random noise vector that follows a standard normal distribution.

[0038] (2) Subpixel correction. For tiny displacements of less than one pixel, bilinear interpolation is used to fine-tune the updated grid to ensure the smoothness of continuous time evolution, while data outside the grid is deleted.

[0039] Step 6: Algorithm loop execution and data analysis (e.g.) Figure 4 (As shown).

[0040] Store the current time light field And determine whether the preset total simulation time has been reached. If it is not achieved, then Return to step 4; if the target has been reached, end the simulation and square the amplitudes of all light fields to obtain the light intensity distribution at the receiver. And store the data in an HDF5 file. According to Theoretical long exposure radius calculated using equal parameters and flicker index This serves as a benchmark for verifying simulation accuracy and for calculating actual data. and , generate as Figure 4 The statistical results shown are as follows: (1) Spot radius verify( Figure 4 Above: Plot the evolution curve of the long exposure spot radius over time. The figure shows that the relative error between the simulation-obtained long exposure radius convergence value and the theoretical formula is less than 1%, verifying the accuracy of the infinitely long phase screen in low-frequency statistical characteristics.

[0041] (2) Scintillation Index statistics( Figure 4 (Below): Calculate the normalized variance of the on-axis light intensity sequence at the receiving end to generate the scintillation index. The curves showing changes in data volume were used to evaluate the impact of light intensity fluctuations caused by atmospheric turbulence on the bit error rate of the laser communication system. The final data error was less than 10%, validating the effectiveness of this method in simulating the log-normal distribution characteristics of light intensity.

[0042] The embodiments described above provide a detailed explanation of the technical solutions and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A simulation method for infinitely long time-series laser atmospheric transport, characterized in that, include: (1) Input the latitude, longitude and time information of the target transmission path, extract the meteorological parameters of the transmission path, and construct the refractive index structure constant that varies with distance. Profile; (2) Configure the wavelength of the laser system Waist radius Sampling grid To construct the initial light field And determine the total number of phase screens. ; (3) Utilizing the refractive index structure constant Total number of profiles and phase screens Determine the axial spacing of each phase screen along the transmission path. and physical dimensions Simultaneously, initial phase screens for each spatial layer are pre-generated. and the prediction matrix used for phase evolution and update matrix , Indicates the first Layered phase screen; (4) Utilize GPU acceleration to execute the step-by-step Fourier transmission method, perform layer-by-layer Fresnel diffraction transmission, and transmit the initial light field sequentially with Each phase screen acts to obtain Light field at the end of time ; (5) The first one extracted according to step (1) Spatial meteorological data, calculate the preset time step Translation of the inner phase screen Based on translation amount Generate new phase data This is combined with the existing phase screen and trimmed to form a new phase screen. ; (6) Using a preset time step Repeat steps (4) to (5) until the preset total time is reached. The simulation sequence yielded a series of light fields transmitted through atmospheric turbulence. , The light intensity distribution map was obtained and statistical analysis was performed.

2. The simulation method for infinitely long time-series laser atmospheric transmission according to claim 1, characterized in that, In step (1), the refractive index structure constant that varies with distance is constructed. The specific process of contouring is as follows: Based on the latitude, longitude, and time of the target transmission path, multi-layer meteorological parameters corresponding to the grid points of the transmission path are retrieved from the ERA5 database. The meteorological parameters include at least: air temperature, air pressure, and meridional and zonal wind speeds. At altitudes of 1 km and below, the refractive index structure constants of each space layer were calculated using the temperature fluctuation method and the Monin-Obukhov similarity theory. Calculations based on the Tatarskii formula at altitudes above 1km .

3. The simulation method for infinitely long time-series laser atmospheric transmission according to claim 1, characterized in that, In step (3), the axial spacing of each phase screen along the transmission path is determined. and physical dimensions The specific process is as follows: Calculate the weighted turbulence intensity along the entire transmission path Using the principle of equal-weighted integration, the turbulence intensity of each phase screen is determined as follows: ,calculate This ensures that the turbulence intensity is equal on each phase screen; subsequently, based on the initial light field... Transmit to the The physical dimensions of the phase screen are dynamically set, and the physical width of the phase screen is dynamically adjusted. .

4. The simulation method for infinitely long time-series laser atmospheric transport according to claim 1, characterized in that, In step (3), the initial phase screen The generation method is as follows: Set atmospheric internal scale and external dimensions For each phase screen layer, for The atmospheric coherence length can be obtained by integrating the distance. ,based on Generate the initial phase screen .

5. The simulation method for infinitely long time-series laser atmospheric transport according to claim 1, characterized in that, In step (3), the prediction matrix and update matrix The generation method is as follows: A template point set consisting of boundary band sampling points and sparse additional points is constructed. The joint covariance between the template points and the new pixels to be generated is calculated based on the von Kármán phase covariance model. Based on this, the prediction matrix for recursively generating new phase data is pre-calculated. and update matrix .

6. The simulation method for infinitely long time-series laser atmospheric transmission according to claim 1, characterized in that, In step (5), the preset time step is calculated. Translation of the inner phase screen The formula is: ； in, The wind speed at each space level is projected perpendicular to the direction of light propagation. (This will be discussed later.) It is decomposed into integer pixel displacement and subpixel displacement.

7. The simulation method for infinitely long time-series laser atmospheric transmission according to claim 1, characterized in that, In step (5), for integer pixel displacement, new phase data is generated on the air inlet side. The formula is: ； in, Given the phase vector of the known template point at the edge of the current phase screen. It is a random noise vector that follows a standard normal distribution; For subpixel displacement, bilinear interpolation is used to fine-tune the updated grid, and then phase data outside the grid is removed.

8. The simulation method for infinitely long time-series laser atmospheric transmission according to claim 1, characterized in that, In step (6), the final time-series light field is obtained. Then, the intensity distribution map is obtained by squaring the complex amplitude of the light field. , The simulation results were analyzed to determine the deviation from the theoretical values, including the long-exposure spot radius. Light intensity scintillation index .

Images