阅读说明：本技术 一种模拟湍流环境uwoc系统信道特性的建模方法 (Modeling method for simulating channel characteristics of UWOC system in turbulent environment ) 是由 李岳衡 贾鹍鹏 黄平 刘陕陕 徐心雲 居美艳 于 2021-07-08 设计创作，主要内容包括：本发明公开了一种模拟湍流环境UWOC系统信道特性的建模方法,将Geldard组合散射相函数表达式中代表湍流的FF散射相函数替换为新湍流散射相函数,获得新的修正组合相函数；将新的修正组合相函数应用到蒙特卡洛UWOC系统信道特性仿真平台,获得反映各种强度湍流效应的接收光强信号及其统计特性。本发明从根本上解决了原组合相函数法只能模拟弱湍流环境的缺陷,可以快速、简单、准确地模拟“弱-中-强”三种状态下湍流信道接收信号的归一化接收光强分布,弥补了传统仿真方法的不足。(The invention discloses a modeling method for simulating the channel characteristics of a UWOC system in a turbulent flow environment, which comprises the steps of replacing an FF scattering phase function representing turbulent flow in a Geldard combined scattering phase function expression with a new turbulent flow scattering phase function to obtain a new corrected combined phase function; and applying the new modified combined phase function to a Monte Carlo UWOC system channel characteristic simulation platform to obtain received light intensity signals reflecting various intensity turbulence effects and statistical characteristics thereof. The invention fundamentally solves the defect that the original combined phase function method can only simulate the weak turbulence environment, can quickly, simply and accurately simulate the normalized received light intensity distribution of the turbulence channel received signal in three states of 'weak-medium-strong', and makes up the defects of the traditional simulation method.)

1. A modeling method for simulating the channel characteristics of a UWOC system in a turbulent environment is characterized by comprising the following steps: the method comprises the following steps:

replacing the FF scattering phase function representing turbulence in the Geldard combined scattering phase function expression with a new turbulence scattering phase function to obtain a new corrected combined phase function;

and applying the new modified combined phase function to a Monte Carlo UWOC system channel characteristic simulation platform to obtain received light intensity signals reflecting various intensity turbulence effects and statistical characteristics thereof.

2. The modeling method for simulating turbulent environment UWOC system channel characteristics as claimed in claim 1, wherein: the new modified combined phase function specific expression is as follows:

in the formula, beta_B(theta) represents a new modified combined phase function, beta_sw(θ)、β_p(theta) represents a pure seawater scattering phase function and a marine particle scattering phase function, respectively, b_sw，b_p，b_tThe values of the scattering coefficients represent the values of the scattering coefficients of pure sea water, sea particles and sea turbulence, respectively, b ═ b_sw+b_p+b_t，Representing the new turbulent scattering phase function.

3. The modeling method for simulating turbulent environment UWOC system channel characteristics as claimed in claim 2, wherein: the new turbulent scattering phase functionCalculating the SPF (epsilon, chi, theta) of the scattering phase function by using Bogucki related measured data, wherein the specific expression of the SPF (epsilon, chi, theta) is as follows:

wherein SPF (ε, χ, θ) represents a scattering phase function, ε is a kinetic energy dissipation ratio, χ is a temperature variance dissipation ratio, θ is a pitch scattering angle with respect to an incident direction, b_tFor the marine turbulence scattering coefficient, VSF (. epsilon., χ, θ) is the volume scattering function.

4. A modeling method for simulating turbulent environment UWOC system channel characteristics according to claim 3, characterized in that: the VSF (ε, χ, θ) is calculated as follows:

in the formula, V₀(epsilon, chi) represents the VSF extreme value when θ becomes 0, θ₀And (. epsilon., X) represents the half width of the scattering angle.

5. The modeling method for simulating turbulent environment UWOC system channel characteristics as claimed in claim 4, wherein: b is_tThe calculation formula is as follows:

6. a modeling method for simulating characteristics of a channel of a UWOC system in a turbulent environment according to any one of claims 3 to 5, wherein: the epsilon and the chi are selected according to Bogucki related measured data, and the Bogucki related measured data are shown in a table 2:

log₁₀ε log₁₀χ -10 -2 -10 -4 -10 -6 -8 -2 -8 -4 -8 -6 -6 -2 -6 -4 -6 -6

table 2.

7. A modeling method for simulating characteristics of a channel of a UWOC system in a turbulent environment according to any one of claims 2 to 5, wherein: when the flicker index value of the ocean turbulence is less than 0.75, the turbulence represents weak turbulence, when the flicker index value of the ocean turbulence is between 0.75 and 1, the turbulence represents medium-intensity turbulence, and when the flicker index value of the ocean turbulence is more than 1, the turbulence represents strong turbulence; when there is weak turbulence, b_tThe value range is (0, 4)]When turbulence of moderate intensity or strong turbulence, b_tThe value range is (4, + ∞).

Technical Field

The invention relates to a modeling method for simulating the channel characteristics of a UWOC system in a turbulent environment, belonging to the technical field of underwater wireless optical communication systems.

Background

Modeling of underwater wireless optical channels in marine environments can be basically generalized into two modes: the most traditional method is the most accurate method, and certainly, the method is a more difficult method, namely a specific underwater light Radiation Transmission Equation (RTE) is established based on a Maxwell equation set, and the path loss of visible light under various water environment parameter conditions can be obtained by solving the equation; obviously, the method is not only computationally intensive, but also has complex theory that is difficult for common researchers to master. The other method is based on RTE simple path solution and the absorption and scattering characteristics of seawater, and adopts Monte Carlo simulation to simulate the specific physical motion process of photons in water, so that important parameters such as motion track, irradiance, channel impulse response and the like transmitted by visible light in seawater can be obtained. The UWOC channel modeling method based on Monte Carlo simulation obtains the favor of a great number of scientific researchers once being put forward by the simple and visual physical concept and steps, is widely applied to channel characteristic simulation and acquisition of UWOC systems in various environments, and is mainly developed based on the mode by domestic and foreign researchers for underwater wireless optical channel modeling.

In computer performance simulation of UWOC systems, turbulence effects are an important factor that will have a large impact on system performance. As early as 1998, the american scholars Bogucki first compared the scattering caused by turbulence with the scattering caused by marine particles, concluding that the scattering phenomenon is dominated by turbulence in the range of scattering angles less than 0.1 °; in 2004, the influence of turbulence caused by temperature change on forward scattering was quantified through experiments, and the value of the scattering coefficient caused by the turbulence was estimated according to the experimental results. In 2018, the English scholars Geldard provides a combined phase function method considering turbulence factors to simulate a turbulence environment channel model; in this model, Geldard considers the turbulence effect as a component factor affecting scattering, and constructs a combined scattering phase function that comprehensively considers the effects of pure seawater, marine particulates, and marine turbulence.

However, after studying a turbulence simulation model based on a combined phase function method proposed by Geldard, we find that a method for embodying a turbulence effect by selecting an FF scattering phase function in the model cannot really embody the influence of turbulence, and received signal data generated by the model can only embody a weak turbulence effect, but cannot accurately embody the statistical characteristics of 'medium-strong' turbulence. This is because the FF scattering phase function represents a traditional scattering model without considering ocean turbulence scattering, and is to fit the comprehensive effect of three water qualities, namely clear ocean, coast and turbid harbor, actually measured by petzild in 1972. Since petiold does not provide measured scattering data at a small angle, the FF model cannot represent a scattering model of an optical signal in ocean turbulence. Obviously, the received signal data simulated by the model cannot fully characterize the influence of the turbulent environment on the system performance, and has very great limitation.

Disclosure of Invention

The purpose is as follows: in order to overcome the defects in the prior art, and in consideration of the fact that the real marine environment inevitably has marine turbulence and the marine turbulence mainly affects the small-angle value of the scattering phase function, the invention provides a modeling method for simulating the channel characteristics of a UWOC system in the turbulent environment, so as to reflect the correct change of the scattering phase function at a small angle in the collision process of photons and particles, thereby comprehensively and accurately simulating the normalized received light intensity distribution of the received signals of the turbulent channel in three states of weak-medium-strong.

The technical scheme is as follows: in order to solve the technical problems, the technical scheme adopted by the invention is as follows:

a modeling method for simulating the channel characteristics of a UWOC system in a turbulent environment comprises the following steps:

and replacing the FF scattering phase function representing the turbulent flow in the Geldard combined scattering phase function expression with a new turbulent flow scattering phase function to obtain a new modified combined phase function.

And applying the new modified combined phase function to a Monte Carlo UWOC system channel characteristic simulation platform to obtain received light intensity signals reflecting various intensity turbulence effects and statistical characteristics thereof.

Preferably, the new turbulent scattering phase functionCalculating the SPF (epsilon, chi, theta) of the scattering phase function by using Bogucki related measured data, wherein the specific expression of the SPF (epsilon, chi, theta) is as follows:

wherein SPF (ε, χ, θ) represents a scattering phase function, ε is a kinetic energy dissipation ratio, χ is a temperature variance dissipation ratio, θ is a pitch scattering angle with respect to an incident direction, b_tFor the marine turbulence scattering coefficient, VSF (. epsilon., χ, θ) is the volume scattering function.

As a preferred scheme, the specific expression of the new modified combined phase function is as follows:

in the formula, beta_B(theta) represents a new modified combined phase function, beta_sw(θ)、β_p(theta) represents a pure seawater scattering phase function and a marine particle scattering phase function, respectively, b_sw，b_p，b_tThe values of the scattering coefficients represent the values of the scattering coefficients of pure sea water, sea particles and sea turbulence, respectively, b ═ b_sw+b_p+b_t，Representing the new turbulent scattering phase function.

Preferably, the VSF (e, χ, θ) is calculated as follows:

VSF(ε,χ,θ)＝V₀(ε,χ)exp{-[θ/θ₀(ε,χ)]²}

in the formula, V₀When (. epsilon., χ) represents 0VSF extreme value of, theta₀And (. epsilon., X) represents the half width of the scattering angle.

Preferably, b is_tThe calculation formula is as follows:

preferably, the epsilon and the χ are selected according to Bogucki related measured data, and the Bogucki related measured data is shown in Table 2:

log10ε	log10χ
		-10	-2
-10	-4
		-10	-6
-8	-2
		-8	-4
-8	-6
		-6	-2
-6	-4
		-6	-6

table 2.

Preferably, when the flicker index value of the ocean turbulence is less than 0.75, the ocean turbulence represents weak turbulence, when the flicker index value of the ocean turbulence is between 0.75 and 1, the ocean turbulence represents medium-intensity turbulence, and when the flicker index value of the ocean turbulence is greater than 1, the ocean turbulence represents strong turbulence; when there is weak turbulence, b_tThe value range is (0, 4)]When turbulence of moderate intensity or strong turbulence, b_tThe value range is (4, + ∞).

Has the advantages that: the modeling method for simulating the channel characteristics of the UWOC system in the turbulent flow environment fundamentally overcomes the defect that the original combined phase function method can only simulate the weak turbulent flow environment, can quickly, simply and accurately simulate the normalized received light intensity distribution of the received signals of the turbulent flow channel in the three states of 'weak-medium-strong', and makes up the defects of the traditional simulation method.

Drawings

FIG. 1 is a flow chart of a Monte Carlo platform simulating turbulent ambient photon motion.

FIG. 2 is a graph of original Geldard phase function as a function of scattering angle.

FIG. 3 is a graph of the improved combination phase function presented herein as a function of scattering angle.

FIG. 4 is a graph comparing PDF curves of received light intensities for two models (b)_t1 weak turbulence log normal distribution).

FIG. 5 is a graph comparing PDF curves of received light intensities for two models (b)_t2 weak turbulence log normal distribution).

FIG. 6 is a graph comparing PDF curves of received light intensities for two models (b)_t3 weak turbulence log normal distribution).

FIG. 7 is a graph comparing PDF curves of received light intensities for two models (b)_t4 weak turbulence log normal distribution).

FIG. 8 is a graph comparing PDF curves of received light intensities for two models (b)_tNegative exponential distribution of strong turbulence in 5).

FIG. 9 is a graph comparing PDF curves of received light intensities for two models (b)_t6 strong turbulent negative exponential distribution).

Detailed Description

The present invention will be further described with reference to the following examples.

A modeling method for simulating the channel characteristics of a UWOC system in a turbulent environment comprises the following steps:

1) calculating the relevant measured data (shown in table 1) of the volume scattering function provided by the experiment by combining with the scattering phase function (shown in formulas (2), (3) and (4)), and solving a new turbulent scattering phase function based on the relevant measured data;

2) replacing the FF scattering phase function representing turbulence in the Geldard combined scattering phase function expression with a new turbulence scattering phase function based on measured data, so as to obtain a new corrected combined phase function;

3) and applying the new modified combined phase function to a Monte Carlo UWOC system channel characteristic simulation platform to obtain received light intensity signals reflecting various intensity turbulence effects and statistical characteristics thereof.

Example (b):

the specific steps and implementation flow of simulating photons through a turbulent channel by the monte carlo method are shown in fig. 1. The new modified combined phase function provided by the invention influences the change of the scattering angle theta after the photon collides with particles in water under the small-angle turbulent flow environment, and further influences the subsequent position of the photon and the absorption weight information, so that the channel characteristics of the system under the turbulent flow environment can be truly reflected.

The invention modifies the original limited FF sub-scattering phase function into a new turbulent scattering phase function calculated and derived by combining the relevant measured data with the scattering phase function, and superposes the pure seawater scattering phase function and the marine particle scattering phase function, thereby constructing and obtaining a new modified combined phase function, wherein the combined phase function is obtained based on the measured body scattering function, and can comprehensively reflect the new modified combined phase function of scattering, absorption and turbulent effects, and the specific expression is as follows:

in the formula, beta_B(theta) represents the new modified composition phase function proposed herein, where beta_sw(theta) and beta_p(theta) is consistent with the neutron scattering phase function of the Geldard model, beta_sw(θ)、β_p(theta) represents a pure seawater scattering phase function and a marine particle scattering phase function, respectively, b_sw，b_p，b_tThe values of the scattering coefficients represent the values of the scattering coefficients of pure sea water, sea particles and sea turbulence, respectively, b ═ b_sw+b_p+b_t，And representing a new turbulent scattering phase function, namely calculating the scattering phase function SPF (epsilon, chi, theta) by using Bogucki related measured data, wherein the specific expression of the scattering phase function SPF (epsilon, chi, theta) is as follows:

in the formula, SPF (e, χ, θ) represents a scattering phase function, e is a kinetic energy dissipation ratio, χ is a temperature variance dissipation ratio, and θ is a pitch scattering angle with respect to an incident direction; and the scattering coefficient b of the ocean turbulence_tIs measured by Bogucki correlation with measured data in combination with a scattering function at [0, [ pi ]]And obtaining interval integral, wherein the specific expression is as follows:

the volume scattering function VSF (e, χ, θ) is calculated by equation (4) and the Bogucki correlation data in table 1.

VSF(ε,χ,θ)＝V₀(ε,χ)exp{-[θ/θ₀(ε,χ)]²} (4)

In the formula, V₀(epsilon, chi) represents the VSF extreme value when θ becomes 0, θ₀(ε, χ) represents the half width of the scattering angle, calculated from the test data by Bogucki and is listed in Table 1.

TABLE 1 correlation of measured data for Bogucki volume scattering function with b_tCalculated value

Note: since the Bogucki original document only provides part of the test data, the values with a mark in the table are obtained by back-stepping according to equation (4).

Example (b):

1. combined scattering phase function curve contrast

In order to clearly show the composition of the sub-phase functions of the proposed new modified combined phase function model and the improvement of the original Geldard turbulence model, the invention provides graphs of the Geldard combined phase function and the new modified combined phase function as a function of the scattering angle theta, as shown in FIGS. 2 and 3, respectively. By observing fig. 2, it can be found that: in a small angle range less than 1 degree, the Geldard combined phase function curve and the FF phase function curve have almost the same change trend, and the values are not different in magnitude. Obviously, the variation trend and magnitude are unreasonable, because the FF phase function is proposed for fitting the volume scattering function data obtained by integrating three typical ocean water quality measured data of clear ocean, turbid harbor and offshore coast, which are obtained by the american scholars petzild in 1972, because the original data does not provide the volume scattering measured data less than 0.1 degree, the FF scattering phase function adopts the interpolation operation method to continue the variation trend of the measured data when the angle is close to 1 degree, that is, the logarithmic monotone variation trend of the small angle range in fig. 2. Considering that the data of the FF scattering phase function at small angles is not actually measured data, it means that the turbulence model of Geldard completely reflects the influence of turbulence at small angles on the scattering angle.

While observing fig. 3, it can be found that: the new modified combined phase function (improved combined phase function) curve is consistent with the trend of the Bogucki actually-measured turbulent phase function curve in a small angle range, and the value difference is small, while in the latter half of the curve larger than 0.1 degree, the value of the turbulent scattering phase function is close to 0, and the latter half of the improved combined phase function curve is consistent with the trend of the scattering phase function curve of the seawater particles. This is precisely reflected in the fact that the ocean turbulence disclosed by Bogucki in his 1998 paper will only have an effect on the system phase function over a small scattering angle range. Therefore, the improved turbulence channel model provided by the invention can well reflect the effect of turbulence.

2. Comparing the PDF of the simulation data of the normalized received light intensity with the flicker index

As can be seen from the foregoing description, the Geldard turbulence channel model and the improved turbulence channel model are both formed by combining three partial functions of pure seawater, ocean particles and ocean turbulence, and as the degree of ocean turbulence increases, the proportion of the pure seawater, the ocean particles and the ocean turbulence in the channel model also gradually changes. And the scattering coefficient b is along with the intensity of the turbulent flow_tThe value is gradually increased, the degree of turbulence is gradually enhanced, the proportion of the turbulence in the channel model is gradually increased, and the played function is also gradually enhanced. Therefore, when the turbulent scattering coefficient b is gradually increased_tThe effect of turbulence on the channel model is also more pronounced. To further demonstrate the rationality of the newly proposed improved turbulent channel model, we fit the turbulent scattering coefficient b_tIncreasing to 1, 2, 3, 4, 5 and 6, then obtaining probability density distribution curves under the 6 turbulence degrees by carrying out Probability Density Function (PDF) statistical characteristic simulation of received light intensity, and comparing the difference between two turbulence channel models before and after improvement. b_tThe probability density distribution curves when 1, 2, 3 and 4 are taken are shown in fig. 4, 5, 6 and 7, respectively, and the flicker index and the coefficient determining value (coefficient determining R) of the log-normal distribution fit corresponding thereto are shown in fig. 4, 5, 6 and 7, respectively²The quality of the fitting degree can be reflected,is used to measure the effect of the fit. The value ranges are generally all [0,1 ]]The closer the value is to 1, the smaller the error, which means the higher the fitting accuracy. ) As shown in table 2.

TABLE 2 comparison of scintillation index and coefficient of determination for two models in a weakly turbulent environment

In fig. 4, 5, 6 and 7, the left graph (a) represents the simulation data of the normalized received light intensity probability density function of the modified turbulence channel model and the log-normal (lognormal) distribution fitting curve thereof, and the right graph (b) represents the simulation value of the Geldard turbulence channel model and the fitting curve thereof. Observing the above 4 figures can find that: in the first three groups of graphs, the fitting effect of the lognormal distribution model is ideal, which shows that when the turbulence scattering coefficients of the Geldard turbulence model and the improved turbulence model take values of 1, 2 and 3 respectively, the received light intensity obeys lognormal distribution; however, the values of the fitting decision coefficients in table 2 can be observed, and the values of the decision coefficients of the two turbulence models are both about 0.9, which means that the error of fitting by using a lognormal distribution model is small, and the possibility that the fitting error influences a simulation conclusion is eliminated. Theoretical research shows that when the turbulent scattering coefficient value is less than or equal to 4, the flicker index values are all less than 0.75, and the error rate research of the underwater optical communication system with strong ocean turbulence is carried out through documents (J)]Signal processing, 2019(5), it is known that when the flicker index of the marine turbulence is less than 0.75, it means that the turbulence is weak, greater than 1 means strong turbulence, and between 0.75 and 1, it means moderate turbulence. And the literature (Optical Wireless Communications: System and Channel modeling with MATLAB [ M)].Taylor&Francis,2012) states that under weak turbulence, the probability density function of the received light intensity can be modeled with a lognormal distribution. However, observing FIG. 7, it can be seen that: b given by Geldard turbulence model_tThe received light intensity simulation data in the case of 4 no longer conforms to the log normal distribution, and the improved turbulence model stillApproximately obey a lognormal distribution; moreover, the simulated value of the flicker index representing the turbulence intensity obtained by the simulation data calculation of the two is far from the theoretical calculation value. This all shows that as the turbulence increases, the deviation between the simulation results of the Geldard turbulence model and the theoretical values has started to become larger, which is well understood: because the stronger the turbulence, the heavier the specific gravity of the turbulence representing the influence of the turbulence on the small-angle scattering of photons in water, and the FF phase function model is selected as the combined phase function of the Geldard model in the small-angle scattering, the influence of the turbulence on the small-angle scattering cannot be truly reflected.

When we continue to enhance the turbulence, i.e. take the turbulence scattering coefficient to 5 and 6, the theoretical calculation of the flicker index indicates that the "medium-strong" turbulence mode has been entered at this time. The received light intensity distributions of the two turbulence channel models were fitted using negative exponential distributions (which are often used to simulate the statistical properties of moderate turbulence) and the resulting simulation data and probability density fitting graphs are shown in fig. 8 and 9, respectively, while the flicker indices corresponding to the two turbulence states and the coefficient of certainty values fitted using the negative exponential distribution model are shown in table 3.

TABLE 3 comparison of scintillation indices and determining coefficients for two models under high turbulence (negative index distribution)

According to the simulation result, when the turbulent scattering coefficient b_tWhen values 5 and 6 are taken, the simulated flicker index values of the improved turbulence channel model are 0.9843 and 1.9183, respectively, which are not much different from the theoretical flicker index values, and the difference between the calculated flicker index value obtained from the Geldard turbulence model data and the theoretical flicker index value is very different. Looking again at the data in table 3 and the fitted curves in fig. 8 and 9, it can be found that: although the light intensity data generated by both turbulence models are fitted with negative exponential distributions, the data flicker index of the Geldard turbulence model deviates too much from the theoretical value, and the curve effect of fitting with negative exponential distributions is not as good as that of the improved turbulence model. This means Geldard turbulenceThe simulation model cannot accurately fit the marine environment under "medium-strong" turbulence, and the degree of fit (such as in terms of percent error of scintillation index) under conditions of weak turbulence is also inferior to the improved turbulence simulation model presented herein above. The improved turbulence simulation model provided by the invention can comprehensively and accurately simulate the normalized received light intensity distribution of the turbulence channel received signal in three states of 'weak-medium-strong'.

The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.