Atmospheric turbulence optical communication capacity expansion method based on entropy power loading

Abstract

The invention discloses an atmospheric turbulence optical communication capacity expansion method based on entropy power loading, which comprises the following steps of: sending a QPSK OFDM (Quadrature Phase Shift Keying Orthogonal Frequency Division Multiplexing) symbol, pre-estimating a channel pre-estimation signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity at a receiving end, and taking the channel pre-estimation signal-to-noise ratio as a calculation basis for entropy power loading in the step 2; step 2, calculating an information source entropy allocated to each subcarrier according to the channel estimation signal-to-noise ratio in the step 1, realizing fine-granularity information source entropy loading through a probabilistic shaping algorithm, then calculating power loading of each subcarrier through a water injection algorithm, and obtaining a symbol corresponding to each subcarrier; after a symbol corresponding to each subcarrier is subjected to the El Mitt symmetry and OFDM modulation, a complex signal is converted into a real signal; 4, the turbulence intensity is changed; and calculating the generalized mutual information of the received signal to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading is used for improving the optical communication capacity under different atmospheric turbulence intensities.

Description

Technical Field

[0001] The present invention relates to the field of space optical communication, and in particular to an atmospheric turbulence optical communication capacity expansion method based on entropy power loading. Background Art

[0002] Free space optics (FSO) is a promising solution to alleviate the scarcity of traditional RF spectrum due to its ultra-wide unlicensed spectrum range. However, atmospheric turbulence is a major challenge that can cause severe degradation of the system's bit error rate performance, thereby undermining the advantage of FSO's high communication capacity.

[0003] Orthogonal frequency division multiplexing (OFDM) has high coding gain and spectrum efficiency under atmospheric turbulence. Existing OFDM-FSO systems lack a mechanism that can dynamically estimate turbulence intensity, so they lack flexibility in dynamic turbulence, resulting in reduced communication capacity and waste of spectrum resources. In addition, existing studies have tested an actual 1.87km long-distance OFDM-FSO link. The spectrum of the actual link is frequency selective. If the power allocation of each subcarrier remains consistent, power will be wasted and communication capacity cannot be maximized. Traditional bit power loading can partially solve the above problems, but due to the lack of fine-grained adjustment of source entropy, its communication capacity performance still needs to be improved. Summary of the invention

[0004] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is the reduction of communication capacity and waste of spectrum resources and power caused by turbulence in atmospheric communication. The present invention provides an atmospheric turbulence optical communication capacity expansion method based on entropy power loading, which is used to improve the optical communication capacity under different atmospheric turbulence intensities. The method obtains channel information by pre-estimating the signal-to-noise ratio of the channel under different turbulence intensities, and then distributes the transmission entropy and power to each OFDM subcarrier through entropy power loading, so as to achieve the purpose of improving communication capacity.

[0005] To achieve the above object, the present invention provides an atmospheric turbulence optical communication capacity expansion method based on entropy power loading, which is characterized by comprising the following steps:

[0006] Step 1: Send QPSK OFDM symbols, and estimate the channel estimated signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity at the receiving end, which serves as the basis for calculating the entropy power loading in step 2;

[0007] Step 2: Entropy power loading includes entropy loading and power loading. According to the channel estimated signal-to-noise ratio in step 1, the source entropy allocated to each subcarrier is calculated, and the fine-grained source entropy loading is realized through the probability shaping algorithm. Then, the power loading of each subcarrier is calculated through the water injection algorithm to obtain the symbol corresponding to each subcarrier.

[0008] Step 3: After the symbol corresponding to each subcarrier is Hermitian symmetric and OFDM modulated, the complex signal is converted into a real signal;

[0009] Step 4: Change the turbulence intensity;

[0010] Step 5: Calculate the generalized mutual information of the received signal to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence.

[0011] Furthermore, the OFDM format includes the number of IFFT points, the number of data subcarriers, a cyclic prefix and a training sequence.

[0012] Furthermore, QPSK OFDM symbols are sent, and the channel estimated signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity is estimated at the receiving end as a basis for calculating the entropy power loading in step two. Specifically, the transmitting end modulates the information sequence into a QPSK OFDM signal and sends it. After the QPSK OFDM signal is transmitted through the communication link, it is demodulated by OFDM at the receiving end to obtain the transmitted QPSK constellation point, and the error vector magnitude EVM of the receiving constellation point is estimated. According to the relationship between EVM and the signal-to-noise ratio SNR, the signal-to-noise ratio SNR of each subcarrier is further estimated.

[0013] Furthermore, the entropy loading specifically includes dividing different signal-to-noise ratio intervals based on the principle of maximizing the generalized mutual information at the receiving end, determining the QAM order corresponding to each subcarrier according to the channel estimated signal-to-noise ratio of each subcarrier, and realizing the source entropy distribution of each subcarrier through fine-grained adjustment of the source entropy through probability shaping; then, the power allocated to each subcarrier is calculated through an ideal water injection algorithm to achieve the purpose of maximizing the overall generalized mutual information.

[0014] Furthermore, the Monte Carlo method is first used to determine the QAM order with the maximum generalized mutual information at each signal-to-noise ratio; after the order is determined, the golden section search method is used to determine the probability shaping coefficient λ that maximizes the generalized mutual information at the specific estimated signal-to-noise ratio and the specific QAM order.

[0015] Furthermore, firstly, through Monte Carlo simulation, the SNR range is from -15dB to 35dB, with a step size of 0.5dB, to simulate the generalized mutual information GMI performance of each SNR at each order in AWGN, and form a lookup table;

[0016] Then, based on the principle of maximizing GMI, for the estimated SNR of each subcarrier in step 1, determine the modulation order of the subcarrier and the range of the shaping coefficient λ in the lookup table, and again perform Monte Carlo simulation to obtain the current order, current SNR, and the precise shaping coefficient λ that maximizes the GMI of the subcarrier and record it.

[0017] Furthermore, when λ=0, the Maxwell-Boltzmann distribution degenerates into a uniform distribution, at which time the transmission entropy is the largest; as λ increases, the transmission entropy decreases.

[0018] Furthermore, when the turbulence intensity is changed, the Fried parameters of turbulences of different intensities are different, resulting in different phase screens; when the turbulence intensity is stronger, the number of phase screens in the group that cause increased optical power attenuation is greater.

[0019] Furthermore, the generalized mutual information of the received signal is calculated to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence. The calculation of the generalized mutual information is as follows:

[0020]

[0021]

[0022] Where H is the source entropy of the transmitted symbol, G is the entropy loss during the transmission process, M is the order of QAM, and σ 2 is the noise variance of the AWGN channel, P xm is the mth constellation point x m The probability of receiving n is the nth symbol, b n,k is the bit value (1 or 0) at the kth position of the nth symbol, and X(k,b) is the constellation point set in step 2 A subset of .

[0023] Technical Effects

[0024] The present invention provides an atmospheric turbulence optical communication capacity expansion method based on entropy power loading, which can reflect different degrees of capacity expansion effects for optical power attenuation caused by different turbulence intensities. In addition, due to the capacity improvement brought by probability shaping and water injection algorithms to the communication system and the fine-grained adjustment of the source entropy, the communication capacity is closer to the Shannon limit.

[0025] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1It is a flow chart of a method for expanding atmospheric turbulence optical communication capacity based on entropy power loading according to a preferred embodiment of the present invention;

[0027] Figure 2 It is a structural schematic diagram of an atmospheric turbulent optical communication system of an atmospheric turbulent optical communication capacity expansion method based on entropy power loading in a preferred embodiment of the present invention;

[0028] Figure 3 It is a schematic diagram of the comparison of mutual information received under a certain phase screen between a modulation mode of an atmospheric turbulence optical communication capacity expansion method based on entropy power loading and other traditional modulation modes in a preferred embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram comparing the mutual information received under different turbulence intensities of a modulation method of an atmospheric turbulence optical communication expansion method based on entropy power loading and other traditional modulation methods in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0031] In the following description, specific details such as specific internal procedures and techniques are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0032] like Figure 1 As shown, an embodiment of the present invention provides an atmospheric turbulence optical communication capacity expansion method based on entropy power loading, and the specific steps are as follows:

[0033] Step 1: Send QPSK OFDM symbols and estimate the signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity at the receiving end, which serves as the basis for calculating the entropy power loading in step 2.

[0034] The transmitter modulates the information sequence into a QPSK OFDM signal. The OFDM format (number of IFFT points, number of data subcarriers, cyclic prefix, training sequence) is consistent with the subsequent step 3. The signal is transmitted through the communication link and demodulated by OFDM at the receiving end to obtain the transmitted QPSK constellation point, estimate the received constellation point (error vector magnitude) EVM, and further estimate the SNR of each subcarrier based on the relationship between EVM and (signal-to-noise ratio) SNR. The details are as follows:

[0035] First, the receiving constellation points are normalized in scale with reference to the standard transmitting constellation points based on the principle of minimum mean square error (MMSE), and then the EVM between the receiving constellation points and the transmitting constellation points is calculated. The EVM formula is as follows:

[0036]

[0037] R represents the receiving constellation point, and S represents the transmitting constellation point. For QPSK, both R and S are complex numbers in mathematics. Therefore, what is calculated here is the square of the vector distance, and the square root represents the vector error. 2 It represents the energy of the transmitted constellation point, and the square root represents the amplitude. Since EVM is usually expressed as a percentage, the end of the formula needs to be multiplied by 100. In summary, EVM represents the ratio of the vector error of the receiving constellation point relative to the transmitting constellation point to the amplitude of the transmitting constellation point, and its size measures the degree of deviation of the transmission system constellation point. The relationship between EVM and SNR is as follows:

[0038] SNR = -20 × log 10 (EVM)dB

[0039] Since both SNR and EVM represent the error relationship between the received signal and the transmitted signal, for a Gaussian white noise channel (AWGN), the SNR and EVM between the transmitted and received symbols of QPSK OFDM conform to the quantitative relationship in the above formula, and SNR is usually expressed in dB.

[0040] Step 2: Entropy power loading Calculate the source entropy and power to be loaded for each subcarrier. The details are as follows:

[0041] First, through Monte Carlo simulation, the SNR range is from -15dB to 35dB, with a step size of 0.5dB, and the generalized mutual information (GMI) performance of each SNR at each order in AWGN is simulated to form a lookup table. Then, based on the principle of maximum GMI, the modulation order of the subcarrier and the range of the shaping coefficient λ in the lookup table are determined for the estimated SNR of each subcarrier in step 1, and the current order and current SNR are obtained through Monte Carlo simulation again. The precise shaping coefficient λ that maximizes the GMI of the subcarrier is recorded. The change of λ determines the probability of each transmission constellation point, and then determines the value of the transmission entropy. The mechanism is as follows:

[0042] Given a QAM symbol set A constellation point The probability of is usually generated according to the Maxwell-Boltzmann (MB) distribution

[0043]

[0044] λ refers to the probability shaping coefficient. When λ = 0, the MB distribution degenerates into a uniform distribution, and the transmission entropy is maximum. As λ increases, the transmission entropy decreases.

[0045] Then, the water injection algorithm is used to calculate the normalized energy allocated to each subcarrier to maximize the overall GMI. The principle of the water injection algorithm is as follows:

[0046] Under the condition of limited total power and uneven signal-to-noise ratio, the problem of solving the maximum capacity can be described as

[0047]

[0048] Among them, p i and SNR i are the normalized power allocation and signal-to-noise ratio of the ith subcarrier, respectively. This mathematical problem can be solved by Lagrange number multiplication to obtain the water injection constant c. Then the power allocation of each subcarrier is p i =c-1 / SNR i The ideal water filling algorithm calculates the loading power used by the entropy power loading method.

[0049] Step 3: After the signal of each subcarrier is Hermitian symmetric and OFDM modulated, the complex signal is converted into a real signal, and the system belongs to intensity modulation direct detection.

[0050] The information sequence is modulated into a constellation point sequence using the shaping coefficient λ and energy of each subcarrier obtained in step 2, and then subjected to Hermitian symmetry and OFDM modulation. The core of OFDM modulation is IFFT, and the basic operation of IFFT is IDFT. The principle is as follows:

[0051]

[0052] Among them, N is the length of the input constellation point sequence, and k is the serial number of the constellation point. If X(k) is conjugate symmetric about the middle point, then x(n) is a real number, and this symmetry is called Hermitian symmetry. That is, after the input constellation point sequence undergoes the Hermitian symmetry operation, the data on the left and right halves of the subcarrier are completely symmetrical, so only half of the subcarrier can be used to transmit data, but the data at this time becomes a real number, that is, no phase information is required, and the receiver can directly restore the constellation point through the amplitude. This low-complexity system is called an intensity modulation direct detection system. After IFFT, it is further modulated into an OFDM signal, where the OFDM format remains the same as step one.

[0053] Step 4: Change the turbulence intensity. Since the corresponding received optical power is different, the optical communication performance is also different. The phase screen determines the attenuation of the optical signal, and thus determines the intensity of the turbulence. The phase screen is generated by the following formula.

[0054]

[0055] Among them, f xn and f ym is the discrete x and y spatial frequency, c n,m are the Fourier series coefficients. c n,m The complex Gaussian statistic follows a zero mean and variance of

[0056]

[0057]

[0058] Among them, Φ Φ (f) is the modified von Karman atmospheric phase power spectral density. f is f xn and f ym Polar diameter after conversion to polar coordinates. L x and L y are the grid lengths in the x and y directions respectively, r0 is the Fried parameter. f0=1 / L0 and f m =5.92 / (2π×l0), where L0 and l0 are the outer and inner scales of turbulence respectively.

[0059] Different intensities of turbulence have different r0, which leads to different phase screens in the experiment, and thus can simulate different intensities of atmospheric turbulence. The stronger the turbulence intensity, the more phase screens in the group that cause increased optical power attenuation, which is statistically manifested as a stronger turbulence intensity.

[0060] Figure 2This is the system block diagram of the entire link, which includes the DSP processing process at the transceiver end, the optical devices involved in the link, and the spatial light modulator (SLM) used for turbulence simulation. The turbulence simulation and intensity change in step 4 are achieved by loading different random phase screens on the SLM.

[0061] Step 5: Calculate the generalized mutual information of the received signal to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence.

[0062] The calculation of generalized mutual information is as follows

[0063]

[0064]

[0065] Where H is the source entropy of the transmitted symbol, G is the entropy loss during the transmission process, M is the order of QAM, and σ 2 is the noise variance of the AWGN channel. xm is the mth constellation point x m The probability of receiving n is the nth symbol. n,k is the bit value (1 or 0) at the kth position of the nth symbol, so X(k,b) is the constellation point set in step 2 GMI measures the amount of information transferred from the information sequence sent by the transmitter to the received sequence parsed by the receiver, objectively and effectively evaluating the performance of the communication system.

[0066] Figure 3 The comparison of the received GMI of entropy power loading, traditional bit power loading, and fixed-order 64QAM under a certain phase screen is shown. The entropy power loading method has the best effect and is closest to the Shannon limit.

[0067] Figure 4 The GMI performances of 50 phase screens under (a) strong turbulence and (b) weak turbulence are statistically analyzed, and the average values ​​are displayed. The entropy power loading method shows a greater advantage.

[0068] The method of the present invention will be described below with a specific example.

[0069] QPSK OFDM symbols are sent, and then the EVM of the QPSK constellation points is used at the receiver to estimate the SNR. In this embodiment, 256 subcarriers are used for OFDM modulation. The Hermitian symmetric operation makes the data real after IFFT, so only half of the subcarriers can be used for data transmission, that is, 128 data subcarriers. Some subcarriers with extremely poor signal-to-noise ratio performance are removed, and finally 123 subcarriers are used to load data. Then a cyclic prefix (CP) of length 16 is inserted to mitigate the impact of ISI.

[0070] Entropy power loading uses the signal-to-noise ratio to calculate the source entropy and power allocation.

[0071] The transmitting end first converts the data to be sent from parallel to serial (S / P) and divides it into several data streams, each of which represents the data of one subcarrier. CCDM coding is used to map the data sequence into a QAM sequence according to the obtained shaping factor λ and allocate the power calculated in the previous step. Then OFDM modulation is performed, and then a parallel-to-serial (P / S) operation is performed to convert the data into a data stream that can actually be sent.

[0072] The optical signal is transmitted to the SLM through the transmitting collimator, then reflected back to the receiving collimator and coupled to the single-mode fiber. A variable optical attenuator is used to adjust the received optical power. The received optical signal is then converted into an electrical signal using an avalanche photodiode APD and recorded using an oscilloscope at a sampling rate of 80Gsa / s. A three-ring polarization controller and a polarizer are used to control the polarization of the optical signal to meet the polarization requirements of the SLM (the polarization of the light must be parallel to the long axis of the SLM). A computer is used to control the switching of the phase screen on the SLM. The SLM is loaded with strong turbulence (r0 = 0.8 mm) and weak turbulence (r0 = 3 mm), and its L0 = 10 m, l0 = 0.1 mm.

[0073] The data recorded by the oscilloscope is processed by the DSP on the receiver. First, synchronization is performed and the start position of the data frame is found using the training sequence cross-correlation algorithm. After the serial-to-parallel operation, CP removal and FFT, the training sequence is used for channel estimation and equalization and then removed. Then the QAM demapping operation is performed to recover the data corresponding to the 123 subcarriers. By comparing the received data with the transmitted data, the GMI can be obtained. The result is shown in Figure 4 shown.

[0074] Entropy power loading should be used on OFDM-FSO systems. It combines entropy loading and water injection algorithms. It is an effective adaptive bit and power loading strategy and has been proven to improve the capacity of OFDM systems. It is generally believed that turbulence remains roughly stable on a time base close to the coherence time (commonly known as the Greenwood frequency). In general, the duration of a data frame is much shorter than the turbulence freezing time, so the channel information required for entropy power loading can be effectively obtained through channel estimation. In addition, the core technology of entropy loading, probability shaping, can fine-grainedly adjust the source entropy, while the water injection algorithm allocates power more reasonably, both of which further improve the system communication capacity. Therefore, entropy power loading is an effective way to improve communication capacity in FSO under atmospheric turbulence. In an embodiment of the present invention, the entropy power loading used obtains an average generalized mutual information (GMI) gain of 0.35 bits / symbol and 0.47 bits / symbol under strong turbulence and weak turbulence, respectively, compared with traditional bit power loading, and an average GMI gain of 0.45 bits / symbol and 0.55 bits / symbol under strong turbulence and weak turbulence, respectively, compared with fixed-order 64QAM modulation.

[0075] The present invention estimates the atmospheric turbulence channel to obtain an estimated signal-to-noise ratio, calculates the transmission entropy and power allocated to each subcarrier under different turbulence intensities through entropy power loading, modulates the signal processed by the transmitter onto light through optical modulation, passes through the atmospheric turbulence channel and is received at the receiving end, and calculates mutual information. The method proposed by the present invention is simple in actual operation and has excellent effects, and has a good capacity improvement for atmospheric turbulence optical communications.

[0076] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for expanding atmospheric turbulence optical communication capacity based on entropy power loading, characterized in that: The following steps are involved: Step 1: Send QPSK OFDM symbols, and estimate the channel estimated signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity at the receiving end, which serves as the basis for calculating the entropy power loading in step 2; Step 2: Entropy power loading includes entropy loading and power loading. According to the channel estimated signal-to-noise ratio in step 1, the source entropy allocated to each subcarrier is calculated, and the fine-grained source entropy loading is realized through the probability shaping algorithm. Then, the power loading of each subcarrier is calculated through the water injection algorithm to obtain the symbol corresponding to each subcarrier. Step 3: After the symbol corresponding to each subcarrier is Hermitian symmetric and OFDM modulated, the complex signal is converted into a real signal; Step 4: Change the turbulence intensity; Step 5: Calculate the generalized mutual information of the received signal to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence.

2. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 1, characterized in that: The OFDM format includes the number of IFFT points, the number of data subcarriers, a cyclic prefix and a training sequence. The OFDM format is consistent with that in the subsequent step three.

3. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 4, characterized in that: Send QPSK OFDM symbols, and estimate the channel estimated signal-to-noise ratio of each OFDM subcarrier under each turbulence intensity at the receiving end, which serves as the calculation basis for the entropy power loading in step 2. Specifically, the sending end modulates the information sequence into a QPSK OFDM signal and sends it. After the QPSK OFDM signal is transmitted through the communication link, it is demodulated by OFDM at the receiving end to obtain the transmitted QPSK constellation point, estimate the error vector magnitude EVM of the receiving constellation point, and further estimate the signal-to-noise ratio SNR of each subcarrier based on the relationship between EVM and the signal-to-noise ratio SNR.

4. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 1, characterized in that: The entropy loading specifically includes dividing different signal-to-noise ratio intervals based on the principle of maximizing the generalized mutual information at the receiving end, determining the QAM order corresponding to each subcarrier according to the channel estimated signal-to-noise ratio of each subcarrier, and realizing the source entropy distribution of each subcarrier through fine-grained adjustment of the source entropy through probability shaping; then calculating the power allocated to each subcarrier through an ideal water injection algorithm to achieve the purpose of maximizing the overall generalized mutual information.

5. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 4, characterized in that: First, the Monte Carlo method is used to determine the QAM order with the maximum generalized mutual information at each signal-to-noise ratio. After the order is determined, the golden section search method is used to determine the probability shaping coefficient λ that maximizes the generalized mutual information at the specific estimated signal-to-noise ratio and the specific QAM order.

6. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 5, characterized in that: First, through Monte Carlo simulation, the SNR range is from -15dB to 35dB, with a step size of 0.5dB, to simulate the generalized mutual information GMI performance of each SNR at each order in AWGN, and form a lookup table; Then, based on the principle of maximizing GMI, for the estimated SNR of each subcarrier in step 1, determine the modulation order of the subcarrier and the range of the shaping coefficient λ in the lookup table, and again perform Monte Carlo simulation to obtain the current order, current SNR, and the precise shaping coefficient λ that maximizes the GMI of the subcarrier and record it.

7. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 1, characterized in that: When λ=0, the Maxwell-Boltzmann distribution degenerates into a uniform distribution, at which time the transmission entropy is the largest; as λ increases, the transmission entropy decreases.

8. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 1, characterized in that: When the turbulence intensity is changed, the Fried parameters of turbulences of different intensities are different, resulting in different phase screens; when the turbulence intensity is stronger, the number of phase screens in the group that cause increased optical power attenuation is greater.

9. The atmospheric turbulence optical communication capacity expansion method based on entropy power loading as claimed in claim 1, characterized in that: The generalized mutual information of the received signal is calculated to evaluate the capacity expansion effect of entropy power loading on optical communication under atmospheric turbulence. The calculation of generalized mutual information is as follows: Where H is the source entropy of the transmitted symbol, G is the entropy loss during the transmission process, M is the order of QAM, and σ 2 is the noise variance of the AWGN channel, P xm is the mth constellation point x m The probability of receiving n is the nth symbol, b n,k is the bit value 1 or 0 at the kth position of the nth symbol, and X(k,b) is the constellation point set in step 2 A subset of .

Images