In Free Space Optical (FSO) communication, atmospheric turbulence can cause a decrease in communication link performance.

In this paper, a polar coding channel coding scheme is proposed for this problem, and Monte Carlo algorithm is used to construct polar codes for Gamma-Gamma distribution atmospheric turbulence channels commonly used in FSO communication. By comparing the Block Error Rates (BER) of the proposed polar coding scheme, uncoded transmission, and Low Density Parity Check (LDPC) codes with similar code lengths under Gamma-Gamma distribution atmospheric turbulence channels.

The simulation results show that when the Signal-to-Noise Ratio (SNR) is greater than 6.7 dB in weak turbulence, 10.3 dB in moderate turbulence, and 11.5 dB in strong turbulence, the proposed polar coding scheme performs better than uncoded transmission. Compared with LDPC codes, under weak turbulence conditions, when the SNR is greater than 7.1 dB, the overall performance of the proposed polar coding scheme is better than that of LDPC codes. Under moderate turbulence conditions, when the SNR is greater than 10.6 dB, the overall performance of the proposed polar coding scheme is better than that of LDPC codes. Under strong turbulence conditions, when the SNR is greater than 12 dB, the overall performance of the proposed polar coding scheme is better than that of LDPC codes. In addition, it is found that when the code length is short, the code rate is low, and the decoding width is small, the proposed polar coding scheme has better performance.

The use of the proposed polar coding scheme effectively improves the atmospheric turbulence effect in FSO communication systems. Moreover, in the harsh atmospheric turbulence channel environment of moderate and high turbulence intensities, the proposed polar coding scheme exhibits more obvious advantages than LDPC codes. This indicates that polar codes have good development prospects in FSO communication.