Atmospheric optical communications channel estimation employing superimposed training sequence under sand-dust weather conditions

The inherent advantages of superimposed training method are restricted by superposition data information, power allocation, and direct current bias in channel estimation of atmospheric optical communications under sand-dust weather. A novel scheme was proposed to perform these issues, especially for the channel with sand-dust particles. In the proposal, the data-dependent superimposed training algorithm was utilized to mitigate the influence of data information, the correlation matching algorithm was utilized for direct current bias elimination, and the maximum output signal-to-noise ratio was utilized to perform the optimal power allocation factor. The performance of mean square error, power allocation factor, bit error rate and algorithm complexity were numerically evaluated. The results demonstrate that the proposed method has a better performance than conventional methods with a slightly increased computational complexity.