The structure of the system is aimed at improving the spot detection algorithm, in order to improve the level of an atmospheric wireless laser communication link alignment accuracy by reducing the impact of turbulence on laser communication. First,we calculate the changes in the local distortion to determine the real-time local decision threshold Tn by a Hartmann-Shack wave-front sensor on the local distortion of the results of real-time detection. We then obtain the sequence of Mean Square Error(MSE) from the sampling information M(Xi，Yi) by a statistical series. At the same time, we can determine the sampling space (Xi，Yi)of the beacon based on the local threshold. Second,we use the center of a circle for fitting to make the MSE a minimum by following the method of least squares. We also analyse the advantages and disadvantages of the method with center-fitting of spot detection based on the sequence information, from the perspective of information theory. The results show that the detection accuracy is better than 4 μrad in the turbulence and more than twice the traditional centroid algorithm. The method avoids a problem with the introduction of strong turbulence noise. It basically meet the requirements of Air-Wireless-Laser alignment.