用于反演云层高度的记忆式滑动窗口积分算法

 记忆式滑动窗口积分算法选取一定宽度的窗口，从地面开始滑动，对窗口内的信号进行积分，根据积分值和前一时刻的云层信息来反演云层高度。该算法可消除信号中的噪声，叠加放大信号中的云层信息，将云层与气溶胶加以区分。用该算法处理了Vaisala云高仪的数据，并与云高仪的结果进行了对比，结果表明，记忆式滑动窗口积分算法在反演尝云层信息信息时具有较高的准确性和可靠性。

Memorable glide window integral algorithm for retrieving cloud height

The algorithm of retrieving cloud height was studied. The intensity and the pulse width of the cloud backscatter is different from the atmospheric aerosol and the random noise. The cloud height could be determined by analyzing lidar returns. A selected window gliding from the ground and the integral of signals in the window is calculated. The random noise could be eliminated and the cloud information could be distinguished from aerosol by window integral. Referred to previous cloud information, the current cloud height is retrieved according to the window integral. Compared with Vaisala ceilometer, the memorable glide window integral algorithm could be used to retrieve the information of cloud exactly and reliably.