Kalman filter for the analysis of simulated multipulse Raman lidar water vapor profiles

Abstract

Kalman filtering theory was used to analyse simulated and real water vapor profiles. The simulation of the Raman lidar water vapor profiles was based on the third harmonic of the Nd-YAG laser. The Kalman filter was applied to both single pulse and multipulse water vapor profiles corrupted by Gaussian white noise. The results obtained by the Kalman filter approach were later compared with conventional signal averaging techniques and the smoothing filter technique. In general, the Kalman filter out-performed the conventional filters in both its smoothing and predictive capabilities. The merits and demerits of the Kalman filter when applied for the analysis of atmospheric water vapor profiles are discussed.