The goal of the proposed research is to use system identification methods to develop algorithms to estimate wind velocity profiles along the path of a laser distortion data measured at a receiver. The receiver will either be a photodetector array or wavefront sensor. A dynamic system model for the index of refraction, in which the wind profile appears as a parameter is presented. Maximum likelihood methods will be used for estimation of the wind profile. Both existing estimation algorithms, based on the Kalman filter, and the development of new computationally efficient algorithms are proposed for the identification of the resulting infinite dimensional stochastic system. The proposed research will contribute to remote sensing by introducing the use of both aerodynamic models and system identification methods into the problem of wind profile measurement. The develop of computationally efficient algorithms represents an important contribution to the identification of infinite dimensional stochastic systems.
