A measurement device that is extremely important for Unmanned Aerial Vehicle (UAV) guidance and control purposes is the airspeed sensor. As the parameters of feedback control laws are conventionally scheduled as a function of airspeed, an incorrect reading (e.g. due to a sensor fault) of the Pitot-static tube could induce an incorrect feedback control action, potentially leading to the loss of control of the UAV. The objective of this study is to establish the accuracy and reliability of the two airspeed estimation techniques for eventual use as the basis for real-time fault detection of anomalies occurring on the Pitot-static tube sensor. The first approach is based on an Extended Kalman Filter (EKF) and the second approach is based on Least Squares (LS) modeling. The EKF technique utilizes nonlinear kinematic relations between GPS, Inertial Measurement Unit and Air Data System signals and has the advantage of independence from knowledge of the aircraft model. The LS method is based on explicit knowledge of the aircraft model and has the advantage of on-line computation of the airspeed estimate, with minimal computational effort. The performance analysis was carried out with flight data from the WVU YF-22 UAV research platform. The results of the analysis indicate that the two methods provide essentially comparable performance in terms of mean (∼1 m/s) and standard deviation (∼1.5 m/s) of the airspeed estimation error which is about the 5% of the mean in-flight velocity of 32 m/s.