GPS (Global Positioning System) is a system used for navigation and positioning with the help of 32 satellites that orbit the earth. GPS satellites emit radio waves to activate the GPS receiver on the Earth to determine the exact location, speed and time for all weather conditions. Currently, GPS has a vital role in navigation drone. However GPS have a lot of noise, it is necessary to overcome the noise filter on the GPS. In this research conducted a simulation of adaptive algorithm in extracting data from the GPS sensor. The method in this research is the Kalman Filter. From the identification and simulation can be concluded that the adaptive algorithm is designed to work well and need to be implemented on the downlink system. The simulation results showed that the error of algorithm has convergent properties are increasingly shrinking.