Accurate sound speed estimation for Unmanned Aerial Vehicle-Based Acoustic Atmospheric Tomography

Acoustic Atmospheric Tomography provides a unique way to measure temperature and wind velocity fields in a slice or volume of atmosphere. There are many applications for these measurements, including atmospheric research, weather forecasting, pollution studies, agricultural monitoring etc. The technique discussed in this paper makes use of an Unmanned Aerial Vehicle that is flown over an array of microphones on the ground. The natural sound emitted by the aircraft is recorded on-board and by the ground microphones. The speed of sound of the intersecting rays are then derived by comparing these measurements. Tomographic inversion is then used to estimate the temperature and wind fields from the sound speed measurements. This technique estimates continuous, three dimensional temperature and wind fields from ground level up to at least 1000 metres. Previous studies have indicated that the sound speed measurements must be accurate to within 0.1% in order to obtain useful results. This paper describes the signal processing techniques that have been employed on field trials data to provide estimates of the atmospheric temperature and wind fields pertaining at that time.