Synthesizing the acoustic Doppler effect in software

We present the theory and implementation of the acoustic Doppler effect in which a sequence of digital audio samples recorded at a sound source is transformed into a sequence of samples to represent the sounds perceived by an observer. The computational model alters the frequencies and amplitude of the original sequence, taking into account the simultaneous effects of a moving source, a moving observer, and a moving medium of propagation. One application is the real-time generation of the sounds of moving objects in a computer game as perceived by players of the game.