You can do better than complete waveform time of arrival. Sound propagation in water is dispersive (higher frequencies travel faster, but lower frequencies tra

You can do better than complete waveform time of arrival. Sound propagation in water is dispersive (higher frequencies travel faster, but lower frequencies travel farther). With an accurate model if the chirp waveform, you can determine distance more accurately by looking at the amount of signal "spreadding" and the amount of attenuation of specific signal components due to this shifting. You can do better than that, too. Theoretical accuracy of time of arrival distance approximation is directly related to the wavelength of the transmission signal. If you can model the transmitted signal and its transmission path accurately enough that you can get a strong correlation between the received signal and the modified signal (in this case taking into account Doppler effect and wave dispersion), then you can accurately measure both the time of arrival and the phase of the signal. Assuming a perfect model (HA!) the accuracy of the distance calculation is soon dominated by the various noise components (typically your timer noise, power supply noise, decimation noise from your ADC, and various thermal noise sources in both the transmission path and your measurement equipment).