To control high speed underwater vehicles, a proximity ranging system is needed to monitor the cavity thickness. In this paper, we study a time-of-flight (TOF) principle based acoustic proximity ranging system. By taking into account the acoustically hard boundary at the air-water interface, we first present a two-stage computationally efficient time delay estimation algorithm, referred to as the PEARS (Parameter Estimation for Acoustic Ranging Systems) algorithm, which is applicable to arbitrary transmitted waveforms. Numerical results based on a simulated waveform demonstrate that the PEARS estimates can approach the Cramér-Rao bound as the signal-to-noise ratio increases. We then present experiments performed by using commercially available acoustic transducers to further verify our method. To update TOF estimates quickly, a specially designed continuous wave (CW) is applied to the transducer. Experimental results show that PEARS can achieve high measurement accuracy for ranging distances less than 100 mm with an achievable parameter update rate of approximately 1.5 kHz.