IceSound: Acoustic-based Ice Thickness Sensing with Smartphones

Accurate measurement of ice thickness is essential for safety in ice-related human activities; however, existing solutions are often cumbersome or expensive. We present IceSound, a novel system that leverages the ubiquity of smartphones to realize portable, cost-effective, and real-time ice thickness sensing. IceSound tackles multiple challenges including overlapping echoes due to the higher sound speed in ice (approximately 3250 m/s), weak bottom reflections due to attenuation in thick ice, and variations of temperature and device orientation during measurements. Extensive field experiments show that IceSound achieves a mean absolute error of 0.3 cm for ice thickness measurement when the ice thickness is in the critical range of 10-15 cm. When assessing whether the ice thickness is safe for human activities based on the well-recognized 15 cm threshold, the system exhibits a remarkably low error rate of 0.13%, demonstrating its effectiveness in enhancing safety for ice-related human activities.