AMULET: Acoustic Metastructure for Direction-of-Arrival Estimation Underwater Using a Single Hydrophone
Andrew Bergey, Nakul Garg, and Akshay Gadre
In Proceedings of the 2026 ACM/IEEE International Conference on Embedded Artificial Intelligence and Sensing Systems, May 2026
Best Demo Award SenSys ’26
Small autonomous underwater vehicles (AUVs) that can perceive and explore the environment at low cost are increasingly being deployed for seafloor mapping and biome sensing. A critical sensing ability that many of the larger submarines possess is to perform directional SONAR using arrays of hydrophones to map the environment. Unfortunately, this ability to identify direction-of-arrival (DoA) remains too inaccurate or even infeasible for the small AUVs due to size, weight, and power (SWaP) constraints. This paper presents a new opportunity in augmenting this critical ability using a single hydrophone. This paper presents AMULET, a single hydrophone direction-of-arrival sensing solution in underwater scenarios which leverages acoustic metastructures to introduce directional signatures in the received signal. AMULET’s metastructure takes inspiration from sea shells with a spiral shape that uses air-as-a-metamaterial to augment directional diversity. We further demonstrate how we can extract and compare these directional signatures to estimate the angle-of-arrival across signals of various modulation schemes and different bandwidths as well as complex multipath. Our evaluation in a real-world lake deployment demonstrates a mean error of 4.1° with robustness to distance, across environments, and signal-to-noise ratios.