NEWT The Internet of Underwater Things

A NEWT Lab research thrust, 20 kHz–1 MHz

The Internet of Underwater Things

Most of the planet has no internet, not because nobody needs one down there, but because radio drowns. Underwater, sound is the only signal that goes the distance. We are teaching it to carry more.

Seawater swallows a WiFi signal within about a meter and blue-green light within about a hundred. Sound crosses kilometers, which is why the ocean's native network is acoustic: slow, narrow, echo-ridden, and, until recently, deaf to direction without a boatload of hardware.

Direction from a single ear

Knowing where a sound comes from usually takes an array. AMULET does it with one hydrophone and a piece of shaped plastic.

Drag the sound source. Black is truth; teal is what the receiver believes. The inset shows what the sensor actually hears.

A single hydrophone is an ear without a head: it hears the source clearly but has no idea which way to turn. Every direction sounds the same, so its belief is a whole ring of bearings.

The classical fix is an array: many hydrophones, a meter of aperture, and a channel of electronics for each, comparing arrival times across elements. It works, but a straight array cannot tell left from its own mirror image: two bearings produce identical phase ramps, and one of them is a lie. And every added element costs power, money, and drag on whatever robot has to carry it.

AMULET takes the opposite path: instead of adding ears, it shapes the sound. A passive, 3D-printed acoustic metastructure wrapped around a single hydrophone imprints a direction-dependent fingerprint onto the spectrum of anything it hears; watch the inset repaint as you drag. One sensor, one channel, zero added power, full 360° bearing. In a real lake deployment, AMULET estimates direction with a mean error of 4.1°, robust across distances, environments, and signal-to-noise ratios. That is the demo that won Best Demo Award at ACM SenSys 2026.

The thrust

The last unconnected frontier. The instruments are already in the water: sensors on moorings, autonomous vehicles, aquaculture pens, and tidal and wave-energy infrastructure. But each one is effectively offline, surfacing data by ship visit or satellite buoy. An Internet of Underwater Things needs links, localization, and low-power sensing that survive the physics of the sea.

Intelligence in hardware, at zero power. AMULET's bet is that the scarcest resource underwater is energy, so computation should move into the physical structure itself. A metastructure that encodes bearing into a spectrum is doing signal processing for free, before the electronics ever wake up. The same philosophy of letting shaped matter do the math extends to filtering, multiplexing, and identification.

From the bench to the Sound. Puget Sound is our test tank. The lab also mentors the PNNL "IoT of the Sea" ENGINE capstone team, building underwater acoustic communication for Pacific Northwest National Laboratory: a pipeline from classroom to deployed ocean systems.

Same toolbox, different fluids: the lab's acoustic methods also sense through the human body (UltraPoser) and through cardboard and enamel (MiLTOn, ToMoBrush). Sound goes where radio can't, above the waterline too.

Publications in this thrust

Lab members in bold. The full lab record lives on the main page.

Demo: AMULET: Acoustic Metastructure for Direction-of-Arrival Estimation Underwater using a Single Hydrophone

Andrew Bergey, Nakul Garg, Akshay Gadre

ACM SenSys 2026 Best Demo Award

The same toolbox, out of the water

Acoustic Signature Management Engine in an Object Integrity Sensing System

Deepak Vasisht, Ranveer Chandra, Manikanta Kotaru, Bodhi Priyantha, Nikunj Raghuvanshi, Akshay Gadre

US Patent 12,650,409, granted June 2026
PNNL

IoT of the Sea: mentoring a UW ECE ENGINE capstone team building underwater acoustic communication for Pacific Northwest National Laboratory.

Capstone partnership, 2025

Recognition

Best Demo Award at ACM SenSys 2026 for AMULET, and the Best Presentation Award at the ACM/IEEE CPS-IoT Week PhD Forum 2026, both to Andrew Bergey.

The underwater thrust, recognized

Work on the Internet of Underwater Things PhD applicants and UW students: the lab is recruiting.