Underwater sensor networks could prove invaluable for applications as diverse as monitoring fish farms, hurricane forecasting, and detecting enemy submarines. Transmitting data through liquids is much trickier than sending it through air, however. Engineers at the Massachusetts Institute of Technology have come up with a solution that could enable long-range and low-power underwater communication.
Recreating an Internet of Things below the waves is challenging because the radio signals that most wireless technologies rely on don’t travel well through water. Most underwater communication is therefore done acoustically, but generating sound waves strong enough to travel long distances through a liquid requires a lot of power.
“It’s a turning point from this being a technology that is intellectually superinteresting that we hope will work, to saying we know that this works, and we have a path to deployment.”
—Fadel Adib, MIT
Recharging or replacing a battery in a sensor deep underwater is a major hassle, says Fadel Adib, an associate professor in the Department of Electrical Engineering and Computer Science at MIT and director of the Signal Kinetics group in the MIT Media Lab. This has limited our ability to build large sensor networks in the ocean, he says, which is why his team has been working on a low-power alternative that relies on the physics of “backscattering”.
Their approach involves unpowered devices known as nodes that receive an acoustic signal from a transmitter, modulate it in a way that encodes information in the wave, and then reflect it back. Previously, they had achieved ranges of only a few meters because the sound waves were reflected in all directions, so only a small amount of the signal reached the receiver. But the team’s latest design is able to direct the reflection back at the receiver, resulting in a 15-fold boost in range and opening the door to communicating over kilometers.
“The reason why this is really exciting is because now you start opening up many of the coastal monitoring applications,” says Adib. “It’s a turning point from this being a technology that is intellectually superinteresting that we hope will work, to saying we know that this works and we have a path to deployment.”
The team first developed their backscatter communication scheme in 2019. It involved nodes that encoded data in the reflected signal by switching between absorbing sound waves or bouncing them back to the receiver. This could be used to send data in binary, with a reflection corresponding to a 1 and the lack of one denoting a 0. The modulated signal was then picked up by a hydrophone next to the transmitter.
The part of the node used to reflect the acoustic signal, known as a transducer, was made of a piezoelectric material, which generates electrical current when a mechanical force is applied to it. That makes it possible for the node to harvest energy from the incoming signals when in absorbing mode, which can be used to power the switching mechanism and potentially a sensor without the need for a battery.
One expert calls the technology—combining low-power and long-distance communication—a breakthrough with “high potential for real-world applications.”
The main limitation of the original approach, says Adib, was that there was no way to direct the reflected audio waves toward the receiver. This weakened the returning signal and therefore reduced the range over which it could operate. To solve this problem, they turned to a 70-year-old idea known as a Van Atta array, which has more recently been used to increase the range of radio-frequency identification (RFID) tags.
The researchers’ underwater signal reader includes a transmitter and hydrophone array, pictured here.MIT Signal Kinetics Lab
It involves creating an array of antennas arranged in a symmetrical pattern and then connecting opposing pairs of them using electrical wires. When a radio wave hits one of the antennas in the array, the signal is transmitted to the one it is paired with and reemitted. This means that if the signal is first received by the leftmost antenna, it is emitted by the rightmost one first. As a result, the antennas in the array emit the signal in the reverse order in which they received them, thus reflecting it back toward the source.
Translating this idea to the acoustic domain proved challenging, though, says Adib. That’s because each piezoelectric transducer resonates at a slightly different frequency, no matter how carefully you control their fabrication. When you connect two of them, these frequencies clash and significantly degrade the efficiency with which they reflect the signal. To solve this, the team added a transformer in between each pair of transducers, which helps to transfer the maximum amount of power between the transducers without impacting their resonance.
In tests of a four-by-two array conducted in the Charles River in Cambridge, Mass., the researchers showed that they could transmit data over a 300-meter round trip at 500 bits per second, which is comparable to other forms of underwater acoustic communication, using just 1.8 watts of power. The team presented their results at the ACM SIGCOMM conference in New York City on Monday.
Adib and colleagues have also created a model to test out the theoretical limits of the approach, which they have validated against their experimental data. In a paper due to be presented in Madrid at ACM MobiCom next month, they have showed that it should be possible to achieve ranges of several kilometers.
Finding a way to direct the acoustic backscatter signal back to the receiver is a breakthrough, says Aijun Song, an associate professor of electrical and computer engineering at the University of Alabama, as it significantly boosts the range of the approach. The combination of low-power and long-distance communication has “high potential for real-world applications,” he says. These could include wireless data transmission on coral-reef health from remote sensors or subsea structure-monitoring for aquaculture or oil and gas operations.
The technology could also have significant implications for navies, says Adib. The nodes could be used as beacons to create a form of underwater positioning system for drones. And low-power, long range passive sensors could have significant implications for the stealthiness of submarines, he adds.
