Underwater Acoustic Source Signal Detection Using Terahertz Radar Sensors

Imagine trying to listen to a secret conversation happening deep underwater while you are standing on a boat above. It's a bit like trying to hear a whisper through a thick, soundproof wall; the water blocks the sound from escaping into the air, and the air blocks the sound from getting back in. This is a huge problem for scientists and engineers who want to build a "Space-Air-Ground-Sea" network, a super-connected system where satellites, planes, and submarines can all talk to each other. Usually, if you want to find a submarine, you have to send a submarine out to listen, because sound waves bounce right off the surface of the ocean like a ball hitting a trampoline, and radio waves (like Wi-Fi) just die out the moment they hit the water.

But nature has a sneaky little trick. When a sound source vibrates underwater, it doesn't just stay hidden; it actually pushes and pulls on the water's surface, creating tiny, tiny ripples. These aren't the big waves you see from a boat; they are microscopic shivers, smaller than the width of a human hair, that dance to the rhythm of the underwater sound. If you could see these invisible shivers, you could figure out what's happening underwater without ever getting wet. The challenge is that these shivers are so small and the ocean is so noisy with wind and big waves that spotting them is like trying to hear a single pin drop in the middle of a rock concert.

This paper tells the story of a team of researchers who built a super-precise "ear" for the air that can actually hear these microscopic shivers. They didn't use a microphone or a camera; they used a special kind of radar that operates at terahertz frequencies—a type of light that sits between microwaves and infrared. Think of this radar as a super-accurate ruler that can measure distance changes smaller than a grain of sand. The team set up experiments to see if they could use this radar to detect underwater sounds by watching the water surface dance. They found that while the ocean is messy and loud, their special radar and a clever new math trick could filter out the noise and spot the tiny, rhythmic shivers caused by underwater sounds, even when the sea was choppy.

The Invisible Dance and the Super-Ruler

The researchers started by asking a simple question: If an underwater speaker makes a sound, does the water surface actually move? The answer is yes, but the movement is incredibly small. To understand how they caught this movement, imagine the ocean surface as a giant, bouncy trampoline. If you jump on it, the whole thing bounces up and down. But if someone whispers a secret rhythm from underneath the trampoline, the fabric might wiggle just a tiny bit in time with that whisper. The problem is that the wind is blowing the trampoline around wildly, and the "whisper" wiggle is so small it gets lost in the chaos.

To catch this, the team built a custom radar system. Most radars are like flashlights that bounce light off objects to see how far away they are. But this radar is like a laser pointer that is so steady and precise it can measure if the trampoline fabric moved a fraction of a hair's width. They used a frequency of 122.5 GHz, which is in the terahertz range. Why so high? Because the higher the frequency, the more sensitive the radar is to tiny movements. It's like using a microscope instead of a magnifying glass; the higher the frequency, the smaller the detail you can see.

The Noise Problem and the Math Magic

The biggest hurdle wasn't building the radar; it was dealing with the ocean. The ocean is never still. Even on a calm day, big waves roll by, and tiny bubbles pop on the surface. These create a massive amount of "noise" that drowns out the tiny signal from the underwater sound. It's like trying to find a specific melody in a song where a thousand people are shouting at once.

The researchers tried a few different ways to listen. First, they looked at the strength of the signal (amplitude), but that was like trying to hear the whisper by looking at how loud the shouting is; it didn't work well because the ocean's movement changed the loudness too much. Then, they tried listening to the timing of the signal (phase). This is like listening to the exact moment a drumbeat hits. Because the radar is so precise, even a tiny shift in the water's surface changes the timing of the echo. This method was much better, but the ocean was still too noisy.

So, they invented a new math trick called "Wavelet Packet Dynamic Thresholding." Imagine you have a recording of a noisy party, and you want to find one specific person humming a tune. A simple filter might just turn down the volume of everything, but that would make the hum quieter too. This new math trick is smarter. It breaks the sound down into tiny pieces and looks at them one by one. It knows that the ocean waves are slow and low-pitched, while the underwater sound is faster and higher-pitched. It acts like a super-smart bouncer at a club, letting the slow, loud waves pass through the door but stopping them from entering the VIP room where the tiny, high-pitched hum is. It then uses a special rule (called the SURE criterion) to decide exactly how much noise to cut out so it doesn't accidentally cut out the whisper.

The Real-World Test

The team didn't just do this on a computer; they went out to the real ocean. They set up their radar on a platform near the coast and later on a drone (a quadrotor UAV) hovering above the water. They placed an underwater sound source at different depths and played different tones.

Here is what they found:

• It works in the real ocean: They successfully detected underwater sounds even when the sea was at "Sea State 2," which means there were small waves and some spray. This is a big deal because previous tests were mostly done in quiet, controlled tanks.
• The depth limit: They found that the system works best when the sound source is close to the surface. When the source was at 0.5 meters deep, the signal was clear. When they lowered it to 2.0 meters, the signal got very weak and almost disappeared. It's like the sound gets "stuck" in the water as it tries to travel up to the surface.
• The height limit: When they used the drone, they found that if the drone flew too high (above 3 meters), the signal got fuzzy and hard to read. The drone's own shaking and the distance made it harder to see the tiny ripples.
• The power limit: The underwater sound had to be loud enough. If the source was too quiet (below 169 dB), the radar couldn't tell the difference between the sound and the background noise.

What This Means

The paper shows that we can "see" underwater sounds by watching the water surface dance, using a super-precise radar and a smart math filter. It's not a magic solution that works for everything yet. The team admits that if the sea gets really rough (higher than Sea State 2), or if the sound source is very deep, the system struggles. They also note that the drone needs to be very steady.

However, this is a major step forward. It proves that we don't always need to send a submarine to listen to the ocean; we might be able to listen from the sky. The researchers suggest that in the future, they could make the radar smarter by using artificial intelligence to recognize different types of underwater sounds, not just their pitch. They also want to test this in rougher seas and with better drones that can stay steady even in the wind. For now, they have shown that the invisible dance of the water surface is real, and with the right tools, we can finally hear the music playing underneath.