A New Wave of Underwater Comms Is Coming | Hacker News Reader

A New Wave of Underwater Comms Is Coming

spectrum.ieee.org

88 pointsBrajeshwar33 comments

I didn't find anything about the potential effects on marine life. I'm not in the market for this, so maybe it's obvious to those who are, but it would be nice to address it somewhere for the uninitiated.

But their modem[1] uses NATO's JANUS standard[2], which communicates at 11.5 kHz, which is just audible to humans, but well within range of marine mammals[3].

1: https://www.subseapulse.com/products/ 2: https://spectrum.ieee.org/nato-develops-first-standardized-a... 3: https://en.wikipedia.org/wiki/Hearing_range#/media/File:Anim...

I completely agree, filling oceans/seas with noise seems like a good recipe to further stress marine ecosystems

To be completely frank, we haven't done anything to stop the rise of water temperature and acidification from CO2 absorption that prevent fish from reproducing, along with continuing overfishing. Most of these ecosystems will be dead from food chain collapse in a decade or two.

I suppose the good news is that jellyfish that will step in to fill the niche don't care about sound.

they will adapt like they always did

Yes, the numbers in the ecosystem will shift. Some to zero. That's an ecosystem adapting. Is it good?

Yeah dinosaurs adapted into chicken.

Overfishing and pollution will wipe out most of the remaining ocean fish in the next couple decades if nothing is done about it. We're on track to destroy 99% or more of the species of fish and large animals in our oceans.

Maybe state funding for privateering navies given license to sink illegal fishing boats would be a good start. Trawling and other wildly destructive practices funded by various south asian countries are doing incalculable harm all over the world. Sink on sight is an appropriate policy.

Humanity has to get their shit together, or jellyfish will be all that's left.

You also don't want to find out the hard way that e.g. your new modem enrages sealions or something.

For underwater acoustic systems there is a well understood and effective marine mammal mitigation process that’s followed. Basically, slowly stepping up in power. Not sure if they’re employing that here but if it involves NATO, I would expect that’s being employed

I'd like to know more about this process. But ultimately, it can only be detrimental to sealife to emit noise into the marine environment.

Imagine someone set up a radio or a noisy machine somewhere in your own vicinity and you had no choice but to listen to it (even at a low volume) all day and night, perhaps for the rest of your life. Tests might or might not show that you were stressed, but you most certainly would be.

Perhaps I am biased from regular snorkelling and enjoying being immersed n a different natural world, and for sure I do have tree-hugger tendencies. But I find it extremely arrogant that our species thinks it's acceptable to inject noise pollution into an environment inhabitated by thousands of other species.

Not to get too deep in the underwater weeds here but sound propagation underwater is not omnidirectional like how sound propagates above water. And I’m not sure how true that is either

All that to say, I certainly understand the concern. But I would personally have to see evidence of harm

The Precautionary Principle calls for proof of safety, not proof of harm. Harm is assumed unless proven otherwise.

I assume that the Precautionary Principle is harmful and will refuse to accept it until it has been proven safe.

> Imagine someone set up a radio or a noisy machine somewhere in your own vicinity and you had no choice but to listen to it (even at a low volume) all day and night, perhaps for the rest of your life. Tests might or might not show that you were stressed, but you most certainly would be.

Otoh, some people purposefully setup noise machines to improve sleep and reduce stress. Sometimes a small amount of noise is a good thing.

"Imagine someone set up a radio or a noisy machine somewhere in your own vicinity and you had no choice but to listen to it", you mean like a mosque :-D

Or church bells.

"but if it involves NATO, I would expect that’s being employed"

Why would that be your expectation?

> a well understood and effective marine mammal mitigation process that’s followed. Basically, slowly stepping up in power.

Uhh... I don't consider myself an eco-hippie, but that "mitigation" sounds like a euphemism or sinister "mitigation of negative PR", rather than actually literally doing less damage.

Like: "Gradually make things worse so that animals flee on their own, and if they die of starvation at least it's somewhere else so it won't make the local news with kids finding corpses on the beach."

Years ago an oceanography-oriented founder told of attempting to patent their nice transducer. The response was unexpected: the prior art is classified, so no patent, and stop all use.

Yes, from what I’ve heard, acoustic listening in the ocean is limited/supervised to avoid picking up secret submarine signatures.

Garmin scuba diving products now have underwater communication features built in. Unfortunately they're locked down so you can't use them for any other purposes.

https://www.garmin.com/en-US/garmin-technology/dive-science/...

For those who aren’t scuba divers, 10m range is very impressive. Most radiowave-based systems have a range of 1-2m, with many being plagued with connectivity issues.

One of the main use-cases is air integration, which allows you to have a transmitter connected to your cylinder which monitors gas pressure and relays it to the dive computer on your wrist.

This is an alternative (or, depending on the school of thought, in addition) to having an SPG (submersible pressure gauge) which is hopefully attached to your left d-ring, but for many recreational divers stuffed in a pocket somewhere or scraping on the bottom.

Air Integration is quite an investment (3-500€ per transmitter) and there is basically no interoperability between brands.

When Garmin came out with this, it made quite a splash, as all of a sudden instructors could keep an eye on the cylinder pressure of their students. This is a very minor use-case though, and I don’t believe it’s used a lot.

As for underwater communication, a good chunk of the scuba community is quite opposed to relying on electronics for communication. Hand-based signals are quite good, underwater paper does exist, and if you’re so far from your buddy that you need to wirelessly communicate you’re in a lot of trouble.

This all being said, as a boat captain responsible for the people underwater, I would love a local area network that allows me to pull the plug on a dive, or get some emergency notification. For example “weather is turning, please call the dive” or “I was blown off the wreck, please pick me up”.

Garmin occasionally sends me customer surveys, and the questions on a recent survey implied that they were looking into boat to diver communications as a use case. I don't know whether they're actively working on this, but it seems like it could possible to add to future versions of their boat sonar products.

> This all being said, as a boat captain responsible for the people underwater, I would love a local area network that allows me to pull the plug on a dive, or get some emergency notification. For example “weather is turning, please call the dive” or “I was blown off the wreck, please pick me up”.

In case you are interested, there is a Finnish company called UWIS (short for underwater information systems), that does exactly this [1]. Their system has buoyes that can track the diver units and provide communication channel between on the surface pc and the divers via wifi and sonar network. The system allows two way communication. Tight bandwidth of course as a limiting factor.

[1] https://uwis.fi/en/

For underwater communication, nothing beats the ol' hit-tank-with-knife (or the slightly more sophisticated tank banger).

https://www.amazon.com/s?k=tank+banger

This sounds like another startup that is ignoring all of the edge cases, testing, calibration, and certification that make this kind of product expensive.

In this case it's a good idea. And they know exactly where their market is. I hope it takes off. :)

There are plenty of applications where the reliability requirements are it works most of the time.

It seems that the algorithm mentioned in the article is here [1], [2]. It has been close to 20 years before a working prototype being implemented, I wonder what caused the delay in the implementation?

The original underwater channel model and propagation however, fittingly the Professor's most cited paper, came later which kind of strange since you usually come up with channel model and propagation first before the modulation, not the reverse [3].

[1] Low Complexity OFDM Detector for Underwater Acoustic Channels (OCEANS 2006):

https://ieeexplore.ieee.org/document/4098993

[2] MIMO-OFDM Over An Underwater Acoustic Channel (OCEANS 2007):

https://ieeexplore.ieee.org/document/4449296

[3] Underwater acoustic communication channels: Propagation models and statistical characterization (IEEE Communications Magazine 2009):

https://ieeexplore.ieee.org/document/4449296

"The team is also working on a cheaper version of traditional transducers, which can cost more than $2,000 a piece" - Do these transducers require very high input power out of interest?

Also what frequency does an acoustic modem typically use?

And finally are they expensive because they're somewhat niche, or difficult to manufacture, both, or other reasons?

Edit: looks like the datasheet for 'Waterlinked M16', mentioned in another comment answers the first couple of questions. Seems a lot lower in power than I'd have thought.

Not very high power usually. They're usually made of piezoceramics.

I think they're expensive because they're niche. Narrow band piezoceramics which are used in fish finders and other commercial equipment are pretty cheap. These wider band transducers are niche to underwater comms.

> Also what frequency does an acoustic modem typically use?

Usually in the 10s of kHz. The NATO JANUS standard is around the 12kHz center frequency.

It seems to me that if you need to generate large acoustic signals underwater, HASEL actuators might be the way to go.[1] Since they're effectively just bags full of incompressible fluid, they could work at any depth.

Sensing sound.... that might be harder. Perhaps you could close-loop control the capacitance of the above, and "listen" to the error signal?

[1] https://www.artimusrobotics.com/

> Their transducer will convert energy into acoustic signals (and vice versa) using a modified device that’s typically used to listen to marine mammals and costs about $400.

That does beat the next cheapest option, the Waterlinked M16 that's $2.4k for 1km range and 10 bits/s. They don't list any specifics about capabilities though.

There's an open source project for that:

https://blog.unetstack.net/Project-Sabine-Low-cost-DIY-under...

Google Lee fritag (sp?) Or the whoi micromodem to get the current sota.

Reply on news.ycombinator.com