Can Orcasound imagine a buoy-based low-cost live-listening hydrophone system for all orca signals?

[scottveirs]

Let's discuss a buoyed solution. Mostly we have been using cabled deployment strategies, but in some otherwise-promising locations a low-cost offshore (probably buoyed**) hydrophone system would be handy...

Pointed questions:
AKA preliminary design constraints for awesome design/engineer groups like the Bainbridge BARN

• 1. Can new battery technology help us not need to change the batteries too often?
• 2. What current wireless technology could get the signal to shore (and ultimately the Orcasound live-listening app?
• 3. What hydrophone deployment techniques will minimize flow, cable, and other sources of system or buoy/mooring noise?
• 4. What hydrophone and hardware combination would let us sense all signals of Southern Resident Killer Whales (calls, whistles, and clicks), or at least infer their presence by detecting calls (with mean fundamentals from 200-2000 Hz and harmonics up to ~20,000 Hz).

** Orcasound is totally open to other deployment strategies, but we are assuming for now that the tides and narrow channels of the Salish Sea rule out other extant approaches for near/continuous PAM, like wave-gliders, autonomous boats, sea gliders, etc.

[veirs]

Hydrophone Links:
(see this repository's wiki: Hardware and field deployment -- for much information on our live streaming notions.)
https://github.com/orcasound/orcanode/wiki

https://www.instructables.com/Lets-Build-Some-World-Class-Hydrophones/

Bioacoustice group in StackExchange:
https://bioacoustics.stackexchange.com/questions/1269/what-affordable-hydrophone-options-exist-for-in-the-field-listening

Val''s current phantom-powered preamp circuit which is under development!!:

[bonnettde]

Hello Val and Scott
We will answer your specific questions and getback to to you soon. Below is an preliminary outline of our Barn Project
WetSound Project Outline .docx

[scottveirs]

Hey David, welcome to Orcasound! As we continue our discussion here, you might enjoy learning about Markdown on Github. To save folks without access to Word the trouble of downloading your Word doc and then opening it a compatible program, I used Pandoc online to convert your nice outline into Github markdown (and added headings):

Wetsound Project

##Things to do

1. Barn Presentation

2. Acquisition System Design

   1. Hydrophone

      1. Research Hydrophone Technologies

      2. Prototype development

         1. Select hydrophone technology

         2. Assemble hydrophone

         3. Test protocol

         4. Test Hydrophone

   2. Acquisition System Design

      1. Subsurface

         1. Anchor System`

            1. Mooring Line

               1. Mooring Line Selection

                  1. Requirements

                     1. Support Buoy System

                     2. Non-fouling

                  2. Design Mooring Line

            2. Anchor

               1. Selection

                  1. Requirements

                     1. Cannot Drag

                     2. Easy to deploy and recover

                  2. Design Anchor

         2. Electronic Acquisition System

            1. Raspberry Pi

               1. Program

               2. A/D conversion

            2. Power Supply

               1. Battery Selection

               2. DC/DC converter 12VDC to 5 VDC

               3. Solar Controller

            3. Configure/modify Raspberry Pi

               1. Design and build send/receive antenna

               2. Data SD Card Memory

            4. Bench Test System

               1. Program functionality

               2. EMI

                  1. Resolve interference issues

      2. Surface

         1. Mooring Buoy

            1. Buoy Selection

               1. Requirements

                  1. Cylindrical

                  2. Support 2 antennas

                  3. Support Solar Panel

                  4. Night light

                  5. Buoy must remain on surface

                  6. Long lifetime

               2. Design Buoy

      3. Shoreside

         1. Requirements

            1. Computer System

               1. WiFi Capable

               2. Audio Output

               3. Wave File memory storage

               4. Operating System-Windows, Mac, Linux

               5. Ram-8GB

               6. Hard Drive 100GB

            2. Software

               1. Audacity

               2. Advanced Analysis

                  1. Raven

                  2. Matlab

         2. Raspberry Pi Code

            1. Develop

            2. Test

[bonnettde]

Thanks Val. I'll try Pandoc

[scottveirs]

This is Scott, Val's son, actually!

[scottveirs]

@bonnettde We can definitely help with 2.i.a-b (Hydrophone R&D) and maybe the streaming/analysis side of things, but you will be helping us learn about anchors and buoys!

Also, I might add one subsurface challenge: minimizing flow and other noise from the mooring system. I think there are some "tricks" for isolating the hydrophone acoustically, but the impacts of flow noise may depend mostly on where you deploy (Race Passage or Agate Pass harder than on east side of Bainbridge facing main basin of Puget Sound).

Possible leads:

• Study and/or ask how the Jupiter Foundation isolated the hydrophone on their buoyed systems off Puako (Kona coast of Big Island, Hawaii) and later on their waveglider prototypes. Their website has changed a lot over the years, so you'll have to dredge through using archive.org -- e.g. this equipment page on "sonobuoy" design. As I recall (from a brief site visit ~2004), they lashed a circular ring/loop of vibration absorbing / elastic material to the mooring line and then mounted hydrophone on the outer edge of the loop.
• Study Sofar's Spotter buoy and Smart buoy.
• Talk to Jason Wood at SMRU, Ltd. in Friday Harbor about lessons learned from engineering their Coastal Acoustic Buoys (CABs)?
• talk to engineers at NOAA/PMEL, UW/APL/OOI about mooring technology in general, and specifically keeping hydrophones quiet on/near moorings
• Consider off-the-shelf components (like this solar buoy) vs DIY-only solutions

[scottveirs]

For the record, from the Internet Archive, here's a version of the old Jupiter Foundation page on the equipment they used to listen for humpbacks back in ~2005...

I have no idea what they used to anchor the sonobuoy, nor how the hydrophone was connected to the transmitter. But in general, I'm very interested in exploring analog-only transmission to shore. If that was easy and fidelity wasn't lost in transmission, then the buoy engineering might be greatly simplified -- effectively moving most of the electronics from the subsurface/buoy to onshore...

[bonnettde]

Hi Scott Thanks for this info. We are hoping to be able to design an in water package with a smaller footprint than Jupiter and do the A-D onboard then transmit via wireless to shore. We also will use a Li Ion battery and have sufficient on board storage for .wav files to be held until the user wants to download them. Of course this is all in the preliminary stage of concept design so any info like Jupiter is welcomed.

…

On Sat, Jan 28, 2023 at 6:29 PM Scott Veirs ***@***.***> wrote: For the record, from the Internet Archive, here's a version of the old Jupiter Foundation page on the equipment they used to listen for humpbacks back in ~2005... [image: Screenshot 2023-01-28 at 6 08 44 PM] <https://user-images.githubusercontent.com/14044595/215301050-64fa57b0-04cc-4ba7-88c0-c727613fdce8.png> I have no idea what they used to anchor the sonobuoy, nor how the hydrophone was connected to the transmitter. But in general, I'm very interested in exploring analog-only transmission to shore. If that was easy and fidelity wasn't lost in transmission, then the buoy engineering might be greatly simplified -- effectively moving most of the electronics from the subsurface/buoy to onshore... — Reply to this email directly, view it on GitHub <#59 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACEPVXEXFQ3RDUKN4BOZ4U3WUXIXLANCNFSM6AAAAAAUJKX5MY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[scottveirs]

Sounds, good. Two thoughts:

1. Why transmit to shore and also store onboard? (Val mentioned you are interested in studying SRKW clicks so focused on 192kHz sampling rate minimum, so maybe you intend to store full lossless data onboard and transmit lower resolution, lossy data to shore?)
2. If you're going to store onboard, you might consider FLAC or X3 lossless compression which are more efficient than WAV. Orcasound currently streams lossy in HLS (but would like to try DASH at some point) and archives in FLAC (so far 48kHz max, but the Pisound ADC in Orcasound's current open hard/soft-ware solution can do 192kHz at 24 bits for 2 channels).

[bonnettde]

We are currently hoping the Raspberry Pi Zero 2W (https://datasheets.raspberrypi.com/rpizero2/raspberry-pi-zero-2-w-product-brief.pdf ) can be set up to accomplish both real time (near) audio to shore and to store .wav files that can be analyzed later. The 2W has a SD sound card slot USB and uses both Bluetooth and 802.11 wireless. With appropriate programming we hope we can set sample rates for A/D at up to 500KHz. Our first priority is to develop a hydrophone that is affordable and has an appropriate sensitivity profile to allow fairly high sample rates. I like your Orcasound solution and will look it over.
Thans for this information
Dave

[scottveirs]

@dbainj1 This is another place to try to push forward on an Orcasound buoyed solution!

[scottveirs]

@bonnettde What radio technology are y'all considering for your first prototype?

For the Sunset Bay hydrophone site we might get by with one of the wireless mics used for sports and entertainment events. One option is the Rode Go II wireless mic that claims to have a range of about 250 meters. Glazer's Rentals in Seattle will rent one of these (for testing) at a rate of $20/day.

Other options they rent for are made by DJI and Sennheiser

[scottveirs]

Hey @bonnettde and @dbainj1 -- have you all seen this buoy/drifter project for real time classification of underwater noise from vessel?
https://github.com/IQTLabs/AISonobuoy#particletrackerbuoy?

Looks like some interesting components and DIY ideas (e.g. PVC pipe for waterproof housing)...

[scottveirs]

On the ride home from our first visit to the BARN, @dbainj1 Dave mentioned long ago using a pager to turn on/off a buoyed acoustic project. It looks like there is a similar low bitrate (e.g. texting or walkie talkie) technology called LoRa that might be of interest — https://en.m.wikipedia.org/wiki/LoRa — e.g. for waking up a buoy?

Not sure they will deliver the promised 5 km messaging and 300 meter “crystal clear” audio transmission, but this active MessengerPi Kickstarter looks interesting! https://www.kickstarter.com/projects/diytech/messengerpi-lora-based-messenger-and-walkie-talkie

[klaplain]

If we assume that we need to accurately transmit sounds with 16 bit resolution at frequencies up to 20kHz, what distance does the WiFi need to reach? Is it line of sight? There's a big difference between 300m and 5km. Lora is viable for low bandwidth pathways but I don't think it is suitable for the real-time transmission of high resolution sound

[dbainj1]

The distance needed will vary with site.  About 500', give or take, would be a good initial goal.  Line of sight can typically be assumed.  When line of sight isn't available, it's likely that relays could be set up that allow line of sight.We've looked at FM radio, which requires a license to go over 200'.  Bluetooth is another option, but probably won't reach 500'.  Wi-Fi is an option that's good for miles, if you can keep the antennas lined up.  Wide band FM is another technology that could go to higher frequencies.  A mobile hotspot can work where there is good service.  Satellite Internet is another option, although expensive and typically limited to less than 3 Mb per second.Of course, if you have a local resident who just needs to get to the end of the dock, the distance could be far shorter.  We've had one application where we needed to connect Wi-Fi a few miles away (it worked most, but not all the time).Radar is a possible source of interference at Wi-Fi frequencies.I'd suggest developing a system that uses technology you're comfortable working with.  We'd apply it where it can work.  If you develop longer range solutions in the future, we can apply them at more challenging sites.DaveSent from Samsung Galaxy smartphone. -------- Original message --------From: klaplain ***@***.***> Date: 11/22/23 1:35 PM (GMT-08:00) To: orcasound/orcanode ***@***.***> Cc: dbainj1 ***@***.***>, Mention ***@***.***> Subject: Re: [orcasound/orcanode] Can Orcasound imagine a buoy-based low-cost live-listening hydrophone system for all orca signals? (Discussion #59) If we assume that we need to accurately transmit sounds with 16 bit resolution at frequencies up to 20kHz, what distance does the WiFi need to reach? Is it line of sight? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[scottveirs]

@klaplain yes I agree with @dbainj1 that a great initial design goal for a wireless orca buoy would be:

• Resolve frequencies up to 20 kHz (could be relaxed to 8-10 kHz IMHO)
• 16 bit depth
• 150m (~500') range to buoy from streaming computer ashore

[scottveirs]

A Google search yesterday revealed this interesting effort to build a smart buoy for Orcasound! (I wasn't aware of this til now.)

I'm new to cellular IoT communication schemes, but it appears the chip on the LTE extension board may support:

"CAT-M1 R14 with DL/UL speeds of 588/1119 Kbps"

Especially with some lossy compression, it seems that upload bandwidth might be sufficient...

[klaplain]

Yep. For distances of a few 100 feet, WiFi/Bluetooth would be OK. But for longer distances LTE is a better solution. Data rates should support the 16bit/20kHz you need. There would be a subscription cost for the LTE data link (estimate $100/year). The Spresense Sony board is a great way to test for viability. An alternative BARN strategy would be to reuse the existing BARN preamplifier & STM32 ADC and to pair this with a board such as the Adafruit FONA 3G Cellular board. (https://www.adafruit.com/product/2687)