Turn Your Old Cell Phones into Listening Devices to Stop Illegal Logging(makezine.com)
makezine.com
Turn Your Old Cell Phones into Listening Devices to Stop Illegal Logging
http://makezine.com/2014/07/25/turn-your-old-cell-phones-into-diy-environmental-listening-devices-to-stop-illegal-logging/
16 comments
Thanks, Chuck, this is Topher, the guy from the article. It's clear that we see eye-to-eye on two of the fundamental tenets of the Rainforest Connection project, namely:
1) The Internet of Things need not wait years for specialized, embeddable hardware. Discarded smartphones offer an immediate shortcut that most developers are able to leverage almost immediately. They're chock full of sensors with mature APIs and built to be power efficient and reliable.
2) Surveillance is a emotional issue, but (to me) it has a different meaning when the hardware and software that produces the data are entirely open-source, and—most importantly—the data feeds produced are made immediately available to anyone who desires access. My hope is that Rainforest Connection will not be interpreted as "spy-[ware]" but rather as an open connection between the forest and those around the world. Forks of the project may be used for other purposes, but at its core, the project is about streaming audio/sensor data openly, and building modules for analyzing the data for distinct patterns (the first example of which is illegal logging).
If either of those statements doesn't make sense based on the Make article, I'd encourage you to check out the Kickstarter project page which contains a lot more information about our overall vision and activities:
https://www.kickstarter.com/projects/topherwhite/rainforest-...
1) The Internet of Things need not wait years for specialized, embeddable hardware. Discarded smartphones offer an immediate shortcut that most developers are able to leverage almost immediately. They're chock full of sensors with mature APIs and built to be power efficient and reliable.
2) Surveillance is a emotional issue, but (to me) it has a different meaning when the hardware and software that produces the data are entirely open-source, and—most importantly—the data feeds produced are made immediately available to anyone who desires access. My hope is that Rainforest Connection will not be interpreted as "spy-[ware]" but rather as an open connection between the forest and those around the world. Forks of the project may be used for other purposes, but at its core, the project is about streaming audio/sensor data openly, and building modules for analyzing the data for distinct patterns (the first example of which is illegal logging).
If either of those statements doesn't make sense based on the Make article, I'd encourage you to check out the Kickstarter project page which contains a lot more information about our overall vision and activities:
https://www.kickstarter.com/projects/topherwhite/rainforest-...
Thousands of phones sounds like a massive cell phone bill, even with the cheaper plans overseas referred to in the article. 10 snapshots a day adds up if you're charged a few pennies per mms. Would be great to see some some further elaboration on this point.
> Thousands of phones sounds like a massive cell phone bill,
> even with the cheaper plans overseas referred to in the
> article. 10 snapshots a day adds up if you're charged a few
> pennies per mms. Would be great to see some some further
> elaboration on this point.
Very good question/assertion.
This is Topher, from the article. In our experience so far (Indonesia & Africa), it's been shocking how affordably we're able to operate the system, though there are caveats when it comes to having to work around local standards (to be explained in a moment).
In Indonesia, in our first pilot test in 2013, we were actually able to get SIM cards with unlimited data in retail outlets the equivalent of $2.89/month. You can see a tweet from around that time in which we're quite pleasantly surprised: https://twitter.com/RainforestCx/status/342834137163505664
The caveat in this case was that these plans are pay-as-you-go, which means that credits must be purchased each month, and applied to the number through SMS. Thus, while in the field, we had to extend our API to keep track of credits, interact with the telecom over SMS, and translate keywords in Bahasa Indonesian SMS's that were returned. When an account was running low, another device in the forest was instructed to use a top-up code from (provided by the API) to "top-up" the phone with the low-balance. This is obviously a hack to emulate what people do for each other—namely purchase credits and top up each others' accounts.
Going forward, Rainforest Connection will be looking for semi-affordable monthly plans where available, and only performing work-arounds like the one described above when absolutely necessary.
It's also worth mentioning that each phone is able to hear chainsaw noises up to a kilometer in the distance, meaning that one single phone can "protect" almost 3 square kilometers (one square mile or 300 hectares) of forest. That means that 100 square kilometers can be protected with only 30 devices, which at nearly any GSM data price point is a good value.
I'd be happy to elaborate more if you have more questions. Thanks! -Topher
Very good question/assertion.
This is Topher, from the article. In our experience so far (Indonesia & Africa), it's been shocking how affordably we're able to operate the system, though there are caveats when it comes to having to work around local standards (to be explained in a moment).
In Indonesia, in our first pilot test in 2013, we were actually able to get SIM cards with unlimited data in retail outlets the equivalent of $2.89/month. You can see a tweet from around that time in which we're quite pleasantly surprised: https://twitter.com/RainforestCx/status/342834137163505664
The caveat in this case was that these plans are pay-as-you-go, which means that credits must be purchased each month, and applied to the number through SMS. Thus, while in the field, we had to extend our API to keep track of credits, interact with the telecom over SMS, and translate keywords in Bahasa Indonesian SMS's that were returned. When an account was running low, another device in the forest was instructed to use a top-up code from (provided by the API) to "top-up" the phone with the low-balance. This is obviously a hack to emulate what people do for each other—namely purchase credits and top up each others' accounts.
Going forward, Rainforest Connection will be looking for semi-affordable monthly plans where available, and only performing work-arounds like the one described above when absolutely necessary.
It's also worth mentioning that each phone is able to hear chainsaw noises up to a kilometer in the distance, meaning that one single phone can "protect" almost 3 square kilometers (one square mile or 300 hectares) of forest. That means that 100 square kilometers can be protected with only 30 devices, which at nearly any GSM data price point is a good value.
I'd be happy to elaborate more if you have more questions. Thanks! -Topher
Are the phones able to tell where in the protected area the sounds come from (eg. distance and radius) or can it just tell us the approximate distance?
Also you might not have to protect all the areas - if the devices are sufficiently hidden from view then people don't necessarily know where it is safe to loot, and so you end up protecting a much wider range.
Also you might not have to protect all the areas - if the devices are sufficiently hidden from view then people don't necessarily know where it is safe to loot, and so you end up protecting a much wider range.
Thanks for the detailed answer. Hopefully you'll get it to the point where the telecommunications companies will be willing to sponsor you and provide you with free data to make your life easier - but those hacks you cooked up are really cool.
I had no idea cellphone microphones were that sensitive btw.
I had no idea cellphone microphones were that sensitive btw.
Trying to understand that title involved double takes and several extra seconds.
A magnificent Stroop effect for the post-2013 era!
https://en.wikipedia.org/wiki/Stroop_effect
A magnificent Stroop effect for the post-2013 era!
https://en.wikipedia.org/wiki/Stroop_effect
Good or bad? Did it catch your attention and curiosity?
I could see something like this being used in large cities to monitor noise levels, especially from construction.
You could then build real-time maps of the noise in the city, identify illegal construction noise on the week-ends for instance, or simply help people choose which area they want to live in.
You could then build real-time maps of the noise in the city, identify illegal construction noise on the week-ends for instance, or simply help people choose which area they want to live in.
Legal logging is a bigger threat.
> Legal logging is a bigger threat.
This is common misconception, which should be true, but in fact isn't. Environmental crime is rampant and highly profitable. Interpol estimates that 50-90% of rainforest logging is illegal:
http://www.interpol.int/Crime-areas/Environmental-crime/Proj...
This is common misconception, which should be true, but in fact isn't. Environmental crime is rampant and highly profitable. Interpol estimates that 50-90% of rainforest logging is illegal:
http://www.interpol.int/Crime-areas/Environmental-crime/Proj...
My thesis is that the surveillance state exists because it was too expensive for anyone other than state actors to create one. But the proliferation of cell phones, and their concomitant ability to record and distribute data is making it possible for individuals and small groups to do things like this rain forest project. One of the things I predict will happen is that at some point a big push will be made to prevent people from randomly disposing of their old phones, instead either rewarding them monetarily or requiring their disposal in an 'approved' way. At the recent visit to the equipment liquidator I sometimes buy from I saw a pallet with a box full of phones. All kinds from Nokia candybar phones to blackberries to off brand android phones. There had to be 2500 to 3000 phones in the box. I doubt they sold for more than breakage (0.03/lb). But a motivated individual could 'fix' that like this guy did. Buy them attach a solar power + battery source and cheap sim cards and doing nothing more than the phone was designed to do (record sound or take a picture, send an MMS message) Lots of interesting intelligence over a wide area could be acquired. Want to know who the drug dealers are in a neighborhood? Or who works during the day? Or which houses have nobody in them? You could do a lot taking 10 snapshots a day and sending them to a central server.
Clearly it's anathema to a control state to have that ability in the hands of someone outside their control, so I predict some mobilization against it, from the phone "trade in" that is too good to be true to new rules and regulations about running an unattended computing device with recording capabilities.