So many $1 wireless ICs have a powerful "SDR", but will never be documented due to certification/compiliance/export-control reasons. The projects listed here rightly limited the scope to RX-only but Espressif might be forced to "patch" this away if arbitrary TX too is possible (or starts making headlines). Hope it doesn't come to that, more cheap options for RF-to-bits is always welcome.
On a technical note, not much I can find about the signal quality. Phase noise, images, DC Offset, etc. Have to get my hands on one, ESP32-S31 seems promising.
It is legal for a person with an appropriate license, yes. But laws (in the US) about RF transmission differ from how many other laws are structured: We're only allowed to transmit what we have permission to transmit -- and by default, we're forbidden.
But right now, I've got an ESP32 WROOM module on a dev board sitting in my windowsill reporting some sensor values over wifi.
The WROOM itself is registered with the FCC[1]. If I stick to using the documented operation, then I have permission to use it in those documented ways. As a practical concept, I don't need a unique license to use it in the way that I am; I'm granted the ability to use it by Part 15 of the FCC rules.
And that's pretty great: The hardware was inexpensive to buy and very easy to find. I was able to spin this sensor up very quickly and get moving on the software side that deals with those readings. It has been a very rewarding little project with a fantastic cost:benefit ratio.
The fear being expressed here (which I don't necessarily subscribe to) is that things like this WROOM module will become less-available when/if methods to use it in undocumented ways are published and start becoming used.
>>"We're only allowed to transmit what we have permission to transmit -- and by default, we're forbidden."
It is not true that by default you are forbidden to transmit. In fact there are quite open rules in some ways for transmissions. Whenever these discussions about transmission come up there's always people who, without really undrstanding, freak out about transmitting anything and declare that all tranmissions are illegal. Not so.
USA hobbyists can legally transmit without a licence under FCC Part 15, but the limits are pretty low and depend on the frequency. AM, up to 100 mW into the final RF stage, with the antenna, feedline and ground limited to a total of 3 metres. FM maximum field strength of 250 µV/m measured 3 metres away. There are special cases, such as 49.82–49.90 MHz, where home-built transmitters can go to 100 mW on a 1 metre antenna. One-off home-built transmitters don't need FCC certification if for personal use not being sold.
We're granted permission to make the transmissions you speak of; it is otherwise deny-by-default. This is described right there at the top of your first link, in the first three paragraphs.
It's broadly possible to sort any rule (or superset or subset of rules) into one of two different buckets:
One bucket contains the rules about the things things we're not allowed to do. Like a firewall with a blacklist, whereby we can do whatever we want that explicitly isn't blacklisted. It is allow-by-default.
The other bucket contains the rules about things that we are permitted to do. This is like a firewall with a whitelist, where we can do whatever the whitelist explicitly allows us to do and nothing more. It is deny-by-default.
Drug laws in the USA are in the first bucket (blacklisting). Radio transmission rules in the USA are in the second bucket (whitelisting).
Certified unlicensed radio equipment are required to be reasonably tamper resistant, so that reasonably everything that end users might do would be practically compliant even with a bit of malice. This is one of reasons why iPhone used to use pentalobe screws.
They are not required to be as secure as credit card chips, but they also can't be wide open.
This would mean throwing away a giant market for what are generally pretty niche use cases. Very cool use cases, sure. But not worth giving all those sales up in favor of gray-market importers.
As far as I follow the research process it's currently quite difficult to get all data (like the 80MSPS@10-Bit showcase on Reddit) to a computer without a FPGA+USB3.
With the new ESP32-S31 we most likely can extract the I/Q data over its new 1 GBit/s interface. Guessing something around 20-40MSPS should work.
These hack will most likely will be a revolution for 13cm ham radio. With the new 5GHz ESP32 modules most likely also for 5cm. As all Espressif components are WiFi-qualified we can expect a very very stable and robust radio.
I ordered a few and waiting for Ali to deliver them to my lab.
>compresses to about 1.2 bytes per pair, around 95–100 MB/s
No mention what FPGA platform/board used, FT600 USB chip is ~$10 alone.
Now you can replace all of it with one of those silly $5 FX2LP USB 2.0 dev boards. CY7C68013A-56 supports 16bit interface and will happily pump 30MB/s all day long for a fraction of the price. $10 total for 2.4GHz ~12MSPS@10-Bit SDR.
The ESP32S3 support PSRAM, can we redirect the sample to PSRAM, then hooking Logic Analyzer on PSRAM, thus getting the sample directly, not through any digital interface.
(There is cache before PSRAM, but should be trivial to sync).
Not with RTL-SDR specifically but by introducing one of the first open-architecture SDRs, the USRP.
RTL-SDR is the cheapest imaginable commodity implementation of SDR technology, while USRP was... well, pricy enough to bring in some solid revenue and get his company bought. More power to him, it was (and is) an incredibly useful platform.
So many $1 wireless ICs have a powerful "SDR", but will never be documented due to certification/compiliance/export-control reasons. The projects listed here rightly limited the scope to RX-only but Espressif might be forced to "patch" this away if arbitrary TX too is possible (or starts making headlines). Hope it doesn't come to that, more cheap options for RF-to-bits is always welcome.
On a technical note, not much I can find about the signal quality. Phase noise, images, DC Offset, etc. Have to get my hands on one, ESP32-S31 seems promising.
I'm pretty sure that that WiFi-7 modems/routers have a similar modes up to 320MHz bandwidth.
The modems could be quite interesting as they are connected via PCIe which can eat a lot of bandwidth.
But they are very difficult (to almost impossible) to find if there isn't anything left in the firmware blob or device driver.
https://github.com/F5OEO/rpitx
You don't need much to spam random RF energy into the air. That project just flips a GPIO pin at the right frequency, and gets a signal into the air.
It won't be very powerful, and the harmonics will probably bleed all over the place if you just hang a wire off of it with no filters.....
> The projects listed here rightly limited the scope to RX-only but Espressif might be forced to "patch" this away if arbitrary TX too is possible
Why would they? It's perfectly legal to operate radio transmitters on any frequency you like, if you hold a suitable licence.
It is legal for a person with an appropriate license, yes. But laws (in the US) about RF transmission differ from how many other laws are structured: We're only allowed to transmit what we have permission to transmit -- and by default, we're forbidden.
But right now, I've got an ESP32 WROOM module on a dev board sitting in my windowsill reporting some sensor values over wifi.
The WROOM itself is registered with the FCC[1]. If I stick to using the documented operation, then I have permission to use it in those documented ways. As a practical concept, I don't need a unique license to use it in the way that I am; I'm granted the ability to use it by Part 15 of the FCC rules.
And that's pretty great: The hardware was inexpensive to buy and very easy to find. I was able to spin this sensor up very quickly and get moving on the software side that deals with those readings. It has been a very rewarding little project with a fantastic cost:benefit ratio.
The fear being expressed here (which I don't necessarily subscribe to) is that things like this WROOM module will become less-available when/if methods to use it in undocumented ways are published and start becoming used.
[1]: https://fccid.io/2AC7Z-ESPWROOM32
>>"We're only allowed to transmit what we have permission to transmit -- and by default, we're forbidden."
It is not true that by default you are forbidden to transmit. In fact there are quite open rules in some ways for transmissions. Whenever these discussions about transmission come up there's always people who, without really undrstanding, freak out about transmitting anything and declare that all tranmissions are illegal. Not so.
USA hobbyists can legally transmit without a licence under FCC Part 15, but the limits are pretty low and depend on the frequency. AM, up to 100 mW into the final RF stage, with the antenna, feedline and ground limited to a total of 3 metres. FM maximum field strength of 250 µV/m measured 3 metres away. There are special cases, such as 49.82–49.90 MHz, where home-built transmitters can go to 100 mW on a 1 metre antenna. One-off home-built transmitters don't need FCC certification if for personal use not being sold.
It's pretty interesting to read if you are a radio head: https://www.ecfr.gov/current/title-47/chapter-I/subchapter-A...
https://www.ecfr.gov/current/title-47/section-15.209
"operation within these frequency bands is permitted under other sections of this part, e.g., §§ 15.231 and 15.241."
-->
0.009-0.490Mhz 0.490-1.705Mhz 1.705-30.0Mhz 30-88Mhz 88-216Mhz 216-960Mhz Above 960Mhz
We're granted permission to make the transmissions you speak of; it is otherwise deny-by-default. This is described right there at the top of your first link, in the first three paragraphs.
It's broadly possible to sort any rule (or superset or subset of rules) into one of two different buckets:
One bucket contains the rules about the things things we're not allowed to do. Like a firewall with a blacklist, whereby we can do whatever we want that explicitly isn't blacklisted. It is allow-by-default.
The other bucket contains the rules about things that we are permitted to do. This is like a firewall with a whitelist, where we can do whatever the whitelist explicitly allows us to do and nothing more. It is deny-by-default.
Drug laws in the USA are in the first bucket (blacklisting). Radio transmission rules in the USA are in the second bucket (whitelisting).
(There is no third bucket.)
The FCC is relevant in exactly one country in the world.
Other countries have equivalent organizations and they all tend to be in sync on this stuff.
Certified unlicensed radio equipment are required to be reasonably tamper resistant, so that reasonably everything that end users might do would be practically compliant even with a bit of malice. This is one of reasons why iPhone used to use pentalobe screws.
They are not required to be as secure as credit card chips, but they also can't be wide open.
Legal for end-user, sure. But Espressif might run into trouble with regulators like FCC.
Then sell them in the UK, where the FCC is entirely irrelevant.
We treat RF like your lot treat Mexican fireworks.
This would mean throwing away a giant market for what are generally pretty niche use cases. Very cool use cases, sure. But not worth giving all those sales up in favor of gray-market importers.
Our border control confiscates Mexican fireworks.
>if arbitrary TX too is possible
https://espargos.net/espsdr/ straight up shows its an ADC/DAC debug interface working both ways.
As far as I follow the research process it's currently quite difficult to get all data (like the 80MSPS@10-Bit showcase on Reddit) to a computer without a FPGA+USB3.
With the new ESP32-S31 we most likely can extract the I/Q data over its new 1 GBit/s interface. Guessing something around 20-40MSPS should work.
These hack will most likely will be a revolution for 13cm ham radio. With the new 5GHz ESP32 modules most likely also for 5cm. As all Espressif components are WiFi-qualified we can expect a very very stable and robust radio.
I ordered a few and waiting for Ali to deliver them to my lab.
>FPGA+USB3 80MSPS@10-Bit
>compresses to about 1.2 bytes per pair, around 95–100 MB/s
No mention what FPGA platform/board used, FT600 USB chip is ~$10 alone.
Now you can replace all of it with one of those silly $5 FX2LP USB 2.0 dev boards. CY7C68013A-56 supports 16bit interface and will happily pump 30MB/s all day long for a fraction of the price. $10 total for 2.4GHz ~12MSPS@10-Bit SDR.
The GitHub README says the FPGA used: Alchitry Au (Artix-7 XC7A35T, 256 MiB DDR3)
The ESP32S3 support PSRAM, can we redirect the sample to PSRAM, then hooking Logic Analyzer on PSRAM, thus getting the sample directly, not through any digital interface. (There is cache before PSRAM, but should be trivial to sync).
Very exciting!
> However, the current prototype uses the FPGA to clock the ESP32, which results in poor phase noise.
Looks like this was solved five days ago:
https://github.com/h0m3us3r/eSpDR/commit/41a0ffe279119df942e...
Any way to finagle a LORA type system (ultra reliable using precise timing) using a few of these on each end?
Since I'm a little slow and it took me a minute:
SDR in this context is software defined radio.
> undocumented feature that lets the firmware bypass the fixed WiFi and Bluetooth functionality and instead capture raw IQ baseband samples
So kind of like those TV USB sticks that you can adapt into proper SDRs?
The ones that the website hosting this article is named after?
Yep, Matt Ettus converted that to hundreds of millions of dollars
RTLSDRs only have a 8-Bit ADC that easily drown next to strong transmitters.
I remember the Ettus devices to be much much much more powerful.
the Ettus USRP SDRs predate RTL-SDR by almost (not quite) a decade.
I was always into the soundcard HF SDRs, they were so simple and elegant. An oscillator, a ring counter, a 4052 MUX and one really fast dual opamp.
This sounds interesting. How do they work, or where can I learn more?
Not with RTL-SDR specifically but by introducing one of the first open-architecture SDRs, the USRP.
RTL-SDR is the cheapest imaginable commodity implementation of SDR technology, while USRP was... well, pricy enough to bring in some solid revenue and get his company bought. More power to him, it was (and is) an incredibly useful platform.