search

cnlohr / esp8266oddclock

What happens when you rip the ESP8266's stable clock source out from under it's feet?
81 stars 13 forks source link

So, the channel width streches? #1

Open Crypter opened 5 years ago

Crypter commented 5 years ago

Are you telling me that we can center or channel 8 (2.447), have channel width of 100MHz and effectively use the entire 2.4GHz band bandwidth all at once?

cnlohr commented 5 years ago

I don't have an SDR to test that, I only tested narrower. I think think 100MHz is possible at all...

The highest theoretical baseband you could use is 1040 / 6 = 173.3333 MHz (overclock) / 80 MHz (Base) = 2.16 (multiplier) * 20 MHz (default band width) so ~43MHz. It is possible the divisor could be upped (i.e. processor overclocking without telling the rest of the system) and MAYBE you could open it up wider, but I can't imagine the hardware working well for it.

Do you have an SDR to do more testing with this???

Crypter commented 5 years ago

Ahh... No, I don't have an SDR that goes up to 2.4GHz, only a regular RTL-SDR, but let me check among my HAM friends. But what we can measure is signal-to-noise ratio relatively easy. We can try to set it as wide as possible on channel 6/8 and make it broadcast junk continuously. If we manage to worsen the S/N ratio on all channels at the same time then I think it works.

And if only the input filters on the radio allow receiving of that same full bandwidth at the same time...

On a side note, I'm curious if we can use the radio to generate signal on arbitrary frequency, or maybe even abuse the hardware as an SDR. Some Broadcom chips can do this.

cnlohr commented 5 years ago

I am very, very curious about this now. I do intend to experiment with the RFPLL soon, but don't expect much.

Without information from espressif, it's unlikely we'll see SDR8266 any time soon.

MAVProxyUser commented 2 years ago

does this work down to 5mhz channel spacing?