Operating Broadcom Wi-Fi Chips as Arbitrary Signal Transmitters, Like SDRs

Secure mixnet using Android devices would be very cool.

> Matthias Schulz. Teaching Your Wireless Card New Tricks: Smartphone Performance and Security Enhancements through Wi-Fi Firmware Modifications. Dr.-Ing. thesis, Technische Universität Darmstadt, Germany, February 2018.

might be coming soon. Feb 2018 is pretty recent, and publication is sometimes delayed due to procedural issues. However, it might just be "published" by having a number of copies in the university's (as well as the state's and German federal) libraries.

I think this only shows the insanity of FCCs statement, as the Broadcom firmware _is_ locked down and got hacked either way. If somebody wanted to transmit arbitrarily on this bands he or she could just use regular SDRs.

Yeah that's not a strong argument at all. The 2.4GHz and 5GHz bands are known widely as "garbage bands" due to how terrible they are to work with (e.g. rain will heavily attenuate your signal, they're literally how microwaves work to cook your food, etc.).

Moreover, there are plenty of radios on the market that transmit all kinds of noise on these channels that isn't WiFi (e.g. my neighbors had a baby monitor that I could tell when it was on by how shit my WiFi was at that moment, FPV drones somewhat frequently use this band with proprietary analog video transports, etc.) This is permitted because these channels are "unregulated" (or, more realistically, "regulated for unlicensed operation").

On top of that, the FCC's official statement on this very topic was that what they really care about is the amount of power being put into these channels, as some aftermarket router firmwares of varying repute would let you crank the output wattage above FCC regulations for these frequencies (<1W power to stay within unlicensed operation regulations for 2.4GHz), and apparently some in-market WiFi routers picked up those firmwares as a "supported" option (I think ASUS and Linksys were both named). Most of these routers have since patched themselves to stay regulation-safe.

So, go nuts, just stay below 30 dBm/1W from the transmitter in 2.4Ghz channels or prepare for a visit from the FCC. (You'll have to do your own research for 5GHz).

Does it? I feel like it would actually have to cause interference before the FCC's position was actually justified. That seems unlikely considering how pervasive and interference tolerant 2.4GHz devices are.

Yes, Wi-Fi bands only.

RF compliance is really complicated in mobile devices, so there's all kinds of filters to prevent out of band emissions from being generated. The analog parts like the amplifiers are also precisely tuned to only use certain bands.

Is this a limitation of the antenna, or the hardware itself?

This particular hardware may be limited to those frequencies but I believe the Mediatek combo chips that do WiFi+BT+GPS+GLONASS+Beidou+FM as well as GSM are capable of a much wider range. WiFi and BT are at 2.4GHz but GSM/UMTS/LTE reach up to 2.6GHz and down to 800-900MHz, and the various navigation systems use frequencies in the 1.x GHz range. From what I've read, only the FM hardware is different due to its vastly different frequency (76-108MHz, although the hardware could probably go beyond those ranges slightly) but everything else is based on the same SDR.

What's the missing half?

Why is that “bullshit”?

The license is MIT but not MIT. It’s the MIT license with two additional conditions imposed;

- The one you mentioned about citation.

- “The Software is not used by, in cooperation with, or on behalf of any armed forces, intelligence agencies, reconnaissance agencies, defense agencies, offense agencies or any supplier, contractor, or research associated.”

https://github.com/seemoo-lab/mobisys2018_nexmon_software_de...

Even just the one about citation is problematic but IMO the condition about armed forces and so on is even worse. I kind of see where they are coming from but I wish people just used standard MIT and acknowledged that yes your software might end up being used for something that you do not agree with morally.

And if they are going to impose such restrictions, why stop there? Why not also say that you cannot use the software to transmit for example child pornography? And how about saying that you cannot use it to spread false information, fake news etc?

I would be exaggerating, but not much, if I were to say that as soon as you introduce one or more moral restrictions into a software license, you are implicitly saying that any moral concern not mentioned is ok.

How about terrorism? They didn’t say anything about terrorism so I guess using their software for terror is fine as long as the terrorists are not an “armed force”, an intelligence agency etc.

Technology has infitinite uses, a whole host of which are good, probably as many that are bad, and likewise an uncountable amount of uses that are either neither or both, and very many that will be one thing for some people and the other for other people.

It should in my opinion not be the job of a software license to pass moral judgement. Either you release your software for anyone to use for any purpose under the terms of an unmodified license accepted by the community or you might as well not bother trying to be open source at all.

It's their work, it's their prerogative to select the license. If you don't like it, leave it.

Or get a ham radio license and go wild (within the limits of that license)

RTL-SDR is RX only.