Linux, *BSD, and other Unix variants have lagged in Wi-Fi support due to chip vendor's stated concerns about access to the low-level radio functions on their chips: But a meeting last month in London, the Linux Wireless Summit, apparently has helped move development along. DesktopLinux.com reports that the meeting included Linux kernel developers, and representatives from Broadcom, Devicescape, Intel, MontaVista, and Nokia. The summit is part of an effort to standardize parts of Linux for reduced maintenance and complexity, as well as greater functionality.
The summit's organizer is quoted and paraphrased as stating that the FCC will only certify Wi-Fi devices that have a closed-source component for handling low-level radio settings, such as frequency choice and power levels. I don't know that there's actual evidence as to this fact, and would love to see. That would be an extra-regulatory step for the FCC, as there is no defined required for releasing radios that cannot be modified; the onus is typically on the purchaser who modifies hardware conforming to regulatory limits, and suffering the penalties if they fail to conform.
For instance, worldwide 802.11a equipment can use the 4.9 GHz band in some countries; it's limited to public safety purposes in the US and military uses elsewhere. Using 4.9 GHz in some parts of the world could get you thrown into jail for a long, long time.
It's interesting that these considerations are now being made openly. A couple of years ago, I was provided with some of this reasoning from sources I won't identify, but told that the concerns about the FCC and other regulators couldn't be discussed publicly.
You can read some of this history in a January 2005 post that starts off discussing an Economist article criticizing Atheros and Broadcom.
Actually I think that the device can not be type accepted by the FCC if the end user can modify the frequency since it is essentially a commercial radio, the same antimodification rules that apply to the commercially available FRS, GMRS or even CB radios would apply here, unlicensed end users must be restricted from controlling the frequency and power of a type accepted radio device. There is a lot of precedent for this and it is nothing new at all.
I'm all for open source but they do have a point, they can't release the information to non-licensees of the FCC or they lose the type acceptance they need to sell the devices. An easy way to deal with this might be having someone who is writing the wireless driver be licensed as a GROL (radiotelephone operator) by the FCC, but they would still be creating an opportunity for violation of FCC regs by opensourcing the code to control the power and frequency of the transmitter which might still be a problem. Arguably as long as the end user did NOT modify the code they would be legal but the FCC needs to issue an opinion on this.
This really is an issue for the FCC to address one way or the other, but good luck getting a timely response from them.
I think I'm reasonably qualified to make these statements as an FCC licensed amateur radio operator who is required to know these things in order to operate a radio. Any ham or FCC Licensee would probably give the same answer, it's common knowledge in the radio community and goes for all kinds of commercial radios, not just wifi. Try not to take it as a slight against open source but a universal regulatory onus :)
73 de KB3NGB
-noah
rabid open source user/evangelist
[Editor's note: These are interesting statements, and I don't deny your qualifications. But it seems like the FCC's application of these sets of rules should be documented in some fashion, perhaps in FCC orders or written into the rules. I have never been shown or found the specifics that would tie the antimodification rules with Part 15 licensing. If it's an accepted thing, then the FCC would be documenting it in some fashion because they document EVERYTHING, even typical application of typical rules.-gf]
This story (http://www.arrl.org/news/stories/2006/06/29/2/?nc=1) is one of many incidents where the FCC has made arrests for the operation of non type accepted equipment on bands which require type acceptance.
Once the radio is modified by a non licensed technician it loses the type acceptance. It refers to a ham radio (does not require type acceptance because it is intended for licensed amateur use) that can be easily modified to operate on bands that require type acceptance.
Note the use of the phrase "easily modified" and how the FCC clearly states that equipment for use on type acceptance required applications must not be able to be easily modified by non licensees.
The example is dealing with hardware modifications but the same standard would apply to software as well -- it must not be "easily modified", and that includes by a reasonably technical user from the view of the FCC. The penalties, as you can see, are very harsh for violating these provisions as well. There are numerous other stories of this kind of arrest being made, on the FRS, GMRS, CB, ham and undoubtedly the parts of the spectrum occupied by wifi cards.
Even as a ham it is hard to get equipment to modify type accepted radios to work on our bands when we legally can!
This is actually an interest of mine, I have had to go out of my way to acquire equipment and software to modify radios I own which I legally can modify (motorolas and GE radios), because of the concern that non licensees could unknowingly cause harmful interference by making unauthorized modifications to commercial radio equipment.]
I hate to be a killjoy but unfortunately they are correct in their assertion that this is a legal issue properly decided by the FCC.
KB3NGB
noah derose
I agree that they should issue a formal opinion on this issue, but they don't exactly move rapidly on much at the FCC.
Regards,
-noah
[Editor's Note: They move slowly, but this issue has been discussed for several years, ever since the first SDR-like chips appeared. Which is a bit disingenuous, too, because the chips aren't really SDR in most cases, as I understand it; just much more flexible in the baseband than previous generations of chips. -gf]
For many old prism cards, the channels (frequencies) were a bitmap in the firmware.
And users had access to the firmware update programs.
Though there should be a line between users and developers - just because source is available doesn't mean it would be easy to change.
If they really need a closed source component, they could create a register-offset, content list file where they put the rest of the uploadable firmware although it wouldn't be for a foreign processor. Then the driver would load the magic FCC approved bytes to the right places to limit the power and frequency and get on with the connection.
Such would also be architecture independent (though endianness could be an issue if it isn't just bytes).