FCC Introduces Rules Banning WiFi Router Firmware Modification

For years we have been graced by cheap consumer electronics that are able to be upgraded through unofficial means. Your Nintendo DS is able to run unsigned code, your old XBox was a capable server for its time, your Android smartphone can be made better with CyanogenMod, and your wireless router could be expanded far beyond what it was originally designed to do thanks to the efforts of open source firmware creators. Now, this may change. In a proposed rule from the US Federal Communications Commission, devices with radios may be required to prevent modifications to firmware.

The proposed rule only affects devices operating in the U-NII bands; the portion of the spectrum used for 5GHz WiFi, and the proposed rule only affects the radios inside these devices. Like all government regulations, the law of unintended consequences rears its ugly head, and the proposed rules effectively ban Open Source router firmware.

The rules require all relevant devices to implement software security to ensure the radios of devices operating in this band cannot be modified. Because of the economics of cheap routers, nearly every router is designed around a System on Chip – a CPU and radio in a single package. Banning the modification of one inevitably bans the modification of the other, and eliminates the possibility of installing proven Open Source firmware on any device.

212 thoughts on “FCC Introduces Rules Banning WiFi Router Firmware Modification

  1. I am getting fed up with these corporate shrills… Is there a kill bucket for these people one can contribute to?
    Not working that is there anything i can sign to? A petition? Anything?

    1. Oooh the 1% better watch out. This guy is so fed up, he might sign an online petition while he’s taking his morning dump. Or worse yet, he might throw in a couple bucks. The winds of change are blowing. I can feel it. The revolution is near.

      1. Does being that much of a repulsive douchebag physically hurt, or does it just come naturally to you?

        This guy is upset by the way things are going, as he damn well should be, and he wants to know if there’s something he can do about it or do to help, and that’s your response?

        You are an insufferable human being. I hope you suffer from uncontrollable rectal itch.

        1. @dnm – thank you for the much-needed troll put-down. Somebody asks for an FCC comments link and the d-bag troll answers with mockery, probably by smartphone from his remedial high school civics class. Wish I had a comments link to publish here.

      2. Trying to bring about change by opposing lobbyists’ interests is so completely uncool right?! We should all just stay quiet and do nothing! (/s)

        Your attitude is part of this sort of toxic learned helplessness that’s so disturbingly in fashion these days. It lets you be used as a tool to make this sort of horrible legislation stick by acting as a thought terminating meme that self silences dissent.

      3. Fucking imbecilic, do nothing ass wipe. How the fuck do you think the 1% get away with fucking the 99% around? Because of lazy fucking, laughing boy, do-nothing, jerk-offs like you. Let’s see how funny you think this is FUCK WAD.

    2. From the down link:Take Action Now!The FCC is asking for comments on this proposal. The most important thing you can do is comment on the FCC’s proposal and tell them you want to be able to control your computing devices. Will you do this?Comment deadline extended to October 9.Instructions:1. Go to the Federal Register and press “Submit a formal comment” (https://www.federalregister.gov/articles/2015/08/06/2015-184…)2. Start your comment by respectfully asking the FCC to not implement rules that take away the ability of users to install the software of their choosing on their computing devices. Additional points of emphasis you should consider adding:- Wireless networking research depends on the ability of researchers to investigate and modify their devices.- Americans need the ability to fix security holes in their devices when the manufacturer chooses to not do so.- Users have in the past fixed serious bugs in their wifi drivers, which would be banned under the NPRM.- Billions of dollars of commerce, such as secure wifi vendors, retail hotspot vendors, depends on the ability of users and companies to install the software of their choosing.3. Enter your name and address. This is a public comment and your personal information provided will be publicly available.Once you’ve submitted your comment, make sure to encourage others to submit comments opposing these restrictions on computing devices. Use the #SaveWifi hashtag on Twitter or your favorite microblogging services.Google cache: http://webcache.googleusercontent.com/search?q=cache:https:/

      From https://news.ycombinator.com/reply?id=10141167&goto=item%3Fid%3D10139679

      1. The LAST TIME they asked for comment was when they came up with their own version of “Net Neutrality” which really was a way for them to seize control over the Internet, while pretending it was about keeping the Internet the same as it has been.

        I don’t know about the rest of you, But I have seen massive amounts of censorship and attempts at control on the Internet.

        1. That’s funny, I thought net neutrality meant that they would regulate it (as is done with other aspects of trade and commerce) in order to prevent certain market players from gaining and then exploiting an unfair advantage. But I could be wrong…

          1. Your correct, the people advocates for Net Neutrality, while the FCC faught it, and instead tried to implement a fast lane, and tiering, ending what little neutralith existed. The FCC director used to be a cable lobbyist, and was appointed by Obama who received campaign funds from his previous employer. The web is vast, and the hole deep.

            Just sign the petition and hopefully with enough vigor and defense of strength, the truth and what’s right will win over.

      2. We could also argue the point that the obvious purpose of such restrictions negatively impacts source recycling in that obsolete routers could not be upgraded or repurposed as new incopatible standards are adopted, and millions of routers will find their way into landfills to leach toxic metals into the ground water for decades.

    3. Buy an English book and sign the front page in case you lose it like you did the last one. The word is shill not shrill. Then there’s “failing that is there anything etc.”.

    1. If you had read the whole thing, you would have gotten to this:
      “nearly every router is designed around a System on Chip – a CPU and radio in a single package.”
      Which is saying that in many cases the radio IS in fact the whole router.

      1. Yeah but in such SoCs, despite being the same chip, the radio and apps CPU are separate CPUs.

        It’s possible (and in fact VERY commonplace already) for an SoC that has an applications CPU and a radio baseband on the same SoC to allow the apps CPU to be “unlocked” while the radio baseband still enforces code signing – look at ANY mobile phone that has an unlocked bootloader and a Qualcomm MSM-family SoC for example. You can modify the Linux kernel on these, the Android userspace – but if you attempt to modify the baseband images, those will fail signature checks and the radio won’t load.

        I’m also 90% certain that attempting to modify the WCNSS (wlan) firmware of a Qualcomm SoC will fail.

        1. I had thought the same thing when I first heard about this however they specifically call out wrt in the document.

          “What prevents third parties from loading non-US versions of the
          software/firmware on the device? Describe in detail how the device is protected
          from “flashing” and the installation of third-party firmware such as DD-WRT.6”

          1. Truly time to bury the term ‘land of the free’ once and for all for americans. Just go silent during that bit. And you might as well skip the next line too, let;s be realistic about what we’ve seen.

        2. +1

          This might make chips slightly more expensive, but not a lot, because instead of some RAM in the radio part of SoC there is going to be PROM with code burnt in the factory and a “US” (or EU) suffix engraved on the chip.

          This MAY cause several side effect some of which I could consider beneficial.

          1) Smaller lots of chips result in higher price. BAD.

          2) Really read-only characteristics of the radio chip.

          a) Improved security thanks to inherent integrity of a device. GOOD

          b) Bug fixes made impossible. BAD

          As far as it goes the outcome seems worse than better (2:1). However, iff the market works as the neoclassical economics claim, the quality of the firmware should improve because OEMs will seek for chips that have less bugs leaving the factory and not hope for future updates. Together with the improved security-by-integrity I would count it as double hit (2:3), hence in the long term the outcome may be GOOD.

          1. I guess most manufacturers will simply use a bootloader that verifies the code signature. Of course you can bypass this if you have JTAG/direct flash access, but from the document: “Describe all the radio frequency parameters that are modified by any software/firmware without any hardware changes. Are these parameters in some way limited, such that, it will not exceed the authorized parameters?”

            Security against hardware changes is not possible, you could just as well connect a power amplifier/frequency transverter to the antenna socket. No security will prevent that.

            The problem may be more on general purpose PC’s, in most WiFi cards the host driver has full control over all settings (except intel cards where the firmware does further verification). Atheros cards have no firmware at all. Some manufacturers could meet this requirement by forcing secure boot to enabled.

          2. I don’t suppose it matters which acronym it has in terms of untouchable NSA snippets in there huh.

            And when found out It’ll be: it was the Chinese! OH LORDY the Chinese are hacking you!.
            Followed by the crowd: “Did you know the chinese are hacking you? My trustworthy news said so and now I’m an expert”

          3. @Dodo

            Please remember that a Turing machine (CPU controlling the radio) can do much more than adjusting output power levels (e.g. transmit things without you even noticing it) and I would like to avoid it. With the firmware in PROM (it would be good that the memory is readable to verify the contents of the firmware) I have some basis for TOFU and one less place to plant OS-flashing-surviving “rootkit”.

      2. Allot of routers use seperate 5GHz chipsets anyway, take a look at the TP-Link Archer C7 as an example, the main SoC includes the 2.4GHz radio, but the 5GHz radio is provided by a Mini PCIe card, thus this rule would not apply to the C7.

        Theres always an alterante route anyway, theres plenty of “kit” routers available that provide radios via Micro PCIe slots so again, rules do not apply here.

    2. Paul, go back to 3rd grade and pass your English Comprehension before embarrassing yourself again.
      It *clearly* states that if the radio is designed to be firmware locked, the entire system will have to be firmware locked. Otherwise you could just overwrite it. Only if the device had an entirely separated out radio could it be locked down separately.

      1. The entire system must be firmware locked, even if the radio is separate. Actually, even a laptop must have locked software, (ergo “Secure Boot” set as “mandatory”) according to the rules, if the laptop contains a U-NII module.

        Note how a FCC granted manufacturer of a network card must “describe how the module grantee ensures that hosts manufactures fully comply with these software security requirements for U-NII devices.”
        Eg, a manufacturer of a FCC granted network card, that sells the network card to a laptop manufacturer, must ensure, through Mutual agreements, (NDA’s, usage agreements, terms of service), that the laptop manufacturer DO LOCK DOWN the software on the laptop to ensure the driver software are not replaced.

        So this might mean that Secure Boot is set to mandatory, and that Secure Boot is the loaded with manufacturer-specific keys, (not microsoft keys), and the computer must then be sold in a state where the customer does not have full administrator rights on the computer, so he cannot replace the driver software.

        So this will lead to that we MUST send the computer in to manufacturer service and not a third-party service if the computer gets a virus and needs to be reformatted. So the manufacturer can charge high for this.

        1. Conversely it might be an expense for the manufacturers to ensure there product complies with FCC rules despite virus vulnerabilities – think of car manufacturers and product recalls model.

          1. I like the analogy.

            But don’t think for a second we don’t all pay ($) for the recalls. With automobile safety, I lean towards regulate. With computers, um…no.

      2. No, you couldn’t just overwrite it if the radio enforces code signing at boot but the applications CPU allows unsigned code to run.

        Example: Any mobile phone with a Qualcomm MSM chipset and an unlocked bootloader for the “apps” CPUs. Mangle the kernel all you want, but code signature enforcement is in play for the WLAN and baseband firmware already (which are dynamically loaded as blobs by the apps CPU at some point in the bootup process).

      3. It’s clearly stated, but not clearly correct. As Entropy512 has explained in another comment, many SoCs can have different levels of code-signing for the baseband and application components, so “inevitably bans” is not necessarily true. When the premise is false, any conclusions about the scariness of the proposed rule are also suspect.

  2. FCC is only something in the states, right? So in the rest of the world we are probably not going to notice a lot. These cheap routers probably simply won’t be sold anymore in the states, but will still be fabricated for the rest of the world?

    1. When a company designs a new product, which would be cheaper:
      – Design one single product to be sold everywhere – including USA. Therefore, firmware locking implemented to comply with FCC, even though it’s not required in EU/Asia/everywhere-but-usa.
      – Design two products, one for US and one for non-US.

      Why double your design/build/test costs?

      A few years in the future – since every product is now locked then other countries can follow along and implement the same laws.

      1. If anything this ban could be death sentence for the tech industry of the US as rules like this will cripple it and cause it to become less and less relevant on the global stage.
        US tech companies already got a black eye from the NSA spying.

      2. Many devices are now made US and rest-of-the-world separately
        Samsung and such even puts different CPU’s and/or screens in their phones for the US vs other regions, but it’s not just phones.
        So yeah it can be done, and is done already, so we’ll see what happens in this case.

        1. Yeah, it sucks. There are so many really nice phones that EU and Asia get but the North American version has less memory, slower CPU or some other feature that’s smaller or not in it at all.

    2. Not really since a manufacturer isn’t going to make 2 products, they will make one product that abides by all the rules and therefore you would be affected by changes in security of the devices

    3. Don’t worry, the EU will follow soon. All governments love restrictive laws, especially those that take away common people freedoms if measures can be easily justified by the usual lies (war on terrorism etc.).

      1. “Don’t worry, the EU will follow soon.” … definitely!
        And they’ll sell that as innovation to the uni(n)formed public because this will make fighting terror easier!

        Quiz: Who said “Verschlüsselung ist die Sprache der Terroristen”?

  3. If your router is modified and does not affect any radio communication capability of another router unmodified, then this is nothing more than a corporate protection act. The 5Ghz range is open and unlicensed aside from weather and military applications.

    from WikiPedia, but mostly accurate…

    In 2007 the FCC (United States) began requiring that devices operating on 5.250–5.350 GHz and 5.470–5.725 GHz must employ dynamic frequency selection (DFS) and transmit power control (TPC) capabilities. This is to avoid interference with weather-radar and military applications.[18] In 2010, the FCC further clarified the use of channels in the 5.470–5.725 GHz band to avoid interference with Terminal Doppler Weather Radar (TDWR) systems.[19] Yet another revision adds 160 and 80 MHz channel identifiers, and re-enables previously prohibited DFS channels, in Publication Number 905462 (10 June 2015).

    So, unless you’re going out of band, or infringing upon military communications, this new ruling should not apply. the military encrypts any sensitive data, and if you think your little router can interfere with doppler, I got a bridge for you.

    More regulation where none is required.

    1. Actually, you’d be surprised the amount of impact these UNII systems can have on TDWR systems. Not particularly the consumer end UNII client, but the provider end. These are high power, and also have cheap knock-offs and clones, as well as replacement firmware.

      I work in a radio sensitive environment (radio astronomy observatory), and actively participate in a spectrum users group, and was privy to a lengthy discussion between an FAA TDWR rep and the local FCC rep as to just how much interference they were subject to, purely due to “misconfigured” UNII equipment. Not to mention how much interference WE receive due to these things too.

        1. You would be surprised how much interference a single WLAN basestation operating on the wrong frequency causes to a weather radar. The radar is massively powerful, sure. However, the echo from the clouds is rather weak. A 100mW WLAN transmitter with LOS will completely blank out a sector from the weather image. You typically get caugt very quickly if you are the cause of this…

          Check page 7 of this report for a screenshot of a weather radar under WiFi interference: http://www.bipt.be/public/files/nl/1740/3033_nl_mededeling_5ghz-meteo_nl.pdf

        2. For us as in the observatory? Yes. We routinely (approx. once per month) have to run around in our tracking van pinpointing 5GHz WISP sources, often miles away (by LOS, not driving). They like to tweak their setting to step 5-20MHz up or down from the allowed band, landing right in the middle of some important astronomy frequencies.

          Heck, we recently lost a significant portion of one of our most critical bands, L band, thanks to a 1.6GHz signal coming from Light Squared, which isn’t supposed to be broadcasting on that band in this area. Things like this make us the bad guy, because your average person wouldn’t notice these things, or even be bothered by them, but we have to take a stand if we want to continue our work.

          For the FAA and their TDWR? From what I could gather in the Spectrum User’s Group, it’s a problem, and one that is scary for them. These issues also crop up in other equipment due to harmonics and trashing noise floors and such. And that is serious, because comms issues for a busy airport can be very dangerous, potentially fatal and disastrous.

          1. ajford,

            Your complaints are very legit. I personally think that the FCC is going about this the wrong way though. I think they need to do emphasis patrols and start issuing fines on violators. There are so many “hobbyists” that violate the rules with an “I don’t see how that can effect someone so I’m going to do it anyway” attitude that there is plenty of extra noise real users have to deal with.

          2. Light Squared aren’t using consumer gear, of course – they’ve got licensed frequencies well away from any of the consumer stuff, which means their kit almost certainly has the capability to broadcast on frequencies they’re not legally allowed to and they’re just trusted not to screw it up.

          3. @makomk Yes, Light Squared isn’t using consumer gear, and yes they use bands in different areas, but the problem we were facing is the fact that in my area (Puerto Rico), they don’t have the license to broadcast at the frequency they are doing so. They are licensed to broadcast at a different frequency, and to receive (again at a different freq).

            The frequency they are broadcasting at is one of their standard frequencies, but they are not licensed in this area. Hence the problem. I probably shouldn’t have brought up Light Squared since they aren’t relevant to the discussion, but it falls into the realm of transmitting without licenses, even if this isn’t due to hacked/cloned hardware/software.

      1. Well that’s great, but then why the blanket rulemaking instead of selective proscriptions, such as those in place at Green Bank and FCC monitoring locations? 5Ghz doesn’t go that far(less than LOS, depending on environmental factors). A better solution would be preclusion zones, much like those already in place, on emissions of whatever type on this band.

        1. The problem lies in people doing bad things with the HW without consideration for others. How does a preclusion zone differ from band limits? The problem with modifying firmware/clones is that people will step outside their limits just because they can. “The band is crowded? Ok, I’ll just go 5 MHz low. None of my competitors are there”. You can put these limits on, but can’t enforce them without a big stick.

          As for the distance, yes, a consumer device can’t pick up your 5GHz signal that far away, but I was speaking of non-consumer devices. The receiver of a TDWR is going to be FAR more sensitive than a USB dongle or some embedded receiver. They have to be. Imagine trying to pick up individual water molecules in a cloud tens of miles away!

          This is not to say I’m for this rulemaking. I am solidly against the FCC in this instance. I was just pointing out that these devices can have a significant impact. I proudly run DD-WRT on all my routers and fiddle with SDR when I can. The real problem is the idiots doing things they aren’t supposed to (ignoring frequency/power restrictions, etc.). They are ruining it for the rest of us.

    1. They are probably aiming at people who like to changes the power settings on their routers or use frequency bands of other regions of the world that might have conflicts in the US. Some like to think that more power is better.

      To be honest, I actually turn down the radio power settings on my router. As long as the signals strengths in my apartment is fine, there is no needs for it to reach the far end of the hallway.

      1. …and better antennas help to turn down the TX power even more without losing range while increasing RX sensitivy. “Better ears” sometime are more important than a “louder mouth”…
        ;-)

        …but we’ll never stop the “louder is better” mentality of the gawks.

    1. I, for one, would like to know what a partisan committee of non-elected individuals with a history of pandering are proposing BEFORE it goes into effect so I can say “NO!!!” on record when those regulation will effect my life in any way, shape, or form.

      I sure as hell didn’t get a letter from the FCC letting me know they what they were planning.

        1. That’s more accurate, except a few of the companies. Despite the privacy concerns, Google has been the most end-user friendly company we’ve got. They aren’t pushing garbage legislation like this. At least not to my knowledge.

  4. I realize that the FCC doesn’t want modding devices to make them hit illegal power levels to be any more trivial than it already is(given that most routers aim for omnidirectional coverage, even a simple directional antenna may be enough, never mind RF amps and the like); but what I don’t understand is why it is necessary to control all the software, rather than just the output power.

    It is my understanding that the use of that part of the 5GHz band is not contingent on correctly implementing any specific standard/standards; so simply spewing static, while pointless and kind of a dick move, would be as legal as spewing 802.11 frames. What is it to the FCC what software is running, just so long as the output power is capped at a legal level?

    Plus, it’s not as though devices that come from the vendor with illegal features(notably the deauth-attack ‘network management’ tools that have attracted a couple of hefty fines recently, which are a standard feature of most campus/’enterprise’ wireless systems) seem to have much trouble getting certified, so it’s not as though there is a big black-market for hacked firmwares that provide the cool tools that the legitimate vendors won’t.

    I’m sure that a lot of routers, at least currently, leave power/amplification entirely fiddleable in software(with some risk of toasting your hardware), which the FCC might object to; but so long as the output power is fixed, who cares what the software does?

    1. Exactly. The FCC only needs care about the RF and licensing. That’s all. Anything prior to the emission on the device is irrelevant, so long as it meets the technical standards.

      1. I changed the firmware in my ECU as the stock firmware so bad it complained about a simple exhaust and cold air intake upgrade.
        Another car of mine the entire stock fuel system went into the trash and was replaced by after market.
        In both cases the drivability is much better.

      2. I’ve never heard of that. There is a huge industry of reflashing or replacing automotive ECUs and if it’s illegal it’s never been enforced because many huge companies openly advertise and sell their services.

    1. “You buy a house, add a garage, no problem.”
      After you get zoning approval, building permits, and possibly the okay from your HOA. Wait depending on the town and or city you may other approval if you are in a Historical district…

  5. Well if the manufacturers made good FW, we wouldn’t have a need to flash custom FW…
    Who gave me dual WAN capability (with auto switch to backup when the first lost the connection), full USB over IP, good QoS etc. for some good price? No one…

    BTW OpenWRT is compiled to perform FCC rules (e.g. They use mainly power and frequency limits from EEPROM of the wireless chip). Of course you can compile the firmware from scratch with “Reghack”…
    But also you can select in FCC proved router another country and violate the rules…

  6. Actually, this applies to the whole device. The article is wrong that it only applies to the radio. If the device contain a radio capable of acting in the 5GHz band, then firmware Control must be secured for the whole device, even if the parameters for the wireless Communication is locked in hardware. Note that even if you as a device manufacturer can certify in point 2 that the radio itself do contain hardware locks preventing the firmware from operating the device outside of permitted ranges, the device still needs to comply with firmware security.

    The only exception, is when a device is a radio module designed for be used with a “host” – for example a USB Network card, or a PCI network card, then the parameters must be hardware-enforced. No more security is required if the network module is sold to end customers.

    HOWEVER, if the network card is sold to , for example a laptop manufacturer, that incorporates the network card for example in a laptop, the manufacturer of the network card must ensure (through agreements) that the software on the laptop is locked.
    This might include setting “Secure Boot” to mandatory. This in *addition* to the hardware-enforcement.

    See this:
    “For Certified Transmitter modular devices, describe how the module grantee ensures that hosts manufactures fully comply with these software security requirements for U-NII devices. If the module is controlled through driver software loaded in the host, describe how the drivers are controlled and managed such that the modular transmitter parameters are not modified outside the grant of authorization.”

    Note how “Hosts manufacturer must fully comply with the security requrements”. Ergo: The laptop manufacturer must ensure the software (Windows 10) of the host (laptop) the “certified transmitter modular device” (network card inside laptop) is incorporated in.

    So if this come true, we will have to see laptops with “Secure Boot” set as mandatory with no recourse of reinstalling or reformatting the drive without permission from the manufacturer. Thus the manufacturer will be able to charge a very high fee for this.

  7. I have given up being a law abiding citizen a decade ago. I hack devices and violate the DMCA all the time (I violated it this weekend with hacking a Wink Hub) this worthless law will not stop me from running a secure firmware that protects me.

  8. It’s easily solved.

    I buy from China and India markets. They’re mostly guaranteed to work well with most things, and they are cheap to boot.

    Unfortunately, what the FCC (and FAA, and FDA, and etc) are doing is making us a weak country in terms of our technological skill and savviness by hand-holding everything they can. Doing so makes us weak.

    The FCC did this.
    The FAA has retarded drone and UAV laws that restrict even taking a video and putting up on youtube. Other countries are being sane, while FAA is dragging knuckles and being completely dumb.
    FDA retards growth by biohacking industries by requiring >$1e9 to fully “vet” a drug. It makes sure that maintenance drugs are made that make your jimmy bigger, yet other advances are ignored or forbidden.
    FDA also smacks drugs on a “drug list”, where I need to convince a doctor to prescribe for me. Now, say I want some nootropics like Provigil. That’s right, nope. But even if I do, the costs are insane.

    The answer is: do it anyways. And buy from your trusted Indian and Chinese businessman. Their stuff isn’t crippled.

    1. “Do it anyways”

      That’s the spirit.

      Unfortunately, in the long term this won’t suffice. The so-called “trade agreements” (TTIP, TIPA, CETA and how the all are called) are the means to ensure that tomorrow, you won’t have access to your “trusted Indian and Chinese businessman”. Most probably they won’t exist for long.

      And no — it’s not “big govt”, as someone is putting it: it’s “big corp”. The govt is just an instrument in their hands. Reps won’t fix that (as won’t Dems). Wimp or Shrimp.

    1. Vote conservative to eliminate bloated bored gov’t entities that have nothing to do all day but dream up stupid bullshit like this. This is a typical symptom of too big gov’t. Yeah yeah… the republicans are for big business.. bla bla bla… but the left has never met a regulation they didn’t like, especially if it puts a boot on the throat of any business, person or entity that dares thinking outside the box or, gasp, doing something without unnecessary gov’t control.

      Elections have consequences, remember that the next time you hear someone telling you they’re gonna give you something for free.

      1. What a silly comment. Pushing one of the party levers effectively supports all laws, regulations and values that the party supports. Consider applying some thought to your selections to align your vote with your values. That is, unless you unilaterally support diametrically-opposed policy that is adopted by a party to serve as a wedge.

        In this case, OP is encouraged to research the positions of politicians on open devices and networks, when available, or infer the positions from historical votes on related issues. Also, the FCC takes public comment on issues: https://www.fcc.gov/encyclopedia/rulemaking-process-fcc#q8

        1. Well, I already know who you voted for last time around. Putting your faith in comments made to an entity that has no accountability to the people making the comments is quite possibly the stupidest idea here yet.

          You want to kill a weed, you kill it at the root, not fuck around swatting at the leaves.

      2. There’s no conservative party in the US. There’s corporate with a smidge of progressive, and corporate without.

        I don’t count those, who on a day when they haven’t got a Medicare appointment but there’s some left from their last SS check, might hop a bus or drive public roads to go downtown to protest for smaller government.

  9. So… What? Will they come to my home looking through my router settings? It might get a wee-bit harder to reflash a soho router, maybe I’ll have to short a pin or something. That’s it.

    1. “maybe I’ll have to short a pin or something”

      My inexperienced guess is that, since most manufacturers will still want to provide for firmware updates, the SOC will have a non-replaceable boot loader that will contain code to verify/decrypt signed/encrypted firmware. I can’t imaging popular models will keep their secret long, but it would limit you to models that have been cracked.

      Worse yet, cheap manufactures could just decide that firmware updates are not necessary, and burn whatever bit locks the flash from being updated ever. That’s unlikely to be an easy fix.

      1. It will, but not because it’s illegal.

        Much like many modern flagship smartphones, they’re locked down to the point where it’s either extremely difficult or actually impossible to root and install custom firmware.

        Kicking back and ignoring this with a “I’ll flash what I want anyways” mindset is ignorant and short-sighted. It *IS* possible to lock down a device either to the point where it’s impossible or so difficult it’s not worth the hassle to crack. And routers in particular have a far smaller crowd of people looking to install custom firmware and root, so far fewer routers will get hacked.

        I get that “hack it anyways” is a motto here, and for good reason. But making it illegal and forcing companies to lock these things down is a terrible, terrible thing.

        1. The firmware change the FCC is interested in stopping increases the RF power output of the device beyond the FCC emission limit for unlicensed users in that band. This is easily gotten around without a firmware change. And open devices will always be available for an individual to bring in from another country.

          1. I understand that. However, locking down the firmware to prevent radio power changes will almost always involve simply locking down the entire device.

            As to bringing in devices from another country, that works if the manufacturer bothers to make multiple versions of the device rather than keeping their development costs down and simply making one locked down device for everyone. Aside from that, you’ll need to find devices that are entirely non-US. Possible, but it really limits your options.

          2. Most of those requires some hardware modding skills that only a small percentage of the population have.
            You can easily have a couple orders of magnitude more people that know their way around hacking poking at the firmware and yet another few orders of magnitude that knows enough to “flash” their firmware and use the webgui to change them after watching a few youtube video. It is the latter group this would prevent.

        2. I don’t know what you’re referring to wrt modern phones being locked down. I just have to hold down the volume down button + home button whilst I turn it on, and it has a loader to download custom code onto the phone. This is in EU though, so maybe that feature is disabled in US?

  10. None of these news stories bother me, and nor should they anybody else who checks this site, even be grateful for the new laws. Hackers only gain skills when they are challenged, and if software protection never changes, where are we supposed to learn new things and invent new ways of circumventing protection?

    1. “What does not kill us makes us stronger!” … is what the monocellular life organized in the form of a T-Rex said to itself as it looked up and saw a burning rock aerobreaking through the sky. And it was right. But it would never again organize itself in the form of T-Rexes.

      Cryptographic primitives and implementations will be broken from time to time. But in the long run it won’t be broken with a soldering iron, nor with debuggers. It will only be broken by doing math.

      Others think, that hackers will build and design cheap open source microfabrication toolchains (direct write systems, spin coaters, …) and derive independence by putting the individual in the post stone age, instead of just the society. But except for the first few systems, they will not be considered acts of hacking, but acts of building…

  11. 1: A few people are always ingenious and resourceful enough, so if software upgrading is an option, in time someone will hack it.
    2: write once only, Un upgradable hardware is nonsense in consumer electronics. Development and support will be so much more expensive and uncompetitively priced compared to imported hardware. One fuckup in production and a zillion chips goes in the trash.
    3: Everyone knows that a certain agency uses wifi routers for man in the middle attacks and general snooping of wifi and lan traffic. Consumers will have another reason to choose a non US manufacturer if they are security concerned.

    1. Good luck getting them into the country…

      I agree with you, but these jack booted thug assholes will stop at nothing. So you get the non US spec gear in, all of a sudden there’s a scan on your network from the ISP side, then low and behold they see non sanctioned equipment on your system, then there’s a knock on your door with an Eilian Gonzalez style raid to get the offending hardware off the network and you leave in handcuffs, or worse.

      Think I’m kidding?

        1. There’s probably a bigger market for drugs, than for hackable routers for the few geeks who actually know what firmware is. It’s an issue of them not even being manufactured. Easier to grow a field of poppies than to turn out your own VLSI chips.

          Then again, good ol’ FPGAs might be the answer, putting custom silicon in everyone’s hands. In that case, open hardware can spread as quickly as software.

      1. For individual consumers ordering internationally, there is very little they can do. On the other hand, you big box store have to import products that follows all the regulations etc. or their bulk shipments get stopped by the customs, so you won’t find it on the shelves.

      2. Your scenario is plausible for a situation where one person or very few people have ‘non-approved’ routers.

        The reality however is that ‘approved’ routers will be more expensive as it’s only the US that is introducing this BS regulation. Economy of scale dictates that the larger consumer market (the rest of the world) will have cheaper devices.

        So cheaper routers will flood into the US from other markets like China. So then the scenario is that we have a whole industry of SWAT team like enforcement agents, and this scenario is not at all plausible.

  12. what is the concern? is it that users will boost the range from the few feet to hundreds or thousands of feet?

    ok then dont modify the U-NII bands

    or maybe comes on the heels of the ProxyHam

        1. Yep. But you would agree operation above a certain power level on any wavelength requires licensing. With the maximum allowed transmitting power depending on wavelength of course.

  13. This is actually a really big deal. For those of you freaking out about your home routers, chillax. Your linksys running dd-wrt couldn’t affect a TDWR system unless it was in the same room. However, there is a huge number of independent businesses that have sprung up called WISPs that will live or die by this rule. They broadcast the Internet for miles using high end consumer hardware, and take about 5% of the customers from the big ISPs (and they don’t like it).
    Want to have an extra layer of anonymity between you and big brother? Buy internet from a wisp. Want to avoid every single mouse click being stored in NSA’s UT facility? Buy Internet from a wisp. Want to shut down thousands of small businesses, and force the masses to get their Internet from only the largest ISPs that pay billions to DC lobbyists? Create a “rule” completely locking out the use of UNI-II bands. This is a fight that has been going on for 15 years now and is only getting stronger. 802.11ac forced the FCC to open up a small portion of the very restricted space, but rules like this show they didn’t like it, and want more control.

    1. From my own research (when I was fighting a ~5.4GHz DFS problem on a medium-range PTP link), rogue radios are a very big deal to the meteorologists that use TDWR. As I understand it, they show up as a huge and unwavering dead/bright radial streak, and are dead-simple to locate: It’s already charted on a map.

      The unlicensed secondary user is then LARTed into compliance.

      My point, then, is multifaceted: I understand that third-party firmware is a way of life for many people/organizations/devices, and I make a point to always buy hardware that can be hacked/rooted/jailbroken (even if I have no intention of doing so). I accept that weather radar is important, and I understand that interfering with such systems is already problematic.

      But additional rulemaking won’t solve the problem: It’s already illegal to interfere with these services, intentionally or not. Additional rulemaking to make it harder/impossible to install different firmware doesn’t accomplish anything to further that goal. All it does is produce additional rules and more compliance checks, all of which will be tacitly ignored by those who wish to do their own thing.

      (In other news, I can easily buy wireless gear from Alibaba or DX or amazon.co.uk – thanks!)

      I remember the last time the FCC tried something like this, when they decided that all consumer unlicensed ISM-band gear with a separate, detachable antenna must have a “non-standard” antenna connection. The intent was to make it very difficult or impossible for an end-user to put together a non-compliant access point, but it didn’t do that. RP-SMA and RP-TNC connectors were deemed acceptably non-standard, and for a brief moment in time, they were.

      But it back-fired immensely. In no time, the market responded by making it easier to find “non-standard” RP-SMA and RP-TNC antennas and cabling for ISM-band gear than to find non-RP (ie, traditional bog-standard) versions of the same.

  14. The fallout from this (if it is put into effect) is to kill the 5GHz band for consumer products. The cheap hardware vendors will likely reshuffle bullet points on the back of the package to make its absence unnoticeable rather than deal with the cost and complexity of securing the radio firmware to get through FCC cert.

  15. So, let’s frame the discussion for those too lazy to read the document.

    Intention: Lock down all devices with a radio frequency emitter, such that they cannot be set to transmit and receive signals beyond the tolerances which the FCC set.

    Methodology: Lock down the firmware of any U-NII device, such that no open source software may be used on it.

    Notations: The FCC specifically cites DD-WRT as an unacceptable modified firmware. They also suggest that the hardware needs a methodology to lock the device based upon location.

    So, let’s be reasonable. The FCC wants to attack the idea of people using parts of the radio spectrum that they shouldn’t be. They also want to limit transmitter power, so that the spectrum isn’t polluted by individuals setting their devices to a high power and drowning out other signals. The paranoid bit of me also suggests they don’t want people potentially monitoring emergency or “critical” sections of the spectrum.

    The reality is that they are using a hammer and chisel to solve what they should be attacking with a piece of 600 grit sand paper. What will the response be from manufacturers? Will they suddenly invest in better firmware for their devices? I think not. What we’ll see is either a decent device crippled by crap firmware, an expensive device that is priced higher because it has a minimally competent firmware, or a device that costs an insane amount of money just to get the features we already have.

    Maybe we could review some history. Linksys decided to use an open source firmware, which required them to release the code. When they were caught using the firmware, it took threat of legal action to get the firmware released. Do you honestly believe every manufacturer out there will develop their own firmware? Eventually we’ll get someone who tries to sneak an open source firmware into their devices to save a buck, and they’ll get caught. Firmware must be released, and some enterprising hacker will find a way around the BS firmware locks. The next week those routers won’t appear on any store shelves, and suddenly there will be another WRT-54G. What did the FCC solve then, that the lawyers won’t be able to tear down with something as simple as licensing agreements?

    Perhaps we can do one up on the FCC, and instead of just pointing out how their ideas retard development we can offer a solution? I don’t have it myself, without functionally fusing the transmitter power and transmission frequencies on the chip. It makes sense for the FCC, gives us our flexibility, without functionally harming anyone. I’m more leaning to simply killing this proposal though. Heck, Intel should be on the hackers side for this. The internet of things would be crippled if this was enforced. Every transmitter having enough hardware and software locking to make them “secure” would push the price well past reasonable. There’s no better way to kill an entirely new market than pushing it out of the hands of those who would develop it, by increasing the price to insane levels.

      1. The world isn’t just the USA. It would likely be like the boondoggle that happened with PGP. It got to point where people were smuggling copies of PGP source code in books because of a loophole. Imagine if you had to disclose the source code in France and a French person uploaded it to a torrent site or worst yet, posted a magnet link. XD

    1. “So, let’s be reasonable. The FCC wants to attack the idea of people using parts of the radio spectrum that they shouldn’t be. They also want to limit transmitter power, so that the spectrum isn’t polluted by individuals setting their devices to a high power and drowning out other signals.”

      this is as much as ‘problem’ as US voter fraud. ie, its not a problem and its a lame excuse.

      there are probably .01% of the users monkeying around with power and rf settings.

      no, there is no ‘user abuse of radio spectrum’ problem. this is just a lock-down of devices that used to be under our control but the trajectory is clear: our rights on our boxes will be reduced each year until we become renters and not owners.

      this is just a first step. you wait.

  16. This is very bad even though it only effects a single band as it means routers using it may not be secure.
    Security through obscurity just doesn’t work when there’s thousand or millions of copies of the same firmware out in the wild.
    Open source firmware means at least everything is documented and can be peer reviewed.

  17. The FCC mandate effectively eliminates the possibility of *end users* installing proven Open Source firmware on these devices. Router vendors for many years now have used open source software as part of their executable image, and do not necessarily have to open up their code to modification by the end user. It depends on the terms of software license of the open source packages used. Most open source packages in this space use one or more versions of the GNU public license, which do require that end users can modify and replace the image. Other packages are more permissive and do not require this.

    Router vendors like Cisco and Verizon got themselves into legal trouble in the early 2000s by including GPL licensed open source code in their software images. The vendors did not make the software available for end users to modify, a requirement of the licenses. These and other companies addressed the problems to be in compliance with the licenses and/or settled with the Free Software Foundation. Because of this closed source/black box mandate, however, the poster here have astutely recognized the potential for router vendors to “backslide” on their open source licensing requirements.

  18. As I understand, this only affects the 5ghz band on the routers. Modified firmware for standard 2.4ghz isn’t part of the scope. (No doubt once the companies dump a shit tonne of money into building firmware security it will get applied to the 2.4 units as well to distribute the cost).

    This might be naive of me, but wouldn’t it be easier to enforce the developers making dd-wrt/tomato/etc firmwares to have an amature radio licence? This way it’s on them if they cause interference, it should help protect other authorized users allocations. In Canada, to run a 5.4-5.8ghz WISP you need to be licenced anyways, not much difference there.

    1. I hate that report button…. 0.o like 2 lines of code in css guys, let us have an option!

      On point, I meant to say, this only affects routers capable of broadcasting on 5ghz not the band on the router itself.

  19. Wow, the whiners and complainers are out in force to cry their pathetic little eyes out about laws and rules that have been in force since 1936. The FCC hasn’t allowed anyone but licensed amateur radio operators to modify their radios for decades, why should the 5 GHz band be any different? Sheesh.

  20. As an ham radio operator having the ability to put 2.4/5 GHz equipment on ham-only allocations just with firmware changes has always been a great, accessible way to make abandoned routers useful again. Non-ISM type accepted stuff for these bands à la ubiquity et al are often steeply priced beyond the means of poor college hackers. I would be surprised if the ARRL didn’t get involved in this during the public comment period.

    1. Indeed… anything prior to the emission is incidental and irrelevant(even “good engineering practices” are irrelevant if the emission if within Part 97 spec). After all, you can build your own and operate 1.5kW as long as it meets Part 97 standards.

  21. I don’t see the point of locking DD-WRT out of routers. I’ve never seen such firmware modification that avoid regulatory limitations, or basically any router radio firmware mods at all.

    Do they really exist? Or is it just FCC thinking there has to be?

  22. I don’t understand what the fuss is about. The FCC is wanting to actually police it’s rules. This is for modifying the radio hardware. Any FCC compliant device cannot be modified and then still used unlicensed. It sounds like they want to put in place software to make it easier to police those rules. Good for them.

  23. I think I have an easy work around this:
    Use a USB / PCI-e wifi card with an open source OS (which have the driver for this wifi card of course).

    You don’t touch the device firmware and still get to play with it.
    On my device I can go in promiscuous mode, change TX/RX power (but it was fairly complex, had to change geolocation, and used a patched driver cause the regular one was FFC law friendly and basically couldnt get the fastest speed to work (something about channel width).

  24. The HaD article doesn’t mention, so I’m curious to know what the reasoning behind the FCC’s proposal is. My gut tells me that one ore more large companies that stand to lose money due to the modifications have lobbied to get this proposal enacted. Normally, the FCC does things to protect the public’s interest. …What interest are they protecting with this?

    1. I’d think, like nearly anything the US government does, it’d be for some corporate interest too. Except in this case the corporations are mostly in China, and that’s pirate-land, laws on tightening software security are not something they’d be interested in.

      If anyone’s interested, the current UK government acts directly for the interests of it’s millionaire friends.

    2. One interesting thing in the USA is the proposal by telcos to run 4G LTE in the unlicensed 5GHz band.
      They want your routers off the frequencies so that they can better use it to re-sell you your own connection.

      I want to be able to run alternative firmware and as USA is a large market, this rule will likely make my life harder.

  25. As a ham, as long as I have band privileges there(and they do exist in the 5GHz/5cm band), and my modification does not make the device violate Part 97, I don’t see how the FCC would have a leg to stand on. After all, I can manufacture a device from scratch in accordance to Part 97 standards and be good to go… at 1.5kW PEP, no less.

    All the FCC needs concern themselves with is the RF, what’s being carried on it, and if the operator is licensed. Anything prior to the actual emission in hardware, software, or firmware is incidental and irrelevant.

    1. You’re a licensed operator so this change would not apply to you. They are only interested in preventing an unlicensed operator from making an easy change to a piece of equipment that had been previously licensed by the FCC to make it non-compliant with said license.

  26. 1. Making something illegal does not make it practical or enforceable

    2. Nearly all routers are made in places where the FCC is just another barrier to export into US markets. Currently, if you have WiFi on a device it will need the FCC number for US import. This labelling issue is where the powerful WiFi kits originated with a lower signal test in the first place.

    3. Most WiFi cards use the ERP rating to appear more powerful, but still stay within regional/FCC guidelines thanks to software (default install).

    4. People who change the transmitter power discover the output section tends to overheat. Most manufacturers like to minimize cost, and will use the cheapest/lowest-power amp.

  27. Trying to push away all the griping and hobby-horses for just a moment – it seems that the FCC is anticipating that wifi is going to get alot more crowded, what with normal adoption rates and now IoT, and this could be seen as part of their effort to ensure that the wifi radios conform to the spec, which hopefully keeps the band as available as possible.

    I do agree that this might pose a problem for routers where everything including RF is running on one chip, and that it would totally suck for some of us to not be able to run DD-WRT. And I can see how this could be prone to abuse by some manufacturers or providers. And yes it could theoretically close off one avenue of firmware hacking.

    But let’s face it, 999 out of every 1000 routers are used by non-hackers who can’t even be arsed to put in a password unless it’s forced on them. Wifi connectivity is now a commodity and the RF spectrum is a public good, like roads. The busier they get, the more control is required…

    Anyway, I have enough old wifi routers and ESP8266s lying around to keep me hacking for the next 5 years. My nearest surplus store still sells old routers by the pound. I expect that hackable wifi systems will continue to be discovered and shared on HaD, and yet there will be no jackboots or loud knocking at 5 AM.

  28. Since the FCC only takes action when some little birdy (corporate interests) tweets something in its ear, this is what they may be thinking about:
    http://arstechnica.com/information-technology/2015/08/verizon-and-t-mobile-join-forces-in-fight-for-wi-fi-airwaves/

    Combined with the previous FCC actions regarding the 5 Ghz band that gave a little spectrum but made it and the existing 5.8 band almost impossible or overly expensive to use. It all points to a back door grab by the spectrum vampires and at the same time an attempt to eliminate the little bit of competition that they get from the WISP’s. Why else would the cellular companies choose the 5 Ghz band for this “new service” when they already own sooooo much other spectrum that is already part of every cell phone.

    The individual actions of the FCC are not too far outside of being reasonable but once you see the whole puzzle then it starts to stink of rotten fish.

    1. do you realize that the only ones who can MAKE chips that work in those routers are the same ones that are paying for this new ‘law’ ?

      who do you think is on our side? (answer: no one with any clout or power).

      the corps now run bartertown. master-blaster is not going to help us.

    1. SDR + LNB is not as cheap as a ESP8266
      ;-)

      The G3/G4 people are pissed as a study showed 73% of users tend to offload their data usage at WiFi hotspots.
      MiFi was a bad idea for many people, but telecoms have been gauging customers for over a century….

      1. The telcos would be pissed, too, if people actually used their wireless data plans.

        I’ve been around VZW long enough to remember PCMCIA AirCards, when 1X was a huge improvement (120kbps-ish!) over whatever dogshit-slow thing came before it. I remember when 3G rolled out in my town, and I got a pretty solid 2Mbps. I remember when LTE popped out, and I was seeing up to 25Mbps more often than not.

        Nowadays, 1X is so packed that it’s useless — the packet loss is intense. 3G is absolutely full of users, and barely works when it does work — instead of enough bandwidth to watch Netflix at low quality (which I used to do on 3G), I -sometimes- can get enough bits through the pipe to listen to an uninterrupted song on Pandora. LTE can be fast, and often is, but can also be ridiculously slow.

        There isn’t enough available spectral bandwidth for every man, woman, and child to participate in a long-range WWAN when a short-range WLAN is available.

        Meanwhile, an ESP8266 is no more an answer than an FOB-Shanzen router from Alibaba: While either can be made to work, neither addresses main issue, which is: We don’t need these new rules, and we (or at least the Americans amongst us) shouldn’t have to sidestep this sort of thing just to run the software we choose on our computers (even if said computer says “router” on the box it came in).

        My own primary router, an aging Asus RT-N16, is faster and has more storage and RAM, along with better connectivity, than all of the first several Linux boxes I had in the middle-90s put together: If it’s not a computer then I don’t know what is. That little RT-N16 is far more useful as a computer than the last-gen higher-end G5 24″ iMac I have in the garage, which I’d gladly unload on Ebay if I could figure out how to ship it safely: My old MIPS computer (ahem, router) still runs current software!

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.