Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
United States Technology

To Save Lives, Issue Connected Vehicle Technology Waiver, NTSB Tells FCC (arstechnica.com) 81

ArsTechnica reports: In mid-August, the Federal Communications Commission succeeded in its long-held plan to reallocate a portion of the spectrum from car-to-car and car-to-infrastructure communication (known as V2X) to Wi-Fi instead. However, the FCC didn't reassign that entire region of bandwidth -- 30 MHz remains set aside for "intelligent transportation systems." And the FCC should grant automakers a waiver to allow them to start deploying cellular-based V2X (C-V2X) safety systems, said the National Transportation Safety Board in a letter it sent the FCC on Monday. The saga of V2X is a long-running one. The FCC originally saved the spectrum around 5.9 GHz for use with V2X in 1999, but despite keen interest from some automakers and industry groups like the American Association of State Highway and Transportation Officials (AASHTO) and the Intelligent Transportation Society of America (ITS America), the technology still has not been deployed.

Seeing that failure, the FCC decided in 2020 to reallocate some of the bandwidth to Wi-Fi, leaving the frequencies between 5.895 and 5.925 GHz for V2X. ITS America and AASHTO sued the FCC to prevent this, but the US Court of Appeals for the District of Columbia ruled in favor of the FCC in August, allowing the commission to go through with its plan. This has dismayed the NTSB, which has written to the FCC as part of the commission's public comment period as it considers a waiver requested by automakers to deploy C-V2X technology. Conceptually, C-V2X works the same as the older V2X -- direct vehicle-to-vehicle or vehicle-to-infrastructure communication but using cellular radio protocols instead of the dedicated short-range radio communication protocol. The FCC should grant this waiver, said the NTSB, which notes in its letter that it has recommended that the nation adopt wireless-based collision-avoidance technology since 1995. Connected vehicle technology would reduce the ever-escalating carnage on US roads, said the NTSB, and the agency also urged the FCC to make sure that Wi-Fi devices don't encroach on the remaining 30 MHz of intelligent transportation system frequencies.

This discussion has been archived. No new comments can be posted.

To Save Lives, Issue Connected Vehicle Technology Waiver, NTSB Tells FCC

Comments Filter:
  • by TWX ( 665546 ) on Tuesday August 30, 2022 @01:23PM (#62837137)

    After that recent disclosure that Hyundai not only used the same keys on their secure connections, but used the EXAMPLE keys common to student literature for those secure connections, I do not trust that automakers are ready for secure communications even within the CANBUS networks within their own cars, let alone car-to-car in a fashion that would actually make driving safer.

    Consider the possibility that automakers ignorant of the real needs implement insecure systems. These systems allow cars to notify each other of road conditions or safety problems. If a notification such as 'stalled vehicle' can be sent without the software generating the alert and the communicating it being secure, then an attacker could simply start squawking this all over, leading the vehicles that now are dependent on these systems abandoning routes or slowing to a stop to prevent collisions with vehicles that don't even exist. The sort of Hollywood scene of gridlock where an attacker actually takes down a city's roads would manifest, not from someone turning lights green and red to cause accidents, but simply because vehicles acknowledge and follow safety notifications that are bogus.

    And if the systems are as poorly implemented as I fear, the attack could come from anything capable of transmitting it, be it insecure vehicle computers that have been compromised, blackbox projects that are placed in areas, or if they still insist on using the spectrum that's been allocated to WiFi, possibly even compromised phones that support those frequencies.

    As much as I would like to see vehicle safety systems allowing inter-vehicle communications and road-system-to-vehicle communications, this is going to have to be something regulated along the lines of how the Nevada Gaming Commission regulates the algorithms in slot machines with the regulatory agency having full control and the ability to compel changes or updates, and where the system is well and thoroughly vetted and continually tested, revised, and updated over time. Think of the headaches involved in updating web browser SSL certificate authorities but now pushed out to ultimately tens of millions of vehicles requiring regular revision.

    In short I don't think that we're ready for that. And that's before we look at the privacy issues.

    • It's hard to think of situations where wireless communication would make a huge difference practically. If a car is sending out "stalled car", what difference does that make? It helps you know why there's a traffic jam? Stalled cars being rear-ended isn't a huge cause of traffic fatalities.

      There are impractical cool things you could do with perfectly synchronized driving and wireless communication, like coordinated intersections that don't require stopping [youtube.com]. But that is not going to be practical any time soo

      • It's hard to think of situations where wireless communication would make a huge difference practically. If a car is sending out "stalled car", what difference does that make? It helps you know why there's a traffic jam?

        Without more information, even that isn't enough. Stalled where, on what roadway, in what direction? Shared vehicle information will need GPS, roadway and directional information -- or better mile-marker data -- for it to be relevant. There will also have to be a way to determine immediate/near-by proximity for any vehicle movement/motion information.

        • by TWX ( 665546 )

          I figured the audience was smart enough to realize that the notification would include more than literally "S-T-A-L-L-E-D-C-A-R" but I guess Slashdot is slipping a bit.

          • I figured the audience was smart enough to realize that the notification would include more than literally "S-T-A-L-L-E-D-C-A-R" but I guess Slashdot is slipping a bit.

            No, you're right, I figured you *probably* meant that :-) but thought it was worth stating as it means vehicles will (and will need to) be sharing a LOT of information and it may be a privacy issue vs. actual benefit situation. For example, will your long-distance travel/destination be important to know for, say, predicting traffic flow/congestion?

      • ob disc: I work in the field.

        v2x is a game changer. distributed knowledge of road issues - even if you are not in the field, you can imagine that seeing 'farther ahead' is useful.

        we need this. we can't get as far along in the science as we need to unless cars talk to each other and to smart infra.

        • even if you are not in the field, you can imagine that seeing 'farther ahead' is useful.

          Not when people are looking at their phones while driving.

          we need this. we can't get as far along in the science as we need to unless cars talk to each other and to smart infra.

          No, we do not need more strung out programmers developing crappy software which will, inevitably, mangle the person's car when it either gets hacked or corrupted. Or both. And let's not get into the increase in insurance costs beca
        • by TWX ( 665546 )

          For what it's worth I am well aware of how useful such information would be, particularly for self-driving vehicles. This sort of information would allow for not only safety information, but general route planning to reduce congestion in any given area by making more intelligent use of the existing road network, including realtime recalculation if conditions on the intended route change.

          I'm just concerned that mistakes during either the design phase or the implementation on a given vendor's system will res

        • even if you are not in the field, you can imagine that seeing 'farther ahead' is useful.

          I don't see how it would be more useful than Google maps.

        • ob disc: I work in the field.

          v2x is a game changer. distributed knowledge of road issues - even if you are not in the field, you can imagine that seeing 'farther ahead' is useful.

          we need this. we can't get as far along in the science as we need to unless cars talk to each other and to smart infra.

          It's an outdated dangerous technology that risks privacy, security and safety for no reason. If I cared about seeing further ahead there is an app for that.

        • by dgatwood ( 11270 )

          ob disc: I work in the field.

          v2x is a game changer. distributed knowledge of road issues - even if you are not in the field, you can imagine that seeing 'farther ahead' is useful.

          No, I really can't. There are two problems that are inherent and unsolvable here:

          • Trust. Unlike with server-based systems, where obvious fake data can be filtered out, your car isn't going to have the data from thousands of other cars telling it that there can't be a real problem. It's going to have data from whatever is in mesh network range. So as soon as you have any sort of vehicle-to-vehicle communication, your car is now making decisions based on what some other car is saying, and if that data is
    • NGC level = DEALER ONLY service for any thing even tires, oil, lights

    • As much as I would like to see vehicle safety systems allowing inter-vehicle communications and road-system-to-vehicle communications, this is going to have to be something regulated along the lines of how the Nevada Gaming Commission regulates the algorithms in slot machines with the regulatory agency having full control and the ability to compel changes or updates, and where the system is well and thoroughly vetted and continually tested, revised, and updated over time.

      Even if that happens you've still got an easily exploited system. The inherit flaw is the idea that no-one could compromise the network because all of the firmware / code is dictated by a regulatory authority. There's absolutely nothing to prevent the vehicle owner, or others, from tampering with the system after it's turned on. You said it yourself, that these vehicle manufacturers are incompetent when it comes to software security. It won't take but one or two mistakes from them for hackers to get acces

  • "To Save Lives, Issue Connected Vehicle Technology Waiver, NTSB Tells FCC"

    -- was there a contest to create the most incomprehensible headline for this? It took me several tries to realize that "issue" is a verb here, and that a "connected technology waiver" is a thing. Extra points off for hiding it in a subordinate clause.

  • What (Score:4, Informative)

    by phantomfive ( 622387 ) on Tuesday August 30, 2022 @01:30PM (#62837167) Journal

    Connected vehicle technology would reduce the ever-escalating carnage on US roads

    Haven't auto deaths been decreasing for a long time now?

    • Connected vehicle technology would reduce the ever-escalating carnage on US roads

      Haven't auto deaths been decreasing for a long time now?

      That doesn't fit the narrative. The narrative is that cars with people driving are completely unsafe and always will be. We must, must, MUST hand control over to the computers, "TO SAVE LIVES." Never mind that will make us easier to track, and once they convince us that we no longer need to own our own vehicles, make it much easier to lock-down urban areas and prevent movement between population centers. For our own safety, of course, because those people that live the next town over are some nefarious moth

      • I don't mind fear-mongering. But I like technically-correct fear mongering.

        Don't say "ever-escalating" if it's not actually escalating. Say "horrifying" or "terrifying" or some other meaningless word.

        • I wouldn't mind fear-mongering if I thought the general public was capable of discerning fear-mongering from fact. But it's become very clear most see fear-mongering as fact and overreact and overcorrect as hard and as fast as they can.

          • It's not that the general public can't discern fear-mongering, it's that they like it. And amplify it.

            • That's not a good thing either.

              The worship of fear as a way of life is unhealthy in the extreme. It may be the most unhealthy part of American society. It's certainly the most egregiously contagious.

              It'd just be nice if people would take a breath every once in a while and stop spreading the breathless, "Is your house, your car, your child, your cookware, your deodorant, trying to kill you?! *GASP*" Just, every once in awhile. It's an incredible time sink, and the energy people waste worrying over nothing be

      • That doesn't fit the narrative. The narrative is that cars with people driving are completely unsafe and always will be

        You are aware that about 38,000 Americans die in car crashes every year, right? Plus 5,000 pedestrian deaths?

        I get that you're worried about the slippery slope, but that is a lot of people getting killed, and you don't seem to have much of a response beyond stating the obvious arguments against your position in a sarcastic tone.

        • I'm aware. And I'm sorry. Living isn't the most safe activity.

          I realize that the modern era safety is far more important than freedom, but I'll never subscribe to that philosophy. Maybe, if we had some sort of proof of concept of self-driving vehicles that weren't more nuisance and headache than solutions, I'd consider this a "problem we can solve." But until we start getting hours logged for the machines with a comparable safety record, it's not even a fight worth having.

          "Please save us from ourselves" is

          • Granted, the TSA still exists and thus far we've gotten actual proof that they do nothing to actually increase safety, but a ton to cause nuisance and harassment.

            Really? Look at the history of airline bombing attempts since 911 [wikipedia.org]. The ones in Russia, Mogadishu, and Egypt were successful. The 3 targeting the US were failures. One was discovered during an inspection, and the other two (the underwear bomber and shoe bomber) failed because their covert explosive devices did not work. Why would they use t

            • Anyways back to automobile transceivers. I don't think they are a material step towards taking away the right to drive. We already use short-distance transmissions to communicate - that's what brake lights, turn signals, and traffic lights are. But those transmissions are in a form that is convenient for people, but less reliable than a digitally encoded radio transmission for a computer.

              What makes transmitters operating at 400, 500 and 600thz not digitally encoded signals? Either the transmitter is on or its off which clearly encodes unambiguous digital information.

              High frequency digital signals have a desirable benefit of highly accurate direction finding allowing precise relative location of brake lights, turn signals and traffic lights to be ascertained without regard for GPS, augmentation systems, LOS satellite obstructions and current atmospheric or ionospheric conditions.

              So, you get problems for example from glare, or snow - these can be hard for cars to see but also for people, but a digital signal will have no problem at all.

              If the arg

        • You are aware that about 38,000 Americans die in car crashes every year, right? Plus 5,000 pedestrian deaths?

          Not a problem. The word "nuclear" isn't included. You remember that, right? The source of power that hasn't killed one percent of your "year's supply of deaths" in all of history, much less annually.

          Okay, take a deep breath, Avenger. You've gotten over that urge before, you can do it again.

          That said, I'd rather have a little more danger in my life than be wrapped up in swaddling and taken care

    • No they’ve been increasing actually https://www.nhtsa.gov/press-re... [nhtsa.gov]

      • No they’ve been increasing actually https://www.nhtsa.gov/press-re... [nhtsa.gov]

        That article says the rate is decreasing

        The total number is up:
        "NHTSA projects that an estimated 42,915 people died in motor vehicle traffic crashes last year, a 10.5% increase from the 38,824 fatalities in 2020."

        But the rate is down:
        "Preliminary data reported by the Federal Highway Administration show that vehicle miles traveled in 2021 increased by about 325 billion miles, or about 11.2%, as compared to 2020. "

        Data estimates show the fatality rate for 2021 was 1.33 fatalities per 100 million VMT, marginal

    • Not everyone killed by car crashes is in a car.

      2021 has been the highest year for pedestrian deaths in 40 years and 2022 is expected to exceed that. In no small part by ever increasingly hostile vehicles such as ridiculous brodozers and the increasing prevalence of SUVs.

      And of course these wireless devices won't do shit to protect pedestrians.

      • And of course these wireless devices won't do shit to protect pedestrians.

        Unless I'm walking around with one of those wireless devices in my pocket spamming out "stop" to every car in range.

        I changed my mind, I support these things.

  • by cnaumann ( 466328 ) on Tuesday August 30, 2022 @01:37PM (#62837183)

    Deaths and Death rates have been (mostly) decreasing.

    https://injuryfacts.nsc.org/mo... [nsc.org]

    • by EvilSS ( 557649 )
      Yep, especially if you ignore everything after 2014.
      • Look that the second graph "Death Rate per 100,000,000 vehicle miles. The trend is downward, and the rate is less than half of what it was 40 years ago. Over the last 60 years, deaths and death rates have dropped significantly. Granted, things have been relatively flat since 2014.

        None of those graphs show "ever-increasing carnage."

        • by EvilSS ( 557649 )
          No, it's not. Remove the other two lines on that graph and you will see it's trending slightly upward. It only looks that way because the scale is skewed by the others. If you look at the table it's (starting at 2014) 1.17, 1.22, 1.27, 1.22, 1.20, 1.46. Up from 2014 to 2018, a dip, then back up.
  • by awwshit ( 6214476 ) on Tuesday August 30, 2022 @01:37PM (#62837185)

    This will bring new meaning to the term 'war driving'.

    I guess the V2X system in my car must be 'broken'. First I show up as 3 cars, then 1 car, then I'm jamming the signal, then protocol errors. The dealer can't seem to figure it out... sorry that chaos on the road is not my fault.

  • The FCC should just give them the waver. Without congress forcing them to adapt it the automakers will just let it sit on a shelf like they did with the original bandwidth allocation anyway.
  • by tomkost ( 944194 ) on Tuesday August 30, 2022 @02:11PM (#62837279)
    Safety technologies must be fail safe. Wireless is not suitable for something the needs to be fail safe. Wireless technologies should only be relied on for non-safety related enhancements like entertainment or other enhanced features.
    • "Fail safe" does not mean "use only components that do not fail", for obvious reasons.

      • by tomkost ( 944194 )
        yep, it's a concept and guideline. Very few things are absolute. It very well can be the case that some interesting and valuable safety improvements can only be done via wireless. So my point is more along the lines that wireless should never be the primary path to increase the safety until such time that it's truly ubiquitous and reliable. We're far from that now. I have a dead spot in the nearest shopping plaza to my house here in Austin and so does pretty much every other provider.. it's like a blac
  • If V2V were unhackable it could act like some kind of air traffic control but on the ground (there is such a thing as Ground Control in large busy airports already. Different crew handles that, on a separate freq.)

    But in practice this will be instantly abused by the authorities.

    Instead, replace all intersections with roundabouts. I bet the gas savings and reduced emissions will be appetizing. You dont stop! Well, until the people that can't handle roundabouts make it stop, jamming up the works for a whil

    • But on a serious note.. I do love roundabouts vs. stoplights. No question. Far more efficient.

      Efficient yes, but people seem very confused as to who has the right of way

      • Don't worry, they'll get used to it.

        Here in FL we have a similar issue with right turns on red. They're allowed, but not if there's oncoming traffic. They still go.

        Right Of Way confuses many, and roundabouts are but a small part.

        Here we have the "Hialeah Maneuver" -- make a right, by first cutting across 3 lanes of traffic. We have more than that, but that one's bad enough it earned a name for itself.

        • Don't worry, they'll get used to it.

          Here in FL we have a similar issue with right turns on red. They're allowed, but not if there's oncoming traffic. They still go.

          Right Of Way confuses many, and roundabouts are but a small part.

          Here we have the "Hialeah Maneuver" -- make a right, by first cutting across 3 lanes of traffic. We have more than that, but that one's bad enough it earned a name for itself.

          Interesting, I thought "traffic circles" and "roundabouts" were the same thing, but apparently not so.

          From: http://www.ronbeitler.com/2015... [ronbeitler.com]
          "Roundabouts are not traffic circles: Nor are they rotaries (New England) or neighborhood circles. Immediately as an almost knee jerk and hysteric reaction folks claim “Jersey is removing their roundabouts”. Not true. Jersey primarily has had traffic circles. These are not roundabouts. The two have almost nothing in common. At one point New Jersey had over

      • The way I explain it to my parents:

        1. Do you know what a yield sign is?
        2. Do you know what a one way road is?

        You know what a round about is!

    • How do pedestrians and cyclists cross over an endless stream of cager traffic to/from a roundabout?
  • by sinij ( 911942 ) on Tuesday August 30, 2022 @02:42PM (#62837367)
    About ten years ago I spoke to someone at a conference who worked on car2car communication. I asked that person - what would prevent a police department setting up a beacon that would collect data, which includes speed, from every car and then automatically mail speeding tickets? There was a loaded pause, followed by "nothing".

    My belief that car2car protocols will be abused, it is a broadcasting beacon built into your car that you can't turn off, it is not in our interest to have it.
    • > what would prevent a police department setting up a beacon

      Furthermore, what would prevent hackers from screwing around with data as well? If cars are designed to respond to data from other cars to avoid accidents (with inevitable security holes), certain software spoofs might be able to cause sudden braking, swerving into the shoulder or worse.

      The primary thing they need to do to prove such a system is safe is show that is it somehow immune from malicious use. Which I don't see how they can pos
  • Truckers (mainly) and others (hobbyists) have had this forever -- CB Radio. Call out where the speed traps are. Call out where the accidents are. Call out where the lot lizards are. Call out where the seat covers are.

    I used to drive extensively back in the day, inter-state, and my Rx7 and Miata both had permanently-installed CB. I was married then, and it was mainly to talk with the wife when we moved from state to state because usaf job. Hers had a magnet mount. Mine were all permanent.

    But I used it

  • NTSB needs to give it up and stop trying to promote antiquated unsafe technologies.

    Whether it is a warning system that warns a human or a sensor that provides input to a self driving vehicle it's way better to respond to reality as it is actually detected in the real world rather than rely on musing of anonymous to you RF signals broadcasted from anywhere.

    NTSB's scheme has been leap frogged by sensors/CV and fancy algorithms. The rest is just "blah blah blahing" about issues with no real world import and m

  • So you have 15,000 office employees who’s only job is to create a list of frequencies and who owns them. For some reason, the FCC solution for every technical problem is a list of frequencies and who owns them. If they had another solution, it might delegitimize thousands of office workers’ jobs.

You are always doing something marginal when the boss drops by your desk.

Working...