Jump to content

alison

Members
  • Posts

    60
  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

alison's Achievements

Member

Member (2/3)

207

Reputation

  1. It seems to me that open video feeds are still a risk when going up against a high-tech adversary. Let's assume and hope that the telemetry like speed, heading and GPS is already fully encrypted and "uncrackable" so it's only the analog video feed that's "in the clear". If we can scan the standard frequencies and find some of those feeds, then pipe those into another system which already has the terrain visually mapped out from its own overflights, it might be possible to geolocate the incoming drone before it's picked up by other detection systems. I remember someone upthread talking about lasers or autocannon targeting the sound of rotors, and if that's the state of the art then intercepting radio signals containing meaningful data is going to give you a much longer lead time. Of course I am just hypothesizing here, but thinking about scanning radio frequencies looking for audio signals... I am sure that right now hobbyists could set a USB-sized SDR to auto scan, then feed the audio hits into a language model to have it quickly detect what kind of stuff is being discussed on each channel (railroad, logging, weather etc). Video is much more complex but the building blocks are there. If these things travel 60km/h and you start to get signal 10km out, that's up to 10 minutes to figure out where it is and call in one of your anti-drone drones to take it out. Even better if your anti-drone mothership was already in the air taking live footage of the same area to feed back into the model. But as soon as their video feed is encrypted, it shuts down the whole (counter) attack vector.
  2. I just read up on that Aeroscope system and it's the most Chinese **** ever. They're not only not hiding the CCP backdoor, but they've productized it so they can make money on both sides! Get that bag, I suppose, but to my eyes it instantly makes DJI a non-starter for anything outside of hobbyist stuff. Thanks for sharing that webinar. It was sales-y, but it really drives home the sad state of affairs when they're promoting the security angle of dumping the raw footage to SD card while remaining evasive on the live feed question ("well it's downsampled anyway, so it's not the good stuff"). I was hanging for a more detailed description of the technology they're using to communicate from the drone to the controller, but it appears "security through obscurity" is still the name of the game here. Or perhaps the industry really just takes it as a baseline assumption that the live feed is open, which is bananas to me. Perhaps for suicide drones it would be okay to skip encryption on the video feed, if it excessively increases the heat and battery drain, or if the lag sucks too bad. It sounds like these things might soon fly autonomously or be carried inside a mothership to a waypoint in the vicinity of the target and only switch over to direct human control for the final targeting and approach, which hopefully would be at a high enough speed that anyone intercepting the feed would not have time to react. There's still a risk if the enemy collect enough of these final approach feeds that they can determine the flight path of the mothership, or extrapolate where you are launching from, but I suppose that's the same risk that exists with artillery. The spy/recon/overwatch drones, on other hand - the ones that hopefully are going to fly back to base at the end of their mission - those seem like a priority for loading up with all the security. We keep talking on this thread about how maneuver is dead because everyone can see everything going on for miles around, but if everyone is watching everyone else's video feed then what's actually dead is opsec. You can't give people your intel for free, at least make them put up a drone of their own to get it, you know? In any case, it seems I have been a bit too presumptuous about how all this gear works. I am definitely biased by my experience of living in China and going to various lengths trying to 翻墻 (jump the Great Firewall), and now working in the internet privacy and security space. When there's bombs raining down on your head then nerding out over encryption is probably less important than trying to take out the source ASAP. On one hand I hope I never have to experience that. On the other hand, here in Taiwan, who knows? I hope to not be completely ignorant if **** hits the fan, so I appreciate any wisdom people can share here on the thread.
  3. While it is a fair point to make that SSL/TLS everywhere is relatively "new", things in tech do move quickly and the days of chunky crypto accelerator daughter boards - at least for networking - are 20+ years in the rear-view window. Nowadays everyone is carrying around a device in their pocket that can comfortably process E2EE real-time video and audio, so it seems absurd to me that much more expensive FPV drones can't manage the same performance. I get that in an EW environment that's heavily saturated with other signals every little piece helps - if you've ever tried to get Steam Remote Play going over wi-fi in a densely-populated urban area you know the pain - but surely it's cheaper to downgrade the video quality than skip encryption? Especially given H.264 and H.265 compression can be done in consumer-level hardware too nowadays. Perhaps I am missing something obvious because my experience in the tech industry is more server rooms than IoT, but it still seems wacky to me that in 2024 someone could trivially "tap into" the video feed of an FPV drone flying overhead. Somehow more believable to me would be oldskool war driving with a Pringles can and brute-force guessing the enemy wi-fi password. To try to add a bit of independent research here... This is the only reference I can find online on the topic, and it's 8 years old, but they imply that the DJI protocol back then was already encrypted: https://mavicpilots.com/threads/how-secure-is-ocusync.363/ DJI's wonderful webpage localization tech means I can only access the Taiwan version of the page here in the beautiful isle, but from their own page: https://enterprise.dji.com/zh-tw/data-security
  4. Do we have any radio grogs in the house? When I was stuck in rural Canada during COVID I messed around with an SDR listening in to local railroad and logging channels, but I was under the impression that anything serious (law enforcement and certainly military) was both digital and encrypted nowadays. This guy is talking about using non-standard frequencies, but surely that is more about mitigating jamming than interception? In my day job it would be (and has been) front page news everywhere if someone was able to reliably intercept encrypted networking protocols, so how is it not a big deal that FPV video feeds are apparently getting hacked so easily? Are they not encrypted at all? Or is it just the custom-built CCP backdoor special version of "encryption"?
  5. I've been waiting for a decent article on the 珠海云 drone mothership to come out, and CSIS delivered: https://features.csis.org/snapshots/china-research-vessel-taiwan/ There's been other articles on this thing but they're all scant on details and pics, so this piece is a nice summary that gathers together most of what we know. It's probably just a slightly overhyped command+control ship, but I'm a sucker for any drone-related hardware, so here you go.
  6. Just a reminder that Perun did a decent video on space weapons last year.
  7. This has been a dream for decades. I remember reading articles about it during the first round of VR/AR/wearable computing buzz in the 90s. I feel like there have been half a dozen Future Land Force Spectrum Warrior Initiative programs to make it happen, and as far as I know it never has. The last thing I remember was Microsoft pivoting HoloLens to the military. Looking at Wikipedia (https://en.wikipedia.org/wiki/Integrated_Visual_Augmentation_System) it seems like that one might have actually gotten somewhat of a rollout. Perhaps our active duty or recently-serving members can comment on if it's any good? I'm skeptical, because this tech has been just around the corner forever, and yet never seems to be worth its literal weight to people who are already lugging a ton of gear around. I'm not a soldier, but I have done a decent amount of long term travel - bike touring, hiking etc - and even though having a permanent HUD is a cool fantasy, in practice just having a cheap and easily-replaceable phone was good enough. The times you really need hands focused on what's right in front of you, you generally need your full field of vision too. The times a fully networked computer would add context to what you are seeing - when you're in a position to contemplate that extra data - then you also have time to take a glance at a device on your wrist or in your pocket or attached to the rest of your gear. I don't think the barrier is that the technology isn't there yet, I think it's more that humans have discovered that integrating a "second brain" works more smoothly when it's a discrete tool that can be held in our hands and manipulated with our fingers, then tucked away while we focus on something else. The best thing about modern networking technologies is that the front line soldiers can do exactly that, while all the data is still being collected and fed back to the command post where dedicated specialists can apply further analysis. Looking at what we can see from what's happening in Ukraine, the modern feedback loop appears to be faster than it was in the days of human scouts and radio signals, but the overall flow of collecting, filtering and distributing information seems largely unchanged. I am not sure if it will ever become efficient to issue devices that multiplex a dozen inputs to a single soldier when they - quite literally - have an army to lean on. Of course, the sci-fi nerd in me would love to be proven wrong.
  8. Not directly related to the events in Ukraine, but I enjoyed this weekend quick read about how rebels in Myanmar are using drones: https://www.theguardian.com/world/2024/jan/20/myanmar-rebels-junta-above-drones The future of warfare is here, folks.
  9. Well, the election is done, went pretty much the way people expected. If anything the deadlock in the legislative yuan is likely to make the status quo even more status quo-y than it already was. The ball remains in China's court, as it has been for the last 8 years since they unilaterally broke off official relations. Council on Foreign Relations released something fun for the wargamers a few days back: https://www.cfr.org/article/why-china-would-struggle-invade-taiwan It's a multimedia presentation style article that shows the challenges China would face if it it attempted a ground war in Taiwan. I think most people agree that in reality this will mostly be a war waged in cyberspace and the financial sphere, and even if it gets hot it'll more be about subs and missiles than infantry and drones, but it's still interesting to see visualizations of the terrain and imagine how it all could go down.
  10. I think we agree more than we disagree, so my apologies for misinterpreting your post. Western leaders have been clear from the beginning that they aim to avoid major escalation, so that aspect of their response should not surprise anyone. That said, although I know it's been a recurring topic on this thread, I don't think many - if any - governments are cynically using this war to affect regime change in Moscow. It seems to me there would be more effective ways to do this than try to keep a war going outside Russia's borders for some indeterminate period of time that overlaps major elections in countries whose opposition candidates are all too happy to use "let's stop wasting money on foreign wars" as a campaign talking point. Perhaps I am just a bit in the weeds right now, since today was election day. I'll pipe down and get back to reading about drone warfare.
  11. Touché. Perhaps this is a long war simply because both sides have access to fairly similar kit and neither side wants to give up yet.
  12. I am really tired of the idea that there is a mysterious cabal of crypto hawks who somehow have the power to influence defense decisions in dozens of countries around the world and are dedicated to ensuring the war in Ukraine never ends. If anything, people in democratic countries have shown time and again that they do not want war, and even when a war does spark off, they certainly don't want long ones. War is not popular with the people. No politicians are running on a platform of "let's keep everyone at war". On the contrary! Obviously not all decisions made by the government are communicated in detail to the people, but the whole point of democracy is that there is freedom of debate and eventual transparency. There are plenty of anti-war politicians around the world who have an interest in exposing a forever war conspiracy, and yet no evidence has been exposed. So why keep suggesting it exists? It's true that a handful of actors here and there have an interest in dragging wars out for economic reasons, or believe it might be advantageous for geopolitical reasons, but they are far from "the ones in charge", and their position is not widely popular. On the other hand, we have plenty of evidence that warfare has changed in a way that it is no longer easy to deal out crushing victories against near-peer adversaries. Perhaps it never was.
  13. I haven't posted in here for a while because the news is all same-old, same-old, but since yesterday everyone in Taiwan got an air raid alert on their phones, which itself turned into a domestic political debate about if unannounced satellite launches should count as psy-ops, or if they should even trigger an alert in the first place, I suppose it's time to share an update. All the usual sources have decent coverage. I liked this morning's article covering the disinformation warfare aspect, which is the primary mechanism China has been using to try influence the results of the upcoming election: https://www.theguardian.com/world/2024/jan/09/taiwan-presidential-election-china-influence My personal thinking is that yesterday's satellite launch was another form of "I'm not touching you" bullying like the spy balloons. If the Taiwanese government responds to the flyovers, then the angle is that they're overhyping totally-not-military incursions, but if they don't respond, then the angle is that they're weak. Still, even though the Chinese government clearly backs a KMT win, and seems determined to continue their snubbing of Taiwan's democratically elected president if the DPP pulls it off... I don't think any more that the elections here are going to majorly affect any policy decisions on the China side. The CCP may use the results as an excuse to lift or apply light economic pressure, but at the end of the day Xi continues to push the notion that "complete unification" (完全統一) is imminent and inevitable, regardless of what anybody in Taiwan thinks. (And, to be clear, the overwhelming majority of Taiwanese today - even the pro-China side - have no appetite for "unification".) Sadly, I think the more consequential election for Taiwan in 2024 will be the US election. If the US elects a non-interventionist leadership, this will probably embolden China to take more aggressive actions in the region.
  14. Do these in production models all need to be one pilot, one drone, or can a single pilot realistically control a flight or even a squadron already? I am trying to imagine how swarming will work if the gimmick is simultaneous attacks. Even if the autonomous drones are smart enough to hold formation en route to the target, I imagine at some point for the swarm to be effective we're going to need human pilots to do the final run. At least, that's the impression I get from the videos we have seen released so far. Are there any games out which simulate this kind of mission already? I've seen a bunch of FPV drone sims on Steam, but they seem more focused on one pilot, one drone setup. Real-time strategy games with fully-autonomous units have been around for years, but to me that still feels a bit sci-fi. I'm imagining some kind of mission control view showing tiled windows of drones en route with on-screen anomaly tracking, a bit like high tech CCTV setups, and then the ability to jump in to first-person control (or assign pixeltruppen to do so) as required. Is that even representative of real world tech yet?
  15. I have been seeing "line goes up" charts about China for at least 10 years now. China has best mobile payments system in the world. China has best bullet train network in the world. China has best surveillance systems in the world. China has best supply chain in the world. China has best drones in the world. China has best EV manufacturing in the world. And while some of these may be true in the abstract, somehow the Chinese government has not been able to convert these economic and technological advantages into global dominance. I am not sure there is enough evidence yet to show that the current Chinese government is capable of influencing foreign affairs in a major way, outside of propping up a few neighboring authoritarian states. If China was really willing and able to flick a switch and plunge the rest of the world into darkness, it would have done so. But it hasn't. On the contrary, despite the self-own of wolf warrior diplomacy, the government largely still tries to project an image of neutrality because it needs the support of the rest of the world to succeed in its domestic goals. If the Chinese government really wanted to flex, it wouldn't just be selling weapons to Russia, it would trigger a remote kill on every DJI in Ukraine. But instead they're happy to play both sides, because making money is more important than dictating the outcome of a regional conflict. And that's going to be true of regional conflicts all over the world. So maybe the Chinese military industrial complex eclipses the American one at some point, so what? They'll sell weapons to everyone, just like the Americans did. The point that I think other commenters are making in this discussion is that over the long term democracy wins out over autocracy. And I think that victory is better seen through a cultural lens than a military one. People all over the world are watching Japanese anime and Korean dramas. Maybe briefly they were watching Chinese palace dramas, but those got too popular so the government squashed them. Remember Chinese hip-hop? Yeah, so do I until the government harmonized it. Maybe gaming is where Chinese pop culture will finally break through? But gaming has been under attack from the government for years now and it remains to be seen if the Disneyfied versions of what the government lets through will continue to resonate with the global audience. And, if they do, how much of that audience will translate gacha games whose stories are deliberately as far removed from political controversy as possible into a greater trust of China-as-global-leader, as the government perhaps hopes? In recent years, the Xi administration has been pushing Chinese people, Chinese corporations and useful idiots overseas to "tell China's story well", so it can parlay that into global influence, but is that really happening? If the only version of the Chinese story that can ever be shared with the world is a sanitized one, is that really a compelling story? And if the Chinese government can't win hearts and minds, can it ever really achieve the kind of hegemony that the US - or, more broadly, "the democratic west" - does today? Are you really the winner if you sell gadgets to everybody but nobody actually trusts you?
×
×
  • Create New...