4 ms·

Well it seems you're determined to dismiss any scenario. I won't try to convince you. But, I hope we do agree without question that it's possible, should we wi

Well it seems you're determined to dismiss any scenario. I won't try to convince you. But, I hope we do agree without question that it's possible, should we wish to do so.

lazide 4y ago
Autonomous ocean going drones have been common place for decades. Fire and forget programmable torpedoes have been a thing for decades. What makes no sense is trying to link them in some operational/controllable/recallable way to submarines which are trying to remain covert, as the communication channels give away the submarine or require them to do risky behaviors like surface for satellite comms. So my point is, you’re not proposing anything novel where you seem to think you are, and where you think they are advantages are actually significant operational disadvantages based on actual submarine mission profiles.
- Schroedingersat 4y ago
  If you wanted to make this work, you could. Step 1. Have a thousand cheap solar (or buoyancy/wave) charged sonar drones spread over hundreds of miles. Step 2. Have a hundred silent listening drones spread over the same area. Sonar drones can tell anyone and everyone that they exist and where they are, and encrypt what they see to transmit over whatever channel. Listening drones learn what the sonar drones saw by whatever means (satellite, VLF radio, etc). One of them gets close enough to the sub to tell it. The others all whisper into the void somewhere. Or the sub just listens directly if it's close enough. There's probably some better way of the sub learning what's around it that's top secret though
  - lazide 4y ago
    FYI, might be interesting reading - [https://en.m.wikipedia.org/wiki/SOSUS https://en.m.wikipedia.org/wiki/SOSUS] [https://www.researchgate.net/publication/340683649_Cellular_Communications_in_Ocean_Waves_for_Maritime_Internet_of_Things https://www.researchgate.net/publication/340683649_Cellular_...] The big issues tend to be: 1) active sonar plays havoc with marine life, and it’s not a problem that can be typically solved by shifting frequencies, as the frequencies used by certain species are used due to beneficial transmission characteristics. Wartime sonar is so powerful it can literally kill or injure people or animals nearby in the water due to the over pressure pulse, so tends to be seen as dangerous and undesirable to use except in specific emergency circumstances. Low energy active sonar is indeed used by a lot of folks, but due to #3, would likely have some economic issues. 2) sound can be listened to in the oceans over far longer distances than a useful return signal can be gleaned by the emitter, and tends to transmit unevenly due to factors that are difficult to impossible to predict or control like thermoclines, currents, etc. [https://oceanservice.noaa.gov/facts/sound.html https://oceanservice.noaa.gov/facts/sound.html] Listeners to an active sonar emitter can identify the location of something using triangulation even 10-100x the distance the ‘pinging’ party could get a useful return signal. (See performance here [https://en.m.wikipedia.org/wiki/Sonar https://en.m.wikipedia.org/wiki/Sonar]. Think of it like shining a flashlight around in the dark. Folks miles around could see the pinprick of light easily, even if the flashlight wielder can see nothing out there, and it’s not bright enough for everyone else to see anything else either. Listeners at 2x the distance would typically have the same clarity of view as the initial ‘pinging’ party within the area energized. This also makes it easy to avoid for parties trying to hide. They stay 5x the distance from the active sonar, and they’re likely good, minus passive sonars listening for engine noises or whatever which are always a hazard. And 3) distances in the ocean are huge. Much, much larger than you might be mentally modeling. Outside of a harbor or port (which generally already have these systems), a hundred square miles of ocean is tiny. The Atlantic Ocean is 41.2 million square miles, the pacific 63.8, and the Indian Ocean 27. So covering any useful segment of a large shoreline can be nearly impossible economically, let alone a segment of the open ocean. Even if you wanted a sensor per 100 square miles over half of the ocean - say the most interesting/busy parts of the ocean, or approx. 70 million square miles, well, 700k drones are not going to be easy to build, maintain, etc. For this reason, equipment tends to be carried by parties who might need it, and only deployed in areas they might need it, or deployed as fixed installations in very high value/interesting areas. Sonobouys from helicopters, sonar arrays on ships or submarines, fixed sonar installations at harbors, ports, clandestine passive sonar arrays near expected strategic areas off coastlines, etc. 4) clandestine bandwidth is very limited. VLF has VERY low bandwidth (depending on frequency), so low even voice transmission is generally not practical, and it still has limited penetration into water. As far as I am aware, it’s used the equivalent of submarine texting (at best), and the submarine still has to come quite close to the surface. ELF can work at operating depths, but communication is one way, and bandwidth is in the ‘bits per minute’ range [https://en.m.wikipedia.org/wiki/Communication_with_submarines https://en.m.wikipedia.org/wiki/Communication_with_submarine...] ELF antennas are measured in miles and use the earths crust. Sound in the ocean has weird transmission effects. If your quiet ‘whispered’ data gets under a thermocline, it could go 100s of km and you would be none the wiser. If it didn’t, it might go just meters. Distortion at high frequencies also makes it difficult to transmit much data this way unless the listener is close. You’d want some kind of mesh network ideally though, as whatever node the sub is near would not be one the sub would want to approach if there was a potential enemy contact nearby, as it’s drawing a giant target on them. If one way, that could be useful - however that also means the crypto better be really good, as one compromised node would turn the array into an enemy asset to hunt your subs, and the passive listening and mesh network model would mean you’d need to be transmitting all potentially interesting data all the time.
    - Schroedingersat 4y ago
      > 2) sound can be listened to in the oceans over far longer distances than a useful return signal can be gleaned by the emitter, and tends to transmit unevenly due to factors that are difficult to impossible to predict or control like thermoclines, currents, etc. [https://oceanservice.noaa.gov/facts/sound.html https://oceanservice.noaa.gov/facts/sound.html] This sounds exactly like the kind of thing a top secret military program would know how to do with a few petaflops of calculating power, a network of linked microphones, and a complete survey of the area. At least to ranges that would be very surprising to someone outside of that program. But yes, anything some rando on hn thinks of in five minutes has been thought of and dismissed or thought of in detail and had problems said rando didn't think of worked through.
- throwaway09223 4y ago
  > you’re not proposing anything novel where you seem to think you are I haven't suggested any novel ideas at all. I'm only pointing out something that should be obvious. This conversation started when someone asked whether it's possible to use sonar without giving away a sub's location. It is possible. That's really all there is to it.
  - lazide 4y ago
    It is not possible to have bi-directional communication using sonar without giving away a subs location. If doing the equivalent of dropping a speaker into the ocean and giving a speech, yes it is possible of course. This might be an interesting paper - https://www.researchgate.net/publication/340683649_Cellular_Communications_in_Ocean_Waves_for_Maritime_Internet_of_Things https://www.researchgate.net/publication/340683649_Cellular_... Covering specific physics reasons, alternatives, etc.
    - throwaway09223 4y ago
      > It is not possible to have bi-directional communication using sonar Yes of course. No one has suggested otherwise. That's the entire point of using a drone, after all.
      - lazide 4y ago
        For the drone to be useful operationally to the submarine, the submarine would need some degree of control over it? They couldn’t control something that they can’t talk to without giving away their position?
        throwaway09223 4y ago
        > For the drone to be useful operationally to the submarine, the submarine would need some degree of control over it? No, it would not. We already discussed this above, vis a vis preprogrammed paths. I don't really want to rehash things we've already discussed. Have a good one.

Well it seems you're determined to dismiss any scenario. I won't try to convince you. But, I hope we do agree without question that it's possible, should we wish to do so.

lazide 4y ago

Autonomous ocean going drones have been common place for decades. Fire and forget programmable torpedoes have been a thing for decades. What makes no sense is trying to link them in some operational/controllable/recallable way to submarines which are trying to remain covert, as the communication channels give away the submarine or require them to do risky behaviors like surface for satellite comms. So my point is, you’re not proposing anything novel where you seem to think you are, and where you think they are advantages are actually significant operational disadvantages based on actual submarine mission profiles.

Schroedingersat 4y ago
If you wanted to make this work, you could. Step 1. Have a thousand cheap solar (or buoyancy/wave) charged sonar drones spread over hundreds of miles. Step 2. Have a hundred silent listening drones spread over the same area. Sonar drones can tell anyone and everyone that they exist and where they are, and encrypt what they see to transmit over whatever channel. Listening drones learn what the sonar drones saw by whatever means (satellite, VLF radio, etc). One of them gets close enough to the sub to tell it. The others all whisper into the void somewhere. Or the sub just listens directly if it's close enough. There's probably some better way of the sub learning what's around it that's top secret though
- lazide 4y ago
  FYI, might be interesting reading - [https://en.m.wikipedia.org/wiki/SOSUS https://en.m.wikipedia.org/wiki/SOSUS] [https://www.researchgate.net/publication/340683649_Cellular_Communications_in_Ocean_Waves_for_Maritime_Internet_of_Things https://www.researchgate.net/publication/340683649_Cellular_...] The big issues tend to be: 1) active sonar plays havoc with marine life, and it’s not a problem that can be typically solved by shifting frequencies, as the frequencies used by certain species are used due to beneficial transmission characteristics. Wartime sonar is so powerful it can literally kill or injure people or animals nearby in the water due to the over pressure pulse, so tends to be seen as dangerous and undesirable to use except in specific emergency circumstances. Low energy active sonar is indeed used by a lot of folks, but due to #3, would likely have some economic issues. 2) sound can be listened to in the oceans over far longer distances than a useful return signal can be gleaned by the emitter, and tends to transmit unevenly due to factors that are difficult to impossible to predict or control like thermoclines, currents, etc. [https://oceanservice.noaa.gov/facts/sound.html https://oceanservice.noaa.gov/facts/sound.html] Listeners to an active sonar emitter can identify the location of something using triangulation even 10-100x the distance the ‘pinging’ party could get a useful return signal. (See performance here [https://en.m.wikipedia.org/wiki/Sonar https://en.m.wikipedia.org/wiki/Sonar]. Think of it like shining a flashlight around in the dark. Folks miles around could see the pinprick of light easily, even if the flashlight wielder can see nothing out there, and it’s not bright enough for everyone else to see anything else either. Listeners at 2x the distance would typically have the same clarity of view as the initial ‘pinging’ party within the area energized. This also makes it easy to avoid for parties trying to hide. They stay 5x the distance from the active sonar, and they’re likely good, minus passive sonars listening for engine noises or whatever which are always a hazard. And 3) distances in the ocean are huge. Much, much larger than you might be mentally modeling. Outside of a harbor or port (which generally already have these systems), a hundred square miles of ocean is tiny. The Atlantic Ocean is 41.2 million square miles, the pacific 63.8, and the Indian Ocean 27. So covering any useful segment of a large shoreline can be nearly impossible economically, let alone a segment of the open ocean. Even if you wanted a sensor per 100 square miles over half of the ocean - say the most interesting/busy parts of the ocean, or approx. 70 million square miles, well, 700k drones are not going to be easy to build, maintain, etc. For this reason, equipment tends to be carried by parties who might need it, and only deployed in areas they might need it, or deployed as fixed installations in very high value/interesting areas. Sonobouys from helicopters, sonar arrays on ships or submarines, fixed sonar installations at harbors, ports, clandestine passive sonar arrays near expected strategic areas off coastlines, etc. 4) clandestine bandwidth is very limited. VLF has VERY low bandwidth (depending on frequency), so low even voice transmission is generally not practical, and it still has limited penetration into water. As far as I am aware, it’s used the equivalent of submarine texting (at best), and the submarine still has to come quite close to the surface. ELF can work at operating depths, but communication is one way, and bandwidth is in the ‘bits per minute’ range [https://en.m.wikipedia.org/wiki/Communication_with_submarines https://en.m.wikipedia.org/wiki/Communication_with_submarine...] ELF antennas are measured in miles and use the earths crust. Sound in the ocean has weird transmission effects. If your quiet ‘whispered’ data gets under a thermocline, it could go 100s of km and you would be none the wiser. If it didn’t, it might go just meters. Distortion at high frequencies also makes it difficult to transmit much data this way unless the listener is close. You’d want some kind of mesh network ideally though, as whatever node the sub is near would not be one the sub would want to approach if there was a potential enemy contact nearby, as it’s drawing a giant target on them. If one way, that could be useful - however that also means the crypto better be really good, as one compromised node would turn the array into an enemy asset to hunt your subs, and the passive listening and mesh network model would mean you’d need to be transmitting all potentially interesting data all the time.
  - Schroedingersat 4y ago
    > 2) sound can be listened to in the oceans over far longer distances than a useful return signal can be gleaned by the emitter, and tends to transmit unevenly due to factors that are difficult to impossible to predict or control like thermoclines, currents, etc. [https://oceanservice.noaa.gov/facts/sound.html https://oceanservice.noaa.gov/facts/sound.html] This sounds exactly like the kind of thing a top secret military program would know how to do with a few petaflops of calculating power, a network of linked microphones, and a complete survey of the area. At least to ranges that would be very surprising to someone outside of that program. But yes, anything some rando on hn thinks of in five minutes has been thought of and dismissed or thought of in detail and had problems said rando didn't think of worked through.
throwaway09223 4y ago
> you’re not proposing anything novel where you seem to think you are I haven't suggested any novel ideas at all. I'm only pointing out something that should be obvious. This conversation started when someone asked whether it's possible to use sonar without giving away a sub's location. It is possible. That's really all there is to it.
- lazide 4y ago
  It is not possible to have bi-directional communication using sonar without giving away a subs location. If doing the equivalent of dropping a speaker into the ocean and giving a speech, yes it is possible of course. This might be an interesting paper - https://www.researchgate.net/publication/340683649_Cellular_Communications_in_Ocean_Waves_for_Maritime_Internet_of_Things https://www.researchgate.net/publication/340683649_Cellular_... Covering specific physics reasons, alternatives, etc.
  - throwaway09223 4y ago
    > It is not possible to have bi-directional communication using sonar Yes of course. No one has suggested otherwise. That's the entire point of using a drone, after all.
    - lazide 4y ago
      For the drone to be useful operationally to the submarine, the submarine would need some degree of control over it? They couldn’t control something that they can’t talk to without giving away their position?
      - throwaway09223 4y ago
        > For the drone to be useful operationally to the submarine, the submarine would need some degree of control over it? No, it would not. We already discussed this above, vis a vis preprogrammed paths. I don't really want to rehash things we've already discussed. Have a good one.