The Strategy of Technology

tlb wrote:Yes, an FTL fire control system could be built with the first generation transmitter; but it would not be Apollo.

It would be "Apollo Mk 0.001" In the first post, I used "Apollo" as a generic term for any FTL fire-control. In the second I used "Apollo + KHII" to specify the actual implementation of Apollo.

I apologize for confusing you.

tlb wrote:

kzt wrote:Each device has a time slot and transmits during that slot. All the weapons listen to all all the commands and execute the ones that apply to them.

Same way it works on a wireless access point.

But salvos might be separated by a distance that causes a data string from a missile to blend into another from a trailing group due to propagation delay - not something you need to worry about in normal communication. You might be able to limit this by restricting transmission to some closing distance from target, so only the leading salvo talks.
I will accept that it can be done, but it is outside my knowledge or ability.

The distance between the signal sources do impose a problem, but it's mostly a computational problem.

I don't how much technical background you have, but here is a technical paper describing the methodology used.

kzt wrote:Each device has a time slot and transmits during that slot. All the weapons listen to all all the commands and execute the ones that apply to them.

Same way it works on a wireless access point.

tlb wrote:But salvos might be separated by a distance that causes a data string from a missile to blend into another from a trailing group due to propagation delay - not something you need to worry about in normal communication. You might be able to limit this by restricting transmission to some closing distance from target, so only the leading salvo talks.
I will accept that it can be done, but it is outside my knowledge or ability.

Joat42 wrote:The distance between the signal sources do impose a problem, but it's mostly a computational problem.

I don't how much technical background you have, but here is a technical paper describing the methodology used.

Thank you very much for the reference; some parts were too deep for me, but I think that I have the gist.
To the extent that I did understand that paper, it seemed to increase my uncertainty about FTL communication with Apollo. Two points stand out:
1. The paper emphasizes that a single frequency network has to be limited in size, otherwise you lose the required coordination (? my interpretation). In their example the Czech Republic was too big for one network to be viable.
2. Implementation required a synchronizing time signal, supplied in their example by GPS. Note that the timing signal is supplied external to the network. What would supply the timing in an FTL communication network? Seems to me that it must either an FTL signal or internal clocks on each member of the network.

I need to add that I expect there is a technology solution to this; but it might mean that what we consider high bandwidth for FTL communication would suffer in comparison to RF communication.

tlb wrote:..snip..
Thank you very much for the reference; some parts were too deep for me, but I think that I have the gist.
To the extent that I did understand that paper, it seemed to increase my uncertainty about FTL communication with Apollo. Two points stand out:
1. The paper emphasizes that a single frequency network has to be limited in size, otherwise you lose the required coordination (? my interpretation). In their example the Czech Republic was too big for one network to be viable.
2. Implementation required a synchronizing time signal, supplied in their example by GPS. Note that the timing signal is supplied external to the network. What would supply the timing in an FTL communication network? Seems to me that it must either an FTL signal or internal clocks on each member of the network.

1. The paper is over 10 years old at this point and technology marches on. Some of the problems he mentioned in the paper doesn't exist for our exercise, like reflections from terrain and other obstacles. Since we don't know how the signal/noise environment looks for gravpulses it's difficult to say what the limit of the network is. You can avoid the problem though by using multiple self organizing networks, ie. each gravpulse transceiver talks with their neighbors and come to a consensus how they should organize the networks - this means you essentially can have many networks spread out over a huge volume. It does introduce some latencies in the network though, but they are minuscule compared to the FTL speed gain.

2. There are several solutions. One solution is to elect 1 transceiver as the clock for each network.

This transmission-technique relies heavily on computational power and from a practical and economic standpoint with what we use today it's geared towards being affordable to the average consumer buying a TV for example. Military solutions may have totally different demands.

In essence, the limitations are computational power coupled with balanced transmission power and receiver sensitivity.

Of course, most of this is pure speculation and guesstimates since we don't really have a clue on how the physics work in Honorverse.

The main issue with time devision multiplexing is clocking.

For the time frames we are talking about, syncing clocks on launch would do.
Time slot size could be adjusted based on expected dispersal of the mssiles.
(Interesting point. Modern mobiles have a fixed range based on timeslots not power, but their range can be doubled if two slots are alocated to a phone instead of one. At a cost in capacity of course)

But timeslot multiplexing only works until a second launch is made, or a fleet launches. Then there's too many devices.

So, there must also be frequency seperation. Lots of frequencies.

We also know that there are tight beams of some sort.


So, for me, given the physical size of the ships (plenty of room for transmitters), the distances involved, and the sheer number of Apolo missiles that can potentially be in a fleet launch, Apollo probably gives each control missile a frequency to use and uses tight beams to do it.

Annachie wrote:The main issue with time devision multiplexing is clocking.

For the time frames we are talking about, syncing clocks on launch would do.
Time slot size could be adjusted based on expected dispersal of the mssiles.
(Interesting point. Modern mobiles have a fixed range based on timeslots not power, but their range can be doubled if two slots are alocated to a phone instead of one. At a cost in capacity of course)

But timeslot multiplexing only works until a second launch is made, or a fleet launches. Then there's too many devices.

So, there must also be frequency separation. Lots of frequencies.

We also know that there are tight beams of some sort.

So, for me, given the physical size of the ships (plenty of room for transmitters), the distances involved, and the sheer number of Apolo missiles that can potentially be in a fleet launch, Apollo probably gives each control missile a frequency to use and uses tight beams to do it.

That is the crux: there is no carrier wave whose frequency can be adjusted. Each bit in a message is a FTL pulse traveling on the hyperspace boundary, so internal clocks are perhaps the only way.

Annachie wrote:So, for me, given the physical size of the ships (plenty of room for transmitters), the distances involved, and the sheer number of Apolo missiles that can potentially be in a fleet launch, Apollo probably gives each control missile a frequency to use and uses tight beams to do it.

However, for reasons that I don't recall yet seeing given it doesn't (at least for now) seem possible to transmit the FTL control signals to an Apollo 23E control missile directly from a ship.[1] They all have to go through the Keyhole II remote platform.

So the ship size doesn't let you physically spread those link transmitters out.
-----------------
[1] Maybe the missile receiver can't pick out the signal from the noise of the ship's wedge, maybe it's something else. But IIRC RFC said pretty clearly that if an SD(P) lost both Keyholes IIs to combat damage they wouldn't be able to FTL control their missiles. If there was a way to distribute those fire control links over the actual ship I have to imagine that they would have.

Jonathan_S wrote:

Annachie wrote:So, for me, given the physical size of the ships (plenty of room for transmitters), the distances involved, and the sheer number of Apolo missiles that can potentially be in a fleet launch, Apollo probably gives each control missile a frequency to use and uses tight beams to do it.

However, for reasons that I don't recall yet seeing given it doesn't (at least for now) seem possible to transmit the FTL control signals to an Apollo 23E control missile directly from a ship.[1] They all have to go through the Keyhole II remote platform.

So the ship size doesn't let you physically spread those link transmitters out.
-----------------


[1] Maybe the missile receiver can't pick out the signal from the noise of the ship's wedge, maybe it's something else. But IIRC RFC said pretty clearly that if an SD(P) lost both Keyholes IIs to combat damage they wouldn't be able to FTL control their missiles. If there was a way to distribute those fire control links over the actual ship I have to imagine that they would have.

Uh, for the most part that's to get the transmitter outside the wedge perimeter of the mother ship. Any individual ship is talking only to [its[/i] 23-Es. Now, those 23-Es are cross-commmunicating with the 23-Es of every other salvo in the overall launch.It is neither necessary nor desirable for the launching ship to have a huge antenna farm (FTL version), and trying to open up its wedge aspects to handle that many of them would significantly increase its own vulnerability.