Southwest Pilot Rant Locks Out Frequency

[Kata]

aidan, you lack any understanding of the levels of cost involved in aviation hardware. You note that it is a mission-critical application; so it needs to be much better engineered than a cell phone. The certification requirements are enormous, entirely different to consumer electronics, and the production volumes are low. You'd be talking many thousands of dollars for even a simple unit.

Look: even a decent model current generation com radio is £3000, not including another £1000 for installation. http://www.gps.co.uk/Transceivers/Ga...p-102-261-273/

Also, with analogue radio you can still make out transmissions even when at the limits of range; it's not a simple "now you hear me, now you don't" like digital would be.

If two people are transmitting on the same digital frequency at once, will they not block each other anyway?

[peterb]

I don't see any reason why it should cost much more than a basic cell phone to make. So it would depend on how scrupulous the manufacturers are.

That is because you have little understanding of aeronautical engineering, or flight safety critical engineering and the standards and type approval a piece of aircraft equipment has to undergo before it is cleared for use.

Kata has covered most of the essential details, but an aircraft radio is not some bit of mass produced consumer kit like a cell 'phone where volume sales keep the price down.

[TheAnimus]

1) why bother, current system works just fine most of the time, there are FAR bigger issues (bad pun there!) that need to be looked at.
2) an ATC operative is still required per channel, if someone fills it with crap, they still have to listen in case they say the important bit.
3) all equipment in aviation has to be of a slightly higher grade than normal, this greatly increases the costs.
4) migration, it would be impossible to migrate in the way you envision if you compare it to the TV switchover.
5) more complexity in the air will lead to more human error in the air.
6) there is already a system in place that notifies pilots when cross talk is occuring with quite high reliability.
7) all the nav aids and stuff that currently work, would you change them or just scap em and force GPS?
8) I think you have a strong failure to understand how in air radio is used in a busy airport, and how many different frequencies there already are on the spectrum, the order in which you get clarences to move.
9) international emergency standards have been met, changing these would require either every country to switch, or have a very confusing and potentially deadly mix match until migrated. (ie when bored tuning 121.5)
10) cost, aviation is damned expensive as is.

I think I missed some out too.

[Kata]

Navaids? Hell yeah, scrap the lot NBDs, at least...

[TheAnimus]

If two people are transmitting on the same digital frequency at once, will they not block each other anyway?

Yes they can, but you greatly reduce the probability of a clash by having a spread spectrum or hopping, or use of frames.

But this gives me an excuse to talk about this lady:
http://en.wikipedia.org/wiki/Hedy_Lamarr

A hollywood actress, star of a very riske film that shocked sensabilities at the time.

A fricken genius!

But it would go off topic to try and drag this out any further, but an amazing if not troubled lady.

[badass]

Ahh yes this reminds me of a conversation an old boss's boss (a finance person) had with the IT dept when we were trying to track down a problem with our backups taking around 14 hours when they should have taken 5. To top it off, they were impacting certain users desktop performance in a way that shouldn't have occurred.

Anyway, I digress. What I remember most was when said boss's boss told us that our backups should take 2 hours. He knew nothing about the amount of data we had to backup, the mechanism we used to perform said backups or even the cost of our at the time current data protection infrastructure.
But of course he knew they should take 2 hours (they were supposed to take 5 hours. Even with the tape drive running flat out, they would have taken 3 1/2 hours)
I suppose what I'm saying is he really looked foolish making such a statement and should probably have stuck to what he knows rather than making stupid statements about things he knows the square root of nothing about.

[aidanjt]

aidan, you lack any understanding of the levels of cost involved in aviation hardware. You note that it is a mission-critical application; so it needs to be much better engineered than a cell phone. The certification requirements are enormous, entirely different to consumer electronics, and the production volumes are low. You'd be talking many thousands of dollars for even a simple unit.

The fact that current hardware is so expensive is all the more reason. Fair point about certification, can't get around that. Although engineering 'mission-critical' boards is a mere matter of making the PCB a bit bigger to reduce resistance, using slightly more expensive power regulation components, and slapping it in a metal chassis, and getting it certified. People often think 'mission-critical' requires some kind of magical engineering voodoo, but it really doesn't. Almost all of 'mission-critical' engineering is satisfying relevant certification requirements to maintain standard spec. and filling in the associated paperwork and sending off the proto for testing. But the overall lower cost design remains.

Look: even a decent model current generation com radio is £3000, not including another £1000 for installation. http://www.gps.co.uk/Transceivers/Ga...p-102-261-273/

Analogue components are more expensive than silicon chips, and the analogue aviation radio market has long passed saturation point. The transition to modern technology wouldn't have either of those impediments.

Also, with analogue radio you can still make out transmissions even when at the limits of range; it's not a simple "now you hear me, now you don't" like digital would be.

On the plus side, the digital signal is much more resilient against noise to begin with, so the actual range will be extended and is easily repeatable anyway.

If two people are transmitting on the same digital frequency at once, will they not block each other anyway?

No. Digital signalling doesn't operate in continuous mode. That's why your phone/laptop/desktop/etc can send and receive files over the internet at the same time.

[santa claus]

you lack any understanding.....

That is because you have little understanding.......

1) I think you have a strong failure to understand.......

.......making stupid statements about things he knows the square root of nothing about.

I dunno, Hexus used to be a happier place .

[peterb]

I don't really know where to begin in correcting the erroneous assumptions made in the above post.

The fact that current hardware is so expensive is all the more reason. Fair point about certification, can't get around that.

Indeed not - and that is one of the most expensive parts of the process - getting and maintaining the appropriate certification.

Although engineering 'mission-critical' boards is a mere matter of making the PCB a bit bigger to reduce resistance, using slightly more expensive power regulation components, and slapping it in a metal chassis, and getting it certified.

Not sure where you dragged these misconceptions from. "Making the PCB a bit bigger to reduce resistance"? Using slightly more expensive power regulation components"? "Slapping it in a metal chassis"?

The whole point about mission critical systems is that their is traceability throughout the individual unit's life, from manufacture through service record and finally disposal. The whole quality control and quality assurance process is complex, before you even get to the design stage. Depending on the system, that may mean that each individual component is tested, the specificatuions of each component more rigorous and that may mean different packaging of ICs (so more limited production runs) and a meticulous audit trail of each productipon unit. In true MC system, software may be developed by independent teams. All this adds to the development costs. Rather than make equipment bigger, the drive is to make it smaller and lighter, as weight impacts aircraft performance. That brings its own problems with heat dissipation.

An aircraft is a hostile environment for electronic systems, and equipment has to be designed to take that into account. Changes to an aircraft fit may involve the design team, maintenance records and maintenance schedules, drawings, configuration management, etc, etc. None of this is cheap. New equipment has to be electromagnetically compatable with the legacy equipment in the airframe.

People often think 'mission-critical' requires some kind of magical engineering voodoo, but it really doesn't.

True, it requires rigorous and reviewed design methods, and meticulous quality control.

Almost all of 'mission-critical' engineering is satisfying relevant certification requirements to maintain standard spec. and filling in the associated paperwork and sending off the proto for testing. But the overall lower cost design remains.

And the testing is not simply plugging it in and seeing if it works! It may (will) involve vibration and environmental testing, often over many weeks.

Analogue components are more expensive than silicon chips, and the analogue aviation radio market has long passed saturation point. The transition to modern technology wouldn't have either of those impediments.

The cost of the components is insignificant against the costs of development and maintaining a continuous QA audit procedure.

On the plus side, the digital signal is much more resilient against noise to begin with, so the actual range will be extended and is easily repeatable anyway.

A sweeping generalisation, much depends on the modulation characteristics of the bearer.

No. Digital signalling doesn't operate in continuous mode. That's why your phone/laptop/desktop/etc can send and receive files over the internet at the same time.

No, your phone/laptop/desktop uses the internet at the same time because it is packet switched. If you are referring to the layer 1 protocol for ethernet (CDMA - CD) then there is an element of truth in what you say, in that the packets are transmitted in turn, so that each device can access a common lan, but that has nothing to do with the mechanism of file sharing protocols. Cellphones do employ frequency hopping over a limited number of channels, but only as part of the hand-off process as they pass from one cell to another.

For radio transmission of data, the transmission is continuous, and some form of modulation is applied (QPSK is a common one) but the transmitted signal is still continuous. As for the robustness of a digital signal, that is down to the error correcting functions built in to the protocol in use.

There are protocols used for data transmission over wide area radio communication nets, using Time Division Multiplexing and/or Code Division Multiplexing, but they are very specialised and not generally suited to voice transmission.

<---snip

If two people are transmitting on the same digital frequency at once, will they not block each other anyway?

Yers, although you need to get away from thinking about frequency, and think about channels. A channel may use multiple frequencies sequentially (TDM or FDM) or spread spectrum (CDM) but the receivers all have to be synchronised with the transmitter. And the whole point of an air traffic control network is that it is all-informed, so that everyone can hear what is going on in that particular piece of airspace, so everyone needs to be using the same frequency/channel.

And if the problem was so serious, and the solution as simple as you seem to think, don't you think it would have been implemented, or in the process of implemented by now?

Communications systems will evolve, possibly along the lines you mention, but it will be a gradual evolution, not a big bang switch over.

[TeePee]

Pilots have to show that they know it all.. It's compulsive.

My company has 60 radios, (only one of which was made after 1990). Assuming a digital radio doesn't cost more than a new analog radio (which I think is a big assumption), then why force us to spend $180,000 for a problem which doesn't exist?

Moving to new technology has always cost more in aviation. A certified GPS will cost $6K upward. The new Mode S transponders are $2K or more. A 406mhz ELT? $600. None of these things really do more than non-aviation equipment. A Garmin 430 is primitive compared to the Tom Tom in your car. That ELT does less than a portable Spot tracker. The problem is certification.

[TheAnimus]

I dunno, Hexus used to be a happier place .

I'm not one to say you should always yield to the heard mentality, but it might be on to something.

The thing is, if someone doesn't raise something as a question "how come we aren't doing x" "is it just an economic proposal", "isn't this a serious problem" or similar, if they state as fact something they demonstrably know nothing about, then they are being a bit stupid. They are just going to make themselves look like a fool.

I don't think any of us have an issue debating the issues of radiotelephony for pilotage, but you can't have someone simply saying things that are so untrue.

It's quite apparent that we've got people who understand radio theory, pilotage, frequency hopping, digital flow control, and economic value proposistions to a much higher standard. We are not all equal.

Peterb is in most things a much better electronic engineer than I am, TeePee I think has considerably more hours than the rest of the pilots in this thread put together and raised to an exponant.

So when you come in, with something that is plainly wrong, your going to get told that.

If you don't want people to think your a complete idiot, ask questions about the stuff you don't know, debate it, don't try and represent it as a complete fact.

[aidanjt]

Rather than make equipment bigger, the drive is to make it smaller and lighter, as weight impacts aircraft performance. That brings its own problems with heat dissipation.

It's a trade off. But smaller brings more error potential, and thermal problems, especially when tolerating with noise, surges, spikes, etc..

And the testing is not simply plugging it in and seeing if it works! It may (will) involve vibration and environmental testing, often over many weeks.

I didn't say they'd be a walk in the park.

The cost of the components is insignificant against the costs of development and maintaining a continuous QA audit procedure.

Right, but we're not speculating about a non-existent device's potential R&D costs. We're also not speculating about the profit margin the manufacturer will be after. We're talking about how much the unit price might be, which is much easier to guestimate.

A sweeping generalisation, much depends on the modulation characteristics of the bearer.

A generalisation in much the same way as little things tend to stick to, and be drawn towards really big heavy things. Watt for watt, a digital signal can take much more degradation than an analogue waveform.

No, your phone/laptop/desktop uses the internet at the same time because it is packet switched.

Which is only made possible by digital transmission.

If you are referring to the layer 1 protocol for ethernet (CDMA - CD) then there is an element of truth in what you say, in that the packets are transmitted in turn, so that each device can access a common lan, but that has nothing to do with the mechanism of file sharing protocols.

Other than it being inherent to digital transmission.

Cellphones do employ frequency hopping over a limited number of channels, but only as part of the hand-off process as they pass from one cell to another.

Exactly, it's easy to develop into an aviation radio standard.

For radio transmission of data, the transmission is continuous and some form of modulation is applied (QPSK is a common one) but the transmitted signal is still continuous.

There's nothing which physically requires continuous transmission or any kind of modulation, that's a holdover from analogue thinking (and rides a digital signal over an analogue carrier). All you need is an omni-directional pulse of photons to represent an on bit. With that said, current modulated digital transmission standards happily perform bidirectionally.

As for the robustness of a digital signal, that is down to the error correcting functions built in to the protocol in use.

It also depends on weather, frequency, atmospheric particulates, transmit power, and a whole bunch of other factors. But ultimately, digital signals will take more of a beating.

There are protocols used for data transmission over wide area radio communication nets, using Time Division Multiplexing and/or Code Division Multiplexing, but they are very specialised and not generally suited to voice transmission.

And yet, it works on GSM, figure that.

Yers, although you need to get away from thinking about frequency, and think about channels. A channel may use multiple frequencies sequentially (TDM or FDM) or spread spectrum (CDM) but the receivers all have to be synchronised with the transmitter. And the whole point of an air traffic control network is that it is all-informed, so that everyone can hear what is going on in that particular piece of airspace, so everyone needs to be using the same frequency/channel.

Nothing stopping other pilots from listening in to traffic on other relevant channels, the main point is that no one person can jam the channel on everyone, since they can have their own channel, and ground control can get the chance to say 'hey, flight #, your transmit button is stuck'. Although here's a idea.. feed ground radar data to aircraft and let it pass the position of other aircraft to the plane's avionics, and let that plot them on a map, or ring a buzzer when a collision is predicted, etc... Whatever.

And if the problem was so serious

I'd say communication halting of a whole channel due to a minor button glitch is kinda serious. If a minor glitch can jam a whole channel, it'd be pretty trivial for an intentional sabotage of comms to jam them all. That's a bit of a security concern. The pure utility and reliability of digital transmission is the cherry on top.

and the solution as simple as you seem to think, don't you think it would have been implemented, or in the process of implemented by now?

The problem isn't with coming up with an implementation, that's the easy part. The problem is convincing flying types that their world wont fall apart and they'll all lose their jobs if changes are made.

Communications systems will evolve, possibly along the lines you mention, but it will be a gradual evolution, not a big bang switch over.

The point is (civilian) aviation comms hasn't evolved. It's pretty much the same technology as the original radio.

You don't have to agree with me that it's a good idea. Fine, whatever. But don't go around calling me a clueless moron because you disagree. It's presumptuous and rude, and puts people off discussing anything.

[aidanjt]

I'm not one to say you should always yield to the heard mentality, but it might be on to something.

The thing is, if someone doesn't raise something as a question "how come we aren't doing x" "is it just an economic proposal", "isn't this a serious problem" or similar, if they state as fact something they demonstrably know nothing about, then they are being a bit stupid. They are just going to make themselves look like a fool.

I don't think any of us have an issue debating the issues of radiotelephony for pilotage, but you can't have someone simply saying things that are so untrue.

It's quite apparent that we've got people who understand radio theory, pilotage, frequency hopping, digital flow control, and economic value proposistions to a much higher standard. We are not all equal.

Peterb is in most things a much better electronic engineer than I am, TeePee I think has considerably more hours than the rest of the pilots in this thread put together and raised to an exponant.

So when you come in, with something that is plainly wrong, your going to get told that.

If you don't want people to think your a complete idiot, ask questions about the stuff you don't know, debate it, don't try and represent it as a complete fact.

If you're going to call someone an idiot, then at least get the sentence right, or try this novel idea instead, explain why you think I'm wrong. Just saying I'm wrong is a bare assertion.

[peterb]

You don't have to agree with me that it's a good idea. Fine, whatever. But don't go around calling me a clueless moron because you disagree. It's presumptuous and rude, and puts people off discussing anything.

I don't think those words have been posted by me at all. However you have demonstrated in this thread that you really don't understand the issues involved, either in basic communications engineering, or in flight safety certification. I disagree with you because your assumptions are incorrect.

And the unit cost of the equipment takes into account the R&D costs amortised over the eventual production run. So if a radio costs (say) £1.5M to develop, and the sales forecast is (say) 100 units, then each unit will cost £15,000 to start with, before taking into account component costs, individual unit testing costs. So even if the component cost is £1,000, the unit cost would be at least £16,000. (and just to be clear, this is an illustrative example, not figures relating to any particular piece of equipment)

And neither did I say it was a not good idea - if you re-read the last line of my post. However I am saying that it is not a trivial task. The incident that was reported is extremely rare, there are procedures to deal with it; and if it was a serious safety issue, legislation would be applied to make it so.

But rather than use a sledghammer approach to completely design the global civil aircraft communication system, a re-design of the switch would be a more cost effective approach - if it was a common event.

[Kata]

Although here's a idea.. feed ground radar data to aircraft and let it pass the position of other aircraft to the plane's avionics, and let that plot them on a map, or ring a buzzer when a collision is predicted, etc... Whatever.

Look up TCAS and ADS-B

[aidanjt]

That's another way to do it.