LightSquared scrabbles to save itself after FCC stops LTE plan LightSquared is reportedly trying to swap radio spectrum with the US military in an attempt to salvage its business model after the FCC pulled the rug from under the mobile broadband biz - but its customers are already abandoning it. First to go is FreedomPop, a free-broadband-for-all operation (whose business plan makes …
For all L^2's squawking, their tech people must have realised this was a non-starter from the beginning. The bands adjacent to GPS are designated as space->earth only, for the good reason that physics makes it *really* hard for strong and very weak signals to coexist close by each other in any practical way. In the bandwidth they use, GPS signals are actually below noise even in good receivers and it's clever signal processing that extracts them.
It was really surprising that the FCC originally gave them hope that they might prevail. That probably says a lot about the pressures that can be brought on federal agencies in the US political system. A comment I read on another forum suggested several other companies with space->earth allocations were waiting to step in and use this as a precedent for their own terrestrial spectrum usage plans, as space->earth spectrum is a lot cheaper than earth<->earth spectrum.
"It was really surprising that the FCC originally gave them hope that they might prevail."
Correct, we've also seen dangerous nonsense from the FCC with travesties such as BPL/PLC/PLT. The fact that the world's power grid is being/is to be used as the world's largest antenna--and a broadband noise-producing one at that which raises the spectrum's noise floor across the whole planet--shows how the politics within the FCC works these days (and the FCC is certainly no longer run by spectrum management engineers as once it was). In science and engineering parlance, BPL/PLC/PLT is 'inmates in charge of the asylum' stuff, and using the word 'engineering' in the same sentence implies its author ought to be locked away with them.
Trouble is, much of the Western World just copy-cats the FCC. They now have to, as most governments have gotten rid of their spectrum management engineers then downsized and outsourced spectrum management to consultants who then pinch ideas from the FCC! (If the FCC does it then it can't be wrong.)
Spectrum management, once engineering, now a mess. It can't be otherwise since spectrum auctions; for these days money always talks louder than engineering.
" The bands adjacent to GPS are designated as space->earth only,"
No they weren't. They were designated for satellite mobile comms. That involves earth->space too, presumably in the same frequency band, or you're in for a very one sided phone call.
Given that, don't you think that the earth based transmitter that fulfilled the earth->space leg would have caused similar problems? Especially as that transmitter would have been co-located with a GPS rx, namely the one in the satellite mobile phone?
The problem with this whole debate is that no-one is thinking clearly about what the actual technical issues are, were, and always have been. In summary,
1) The GPS industry have been lazy in ignoring frequency allocations that were always going to cause them problems
2) The FCC didn't even begin to think what the technical issues would have been resulting from the sat phone band allocation
3) The FCC were negligent (as you hinted) when permitting the change use; it would have been a good time to have re-assessed the band allocation given widespread GPS usage
4) The FCC are being cowards in not telling the GPS industry "tough luck”
5) LightSquared have been naive in trusting the cowardly FCC to do their job properly. A little testing would have revealed the problem ages ago and avoided the whole thing.
As a result:
1) LightSquared's investors are going to lose a lot of money
2) The US taxpayer is going to lose a lot of money through under exploitation of valuable spectrum space, and possibly as a result of being sued by LightSquared for negligence
3) A viable technical solution to the whole issue (the filters that LightSquared developed) is probably not going to see the light of day because the freetard GPS industry's lobbying looks to have paid off
Many people will crow if and as seems likely when LightSquared are dead and buried. But really it is the tax payer and consumer who is going to lose out the most. That's not good for anyone?
"Given that, don't you think that the earth based transmitter that fulfilled the earth->space leg would have caused similar problems? Especially as that transmitter would have been co-located with a GPS rx, namely the one in the satellite mobile phone?"
Yes there is earth to space communication BUT the power levels are far below what LS was trying to do. Your phone is has a transmission power of up to 210mW and the average satellite phone is between 50 and 300 mW. An LTE tower can be up to 200 Watts!!!!!!!! See the difference? The issue was the output of the tower overloading the weak GPS signal. 300mW wouldn't overload it.
@lance 3
Firstly your phone is capable of 2watts. Secondly a sat phone would have needed at least that, probably more. Thirdly, a sat phone might only have been 2watts but being only an inch or so from the GPS receiver in the sat phone it would have been far worse than a L^2 basestation a few miles away. Or haven't you heard of the inverse square law? Or do you some how imagine that a sat phone user would never have wanted a built in GPS? Or that they wouldn't have wanted to do something like use Google Maps? Or do you imagine that a sat phone GPS jammed by the sat phone itself would somehow have been less annoying to every sat phone user than what LTE will cause?
The problem (yes it is real enough) has *always* been there lurking in the frequency allocation, satellite or terrestrial, but it has taken L^2 to make everyone realise it. No one noticed before because sat phone was a commercial non starter. That's how badly the FCC have dealt with this.
I'm no happier about this than anyone else. But don't blame L^2, blame the FCC for not keeping their eye on the ball. I mean, hadn't the FCC heard of built in GPS?
@bazza,
No it isn't. There is partail truth to the figure you have provided. GSM is a TDM technology and there are 8 time slots; 2 watts /8 = 250mW and that 2Watts is for 850/1900 and 1 watt in the 1800/1900 band; so 125mW.
To further prove how wrong you are. Lithium ion batteries are around 3.6 volts. The average smartphone has between 1000 and 1500 mAh. If we take 1.5 Ah * 3.6v = 5.4 Wh. 2 watts would mean less than three hours of talk time. Most phones beat that hands down.
A USB port also only supplies 2.5 watts. 2 Watts would be awfully close to that especially since the USB broadband modem has more than just a transmitter on it.
We are also talking about LTE and not GSM. UMTS and LTE are far below that of what GSM can use.
More fail from you? All the FCC did, give LightSquared the chance to show that their proposal would not interfere with GPS. It was the job of the FCC. They sold the spectrum for a use and it was LightSquared who wanted to use it for a different purpose. So the FCC gave them a chance to present their case. LightSquared went to the FCC, not the other way around.
>To further prove how wrong you are. Lithium ion batteries are around 3.6 volts. The average smartphone has between 1000 and 1500 mAh. If we take 1.5 Ah * 3.6v = 5.4 Wh. 2 watts would mean less than three hours of talk time. Most phones beat that hands down.
The 2W is *max* transmit power - there's all sorts of funky algorithms tuning that depending on signal quality to the base station. If can go much lower and still support a call under certain channel conditions.
@JetSetJim
The power specs are out there. A sat phone is not anywhere close to 2 watts or even 1 watt. If you look at a CDMA/EV-DO phone, they too are not even lose to that, they are FAR less. Analog phones were 2 watts for a car installed and 600 mW for a handheld.
My phone will not operate anywhere close to 2 watts. I have it set for 3G mode only and as such, it will never be close to 2 watts or even 1 watt. LS was going to be LTE.
W-CDMA (UMTS) has far less transmit power than GSM. LTE has even less than W-CDMA. So LTE is far lower than what GSM is. Even CDMA on a satellite phone isn't even close to 1 watt let alone 2 watts!
I think you'll find that RF front ends care very much indeed about peak received power. Your, er, fail is to not read what I wrote. So I shall write it once more. That band, even if used for sat phones, makes it difficult to get a co-located built in GPS receiver to work at the same time. The FCC should have realised it, the GPS industry should have realised it, and maybe LS were naïve in assuming that the others already had.
As for your spurious refereneces to mobile phone power management, if you do happen to be 35km from a base station (traditionally the GSM max range) then your phone will be running at 2watts and yes it will go flat very quickly indeed. If anything your analysis illustrates that a sat phone in that band co-located with a GPS rx would have been in more trouble than if afflicted by LTE; you couldn't have got the satellites closer to allow the phones to put out less power. Even more reason for the FCC to spot that the band allocation was going to be problematic.
And on the topic of the FCC, if it isn't their job (as you hint) to know what band allocations are viable and best for all US users, what exactly are they for?
It would be interesting to know what plans for built in GPS LS had/have. Their whole proposition depends on GPS working in LS handsets. If they can show a working technical solution then that is the end of the debate. Alas I think that whether LS's filters are 'working' is not going to be assessed objectively by the press, commentards, lobbyists, politicians, CEOs, shareholders, accountants, bureaucrats or the courts, and I doubt that engineers will get a look in.
It is YOU that CANNOT READ. We are NOT talking about GSM, we are talking about LTE.
LTE has a FAR lower output power than GSM or W-CDMA. That is a fact, that is part of the specification. The fact is, n one really uses GSM anymore, they are using W-CDMA. It is no where even close to 2 watts or even 1 watt!
Don't get me wrong here: the FCC made the right call. Eventually. The problem I have with it, and the one Lightsquared should be kicking up a fuss about, is that the FCC did so by changing the rules on them. LightSquared was playing by the rules and are now out shedloads of money because the FCC stupidly told them they could do something that anyone who understands radio communications should have realized was going to cause problems to a network that has now become an essential service. Daft as LightSquared's bussiness plan was the blame for it, and their loss of capital, lies squarely with the FCC IMHO.
Hardly, it wasn't as though the FCC approached them and said "you know that bandwidth you've bought for a satellite phone, how about using it for something different and we'll give you a waiver to do it".
Lightsquared always knew they were pushing it and sailing close to the wind - trying to find loopholes in the system. There is a reason why such prime frequency was able to be purchased so cheap.
This isn't like a tax loophole where as long as you can argue that the law allows it even though morally you are paying a fraction of the intended tax you can get away with it. The FCC also have a remit to protect the spectrum and other license holders and guard against abuse.
If anything the FCC are guilty of bowing to pressure from Lightsquared and giving them false hope by allowing them to try to come up with a solution where they should have just said "no" to start with.
Even so, LightSquared knew they were exploiting a loop-hole in the rules. If it never crossed their minds that technical and political issues could come up rendering their plan a failure, then they deserve to be out of all the cash they spent. They bet big and the house won, and they only have themselves to blame.
There are two possibilities here as I see it:
1) They never predicted the interference problem would be insurmountable and that they'd be shot down in flames.
Verdict: Their scientific analysis was crap and they deserved everything they got.
2) They knew about the potential issues, but thought they'd gamble on getting away with it and making a shitload of cash.
Verdict: Their business model was crap and they deserved everything they got.
What have I missed?
They must have known about the issues (actual, not potential btw) as a lot of research went into this in the 90s in the context of using pseudolites to augment the NAVSTAR network in order to develop GPS-based VNAV approaches [i.e., getting aeroplanes on the ground in low visibility conditions]. Nearly twenty years latter, we're starting to have RNAV (horizontal navigation) GNSS approaches but VNAV is still on the drawing board and I haven't heard about pseudolites for a while, even though in the limited context at hand they could have been viable.
1) The US already has GPS-based VNAV approaches that go as low as 200-feet AGL decision heights (the same as basic ILS). It's called WAAS, and doesn't use any ground-based transmitters (there are ground-based stations that relay correction data up to the WAAS geo-sync satellites, though).
2) RNAV is short for "area navigation" and a generic term. I think you mean LNAV (lateral navigation).
3) I'm not disputing your claim, but I'm curious what research you're specifically referring to. I'm sure multiple studies/research efforts have been done on this topic.
The following: "the fault lies with the GPS manufacturers, who have been happily making making devices that pick up signals outside the GPS band"
If the GPS picks a signal, then it means someone is transmitting it. Are the GPS satellites emitting outside the approved GPS bands?
No, what it comes down to is that the signal is weak. They are medium Earth orbit satellites and as such are over 11,000 miles away from the GPS receiver. So the signal is quite weak even if it was quite strong to start. LS was going to be right next door with an massively more powerful signal that would overload the and drown out the GPS signal.
Analogy. Someone is 5 feet from you and quietly whispering; you can barely hear them. That is the GPS signal. Someone pulls up in a car 100 feet away with a powerful stereo system and has it cranked all the way up. Can you hear that whisper? The car stereo is LightSquared.
If LS had bought the correct spectrum, this would never have been an issue.
The GPS will be most sensitive in the GPS band. The problem is you can't be very sensitive in one band and still block out everything in the adjacent band. Filters aren't solid barriers, they reduce the signal but can't ever completely block it. The further away the signal is from the one the filter want's to let through the more it is reduced. You can make that cut off sharper but going beyond a certain rate requires large increases in both the cost and size of the filter.
If the lightsquared signal was a few times more powerful than the GPS signal then the filters in the GPS would have been sufficient. Since that band is for satellite coms that is the signal strength that was assumed when the GPS was designed. But lightsquared is not a little stronger, it's hundreds of millions of times stronger. No, that's not an exaggeration, you do get such large scales in the RF world.
The filters in the GPS receivers simply don't cut the signals down sufficiently, any filter that did would either be massive, expensive or risk either weakening or distorting the GPS signal itself.
Less accurate GPS systems (e.g. cell phones) can put up with a distorted GPS signal and the slight loss of accuracy that produces meaning that suitable filtering is possible and not too expensive for them. In fact a lot of them already have it to stop their own transmissions jamming the GPS (the cell phone signal is further from GPS frequencies but it's also very very close so the amount the filter needs to reduce it by is far higher).
High accuracy GPS systems need more of the GPS signal to get through with less distortion. That means their filters can't cut off as quickly as in a cheap system and so they suffer more jamming from lightsquared.
For some high precision receivers, they do listen for a signal outside of the GPS band. But they do so because they are using GPS correction services that pay satellite companies like LightSquared to broadcast them. So they are perfectly legitimate, legal and licensed. These receivers are heavily affected since they were listening for a weak satellite signal and now get a strong terrestrial one.
For other GPS receivers, they do not listen in the LS spectrum. However, because brickwall filters don't exist in real-life, at the power level that LS is broadcasting at, these GPS receivers are forced to hear the unwanted LS signal and causing interference.
Well, the FCC should have stomped on this before billions of Dollars in investment was at stake, but the real issue is that Lightsquared tried to press too far with spectrum that was ill-suited for their high-powered signal.
A shame, but GPS has become too important in too many applications to tolerate anything but minor signal jamming.
Maybe we can salvage something from this by putting LightSquared right next to the frequencies that Galileo will use--thereby killing Galileo and saving Euro-taxpayers some much-needed dosh!!
Galileo used the same frequencies as GPS.
As do the Russian GLONASS and Chinese COMPASS systems.
As does WAAS, EGNOS and all the other regional SBAS systems.
This is possible because the signals are all about the same strength and are encoded in such a way as they look like background noise unless you know what to look for.
The Republican government from 2000 set this up to fail. They allowed the creation of "too big to fail companies" creating territorial monopolies out of cable and telecoms who see the death threat of a lightsquared company.
The GPS consortuim was simply the front man, dig deeper and you will see that the big money came from the telecoms and cable industry.
When the Bush crowd, and GOP led FCC of that era ruled that Cable didn't have to follow the rules and serve everyone, they opened the door to allowing maximized profits, minimized customers. They strpped the profitable areas, gave shoddy service, and allowed no competition in - not even to service those who could not beg, borrow or steal broadband connections.
Light squared would change that and therefore had to be stopped at all costs - and what better way than to use the GPS system, one of the most useful tools out there, as a surefire excuse to deny light squared a shot?
cable stopped two miles from my home 20 YEARS AGO. there is no fiber optic cable, no fiber optic telephone abd only 13 percent of the county I live in has access to broadband of anykind -including wireless cell service. We are 30 miles from Williamsburg, 40 miles from richmond, 120 miles from Washington DC.
If a company like Light Squared had started service in this area, the cable users would have begun dropping their poorly serviced product like files, disrupting the telecoms and Cable monopolies.
Of course the FCC said "no" to them.
"LightSquared scrabbles to save itself after FCC stops LTE plan"
The FCC didn't stop anything. The waiver was conditional, one of which was that the interference issue had to be resolved per the satisfaction of the commission. If LightSquared made plans before they received an unconditional waiver, then that was their fault. LightSquared should have focused on the task at hand an not start counting their eggs before they hatched.
The fact that you were struck by a car is because you made plans before the car struck you.
At least according to the logic used in coming up with "The FCC didn't stop anything". statement.
The fact that the car that hit you is operating outside the law is as irrelevant as the fact that the companies the FCC is not enforcing the law on for utilizing out of spectrum frquencies is giving the FCC a reason to stop the legitimate owner of those frequencies from having the right to use them, Right? Of course, Big Corporations operating under the "might makes right" doctrine.
The FCC screwed up, violated the public trust and penalized the wrong party.
You almost have it right. However, in this case LightSquared would be the car operating outside of the law, seeing as how it has no legal right to setup a stand-alone terrestrial network and under Federal Regulations CFR 25.255, ground transmitters operated by LightSquared cannot interfere with other users:
"If harmful interference is caused to other services by ancillary MSS ATC operations, either from ATC base stations or mobile terminals, the MSS ATC operator must resolve any such interference. If the MSS ATC operator claims to have resolved the interference and other operators claim that interference has not been resolved, then the parties to the dispute may petition the Commission for a resolution of their claims."
http://ecfr.gpoaccess.gov/cgi/t/text/text-idx?type=simple;c=ecfr;cc=ecfr;idno=47;region=DIV1;q1=25;rgn=div5;sid=7002a7770b4cdaf775b55914c0f87c4d;view=text;node=47%3A2.0.1.1.4#47:2.0.1.1.4.3.41.33
To use a car analogy:
LightSquared bought some train wheels. Then they realised that trains are too expensive and wanted to make cars with those wheels.
So they asked the FCC for permission to use these train wheels on normal roads.
The FCC replied with "You can use those new wheels if you can prove to other road users satisfaction that they don't damage the road".
It turns out that train wheels do damage tarmac, so they can't use them. End of story.
Now you could argue that the FCC should never have allowed LightSquared to try to show that they could co-exist with GPS. Unfortunately that does mean you want the FCC to always give a flat "No. Go Away" answer to any possible change-of-use in a band.
That's probably going a bit too far.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
