LightSquared screams 'conspiracy' over leaky test results LightSquared's CEO is demanding an investigation into how draft test results on its technology were leaked from a government-assigned testing house to Bloomberg. The testing was to establish if satellite broadband biz LightSquared's mitigation techniques will let it coexist amicably with equipment which relies on signals from …
Exactly when will the man from LightSquared be showing up to retrofit this filter into my 2008 Mercedes' tightly integrated navigation system? Because I'll probably still have same car in ten years, and it'd be nice if the embedded GPS wasn't screwed-up by these discount-by-cheating-the-system mobile towers.
No, I will not accept a replacement $80 portable GPS with wires draped across the dash.
I'm sure that the GPS receiver in his car receives the correct bands just fine. And after all, when the car was manufactured, nobody expected it to need satellite navigation to work out what road it is on when it is next to a SATELLITE transmitting at an altitude of about 26000km.
Only the difference now is that Lightsquared want to put the "satellite transmitters" on the ground!
Tell me, how well can you see the stars if I am standing next to you, shining a million candle power torch in your eyes?
It is already a fantastic achievement being able to reliably pick up satellite transmissions - and not just one, but a whole bunch of them. To then go and say it's up to the GPS lot to find a way to eliminate interference from VERY close signals many thousands of times stronger than the ones they want is way beyond ludicrous.
Oh, and they were there first!
Perhaps you should read the article instead. Here, let me help you a tad:
"LightSquared's mad-as-a-box-of-frogs plan is to use radio frequencies formerly reserved for satellite communications to host a mobile-phone network. "
They also talk about needing thousands of base stations. As an added bonus, you can see the word formerly right there in the quote.
So, the GPS makers built their kit with the silly idea that neighboring bands reserved for satellite signals was never going to be used for nearby mobile phones. How silly of them.
How do you retro-fit filters to existing GPS kit with everything built in? From my UK viewpoint, I fail to see why their scheme should be allowed to get off the ground. Surely the whole idea of regulation is to set aside and allocate frequencies for specific purposes? To just tear up the rule-book and allow satellite frequencies to be used for other systems seems total folly to me. It sets a dangerous precedent. Where will it all end?
LightSquared bought spectrum on the cheap...because it was reserved for satellite to earth comms. If used in that way, it would not have interfered with GPS. In spite of this, LightSquared has tried to game the system, by ignoring international spectrum allocations and attempting to build a terrestrial network in a frequency band reserved for another use entirely.
Of course they're interfering with GPS receivers. They were designed for use in a band where the only transmitters are supposed to be on satellites!
"They should have stayed "reserved for satellite communications"
Wouldn't have helped. The satcom phone itself is also a transmitter operating in the very same band. It has to be, unless you want a one way conversation... And guess where the satcom phone would be? Right next to the GPS receiver that the owner also has.
GPS receivers aren't fussy - they'll quite happily get jammed by any adjacent interference, whether its coming from a satcom phone or a LightSquared base station. Had that old satcom service ever have become popular we would have seen this problem years ago I reckon.
instead of playing one sided political games or egomaniacal trips or functioning as an espionage clearing house, Wikileaks used to go around and find the dirty secrets of corporate ne'er-do-wells and backdoor deals between politicos of ALL parties and their big business tycoons, breaking out the dirty details of how we were going to be screwed next.
Looks like someone else stepped up to the plate. kudos!
regardless of the desirability of the technology, the results should not have been leaked!
I am not surprised that LightSquared's CEO is hopping mad - this has just killed his company. No other investor will now come on board even if they bullet-proof the system and show no interference.
This is serious stuff from a government lab - I work with a lot of clients submitting data to governments and I know my clients are not going to like the thought that preliminary results can get leaked to the press.
It is my understanding GPS receivers fall under Part 15 of the FCC regulations. Part 15 clearly and unequivocally states consumer receiving equipment MUST tolerate all interference from any licensed service which is operating within the terms and conditions of its license.
The battle between manufacturers of "receiving equipment" (think, e.g. of televisions, broadcast radio receivers[AM, FM, Shortwave, etc.] , garage door openers, FRS radios,et. al.) and operators of transmitting equipment. Amateur radio operators battled for years with television manufacturers, VHF commercial radio operators battled with high-power paging service operators. And so on, ad nauseum.
In every case, the real issue in the battle was the cost of designing receiving equipment to reject unwanted signals. Many television receivers, as an example, had little to no filtering on the input and RF input stages highly sensitive to frequencies well outside the television broadcast frequencies due in large part to lack of shielding, poor component and stage layout (optimized for lowest manufacturing cost instead of best RF performance), cheapest usable components, etc. etc. etc..The issue was manufacturing cost which translates into profit margins which translates into consumer cost.
I highly doubt this GPS vs. KightSquared fracas is any different. A pox on both their houses!
(I much prefer dead-tree maps and 3G is far more than fast enough for my use.)
"from any licensed service which is operating within the terms and conditions of its license."
And there you have hit the nail on the head. The license is for transmissions FROM satellites. By the time they reach the ground, they are very weak, similar in strength to the GPS signals, and hence very easy to ignore.
However, Lightsquared has said that satellites are too expensive, yeah, sure they'll have 1, but they want all the other transmitters on the ground; where their signals will be many, many times the strength of GPS signals.
I suppose that in theory you could do it, but there may be some issues...
The GPS receiver does some simultaneous equations to work out it's distance to each satellite. Each satellite also transmits it's orbit information, so the GPS knows the location of each satellite, and it's distance to each, so can work out it's own location.
Each satellite is about 20,000km away. Now imagine that you add another "satellite" that is 1km away. Perhaps the calculations don't bother trying to work out distances for 1km? If they do, should a sanity check reject the distance being 1km? Does the location information allow for a "satellite" with an altitude of 30m instead of 20,000km?
Then there is the other issue. If the relatively high-power signals transmitted from the ground blot out _all_ the satellite signals, the GPS receiver will now need to get a lock from 3 or more ground based transmitters to calculate it's position instead of using the sats. That means a lot of groundstations!
Obviously Lightsquared is having trouble making its target numbers so its pushing the modulation so it bleeds over GPS's frequences. That's dumb; their engineers should have seen this coming.
GPS is one of the few services that absolutely has to be kept running. Its not a convenience function for phone users, it provides safety critical data and a high precision time reference. Over time it may be improved and hardened but for now Lightsquared will either have to reduce their product's footprint or move somewhere else.
I know that in modern society the idea of the public good is a bit old fashioned -- nothing must get in the way of making money -- but the idea that you could render a whole bunch of GPS units inoperable just so that companies can push streaming media is ridiculous. Satellites have very limited transmission power and GPS, to be usable, cannot carry dish antennas like you have with satellite TV. So expecting some kind of magic GPS receiver to appear is just not going to happen.
The GPS receiver manufacturers took the cheap route for the most part when they designed their receivers. They never expected the spill over from ground based stations and designed the receivers with very a wide bandpass. Additional filtering would add a few bucks to the cost of a receiver. Don't add the filters and you end up with a crap device in the presence of extraneous signals.
Hmmm. On reflection that may explain various Landrovers with sat phones unexpectly ending up in a river ;-)
Not as simple as "Additional filtering would add a few bucks to the cost of a receiver". You forget to consider the size impact for portable devices and the increase in system noise due to the filter's loss (which gets bigger as you make it narrower and 'smaller' for a given material choice) wich impacts antenna requirements and/or the ability to work in poor reception conditions (indoors, under forest canopy, etc).
Further more you also neglect the power consumption implications of higher overload margin LNAs, mixers, and low phase noise local oscillators (as even a very good pre-LNA filter won't stop everything). Again, that works against battery powered stuff, which you might notice is a common GPS requirement.
"may explain various Landrovers with sat phones unexpectly ending up in a river" - I am willing to bet that is usually a faulty meat-based processing issue...
(For the record, neither am I but I know a few)
The tighter your filter, the longer it takes for the signal to get through and the higher the attenuation.
The tighter filter also tends to skew the signal more, thus more noise.
Higher attentuation means you need more powerful amplifiers and more sensitive detectors - meaning more noise.
More noise means it's harder to extract the useful data from it and may even make it unusable in places where it would be ok with a less-tight filter.
Given that GPS is a tiny and time-critical signal that's already dealing with a very high noise floor, what do you think they should do?
There is a reason why filters are the way they are, and if the leaked info is true then Lightsquared should never get FCC certification - and would not even get seriously considered in Europe!
I'm not surprised the CEO is pissed off though - that kind of data shouldn't be leaked before the FCC finally tell Lightsquared to go back to the drawing board.
It also probably means that they won't get much more funding, so said drawing boards may be unaffordable.
Band-pass filters are well understood and they work. You install them inline with the antenna and a good filter absolutely will solve the problem in your 2008 Mercedes.
These problems were solved in the 60s... but it's cheaper to build a less selective receiver. The lower the Q, the lower the cost.
There is no modulation overlap, GPSs just let almost anything in, and their amplification sections get overwhelmed by frequencies that are of no interest. Any application that uses a barn door as the first filter will have this... which is why commercial equipment doesn't.
It also depends on the difference in signal strength we're talking about. There's no such thing as a "brick wall" filter that will perfectly block an out-of-band signal. And we're talking about a *very* large difference in signal strengths here, and a very low tolerance for losses on the desired signal. It's hard to say without knowing exactly what kind of frequency difference is involved, but it wouldn't shock me if something unreasonably bulky like a cavity filter was necessary to make this work close to a transmitting site. Radio repeaters (which have to null out really strong signals on adjacent frequencies) often have cavity filters the size of dishwashers, which would really cut down the trunk space in your Mercedes. ;)
"The company maintains those results were based on hugely-inflated transmission power, and that the leak was deliberately "intended to damage LightSquared's reputation, spread false information in the marketplace, and prejudice public opinion against LightSquared"."
The same company that also claimed that there was no interference issues at all and the GPS manufacturers were wrong. Then tests were performed and there was interference just as the GPS manufacturers stated. Sorry LightSquared, but you have lied, lied and lied every chance you had.
The FCC needs to revoke their waiver for terrestrial use and force them to use the band for the purpose it was sold for. If they didn't like it, they shouldn't have bought it.
"The Pentagon has worried for months that a project backed by a prominent Democratic donor might interfere with military GPS. Now Congress wants to know if the White House pressured a general to change his testimony. "
http://www.thedailybeast.com/articles/2011/09/15/lightsquared-did-white-house-pressure-general-shelton-to-help-donor.html
The point that whistled over your head is that the idiot proposal to use filters on the receivers is a non-starter for almost all existing GPS receivers. It may work for future receivers (new production, starting in perhaps 2012), but few existing receivers would have the physical space to install a filter (let alone the monumental expense).
And the other missed point about the Mercedes is that some existing receivers have about a fifteen year lifespan *ahead of them*.
So, the solution is obvious: agree to LightSquared's proposal, but delay their start-up for about 20 years to allow their proposed filter solution to permeate the marketplace. That would be perfectly fair for such a disruptive approach.
on its technology were leaked from a government-assigned testing..."
Because yuns didn't keep up oy yur protection money payments, I mean make your political donations, in a timely manner, and according to the precedents long established, you are being chastised in The Chicago way.
"Band-pass filters are well understood and they work. >>>You<<< install them inline with the antenna and a good filter absolutely will solve the problem in your 2008 Mercedes."
No, LightSquared sends a man around and *he* installs the filter into my 2008 Mercedes. Along with millions others...
... to fix the LightSquared problem in GPS, at least without a major redesign of the system. Filters change the signals passing through them, even when those signals are the desired ones; that's particularly true near the cutoff frequencies. The sharper the cutoff, the greater the issue. Delays and phase shift on some satellite signals will change the accuracy of the fix.
In any case we should consider the big picture: A company came up with a business plan -- a satellite providing wideband wireless service. They got a frequency allocation that would have been okay for that idea but then found the plan wasn't economically feasible. They changed the plan to deliver most of the service via ground based towers. Experiments showed what theory had predicted: Serious interference with a critical established service. Now LightSquared wants the government to tell that existing service (and millions of customers) "You fix this." Perhaps morality doesn't matter but this makes zero economic sense -- absolutely ZERO.
What's the loss if LightSquared simply goes away, compared to the loss to GPS users and society if this can ultimately be fixed on that side? This is simply an effort to coerce a bailout of a well connected company that had an idea that doesn't work. You'd think we would have learned by now ...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
