Thursday, July 7, 2011

Andrew Seybold: LightSquared System is a Potential Disaster

This post was used with Andrew Seybold's permission, from Andrewseybold.com.

 Okay, I admit it. I don’t believe LightSquared should be permitted to use what was supposed to be satellite spectrum for a terrestrial broadband network. Not only that, I don’t believe LightSquared has a sound business plan. Building more than 40,000 cell sites, maintaining them, and reselling the bandwidth to others who want to sell it to their customers does not pencil out in my book. The margins will be too slim, especially given the fact that prices for both voice and broadband services keep falling in the United States so margins will continue to be squeezed. But apart from a faulty business plan, the main reason I am opposed to LightSquared’s plan to build this network is that if there is the slightest chance it will interfere with GPS receivers, it simply should not be permitted to be built.


"...if LightSquared is permitted to build out its system it can only result in a potential disaster that will take years and cost billions of dollars to resolve if it can be resolved at all."


The history of wireless is littered with examples where the FCC acted on a request by a vendor, approved the deployment of a system, and later learned that the new system created interference to other services. The most notable of these mistakes by the FCC is the ongoing saga of the Nextel service. Nextel started out providing Specialized Mobile Radio services (SMR) for two-way radio systems as Fleet Call and then was re-invented to become a full cellular service but not in a cellular band. In fact, as Nextel went around the country and bought up other SMR operators’ systems (each one covering a specific local area), it amassed channels that were intermingled with other SMR operators, as well as fire and police radio systems.

Nextel’s founders worked hard to convince both Congress and the FCC that this type of system would be compatible with the existing users in the 800-MHz frequency band. Since there were only two cellular operators for a city at that point in time, both the FCC and Congress approved the concept in order to bring more competition into the cellular market. During the discussions, fire, police, and many SMR operators pointed out that Nextel’s cell towers would cause interference to these other services but their warnings fell on deaf ears.

If you were the FCC Chairman, would you permit LightSquared to continue? I, for one, would not.

The system was rolled out and, in fact, it caused significant interference to others in the same band including public safety. The reason was simple: SMR and public safety systems made use of tall towers or mountain top sites and only a few of them per city. The radios were high power and the systems worked well. Then Nextel began building towers using the cellular model, which is lots of towers at low locations in order to provide service over wide areas. The result was that when an SMR or public safety radio was within sight of a Nextel cell site, or even close to one, the cell site put out so much power 24/7 that the Nextel signals interfered with public safety and SMR radios and in many cases put public safety personnel in danger.

The solution, which is still ongoing today, was to re-band this portion of the spectrum moving Nextel’s frequencies to a contiguous block of spectrum instead of permitting Nextel to be intermingled with the other services. A lot of this re-banding, as it is called, has been completed but not without Nextel and now Sprint/Nextel having to spend billions of dollars and huge logistical issues for both public safety and SMR systems. Some areas of the country have not completed the re-banding process, including along both our borders where the process must also include coordination with either Mexico or Canada, and then there are many areas where there are disputes about the cost of re-banding or radio replacement.

Nextel got its way based on the belief in Washington at the time that we needed more than two cellular operators in order to create more competition and therefore lower prices. After Nextel was approved, the PCS spectrum (1900 MHz) was auctioned again to ensure more competitors in every market. Fortunately, the 1900-MHz users were relocated by the auction winners so there were no further interference issues.

Fast forward to 2011. The FCC’s broadband report says that we need more commercial and unlicensed broadband spectrum. The FCC’s OBI (Omnibus Broadband Initiative) has cited that in the United States we will be more than 200 MHz of broadband spectrum short to meet the demand by 2013, and the FCC says it will “find” 300 MHz of spectrum within five years and another 200 MHz within another five years. Further, there are two other attitudes in DC that influence the decision-making process. The first is the belief (which I disagree with) that broadband services will replace all wireless services going forward and therefore broadband is the most important form of wireless communications going forward. The other belief is the more competition there is in every market in the United States the lower the pricing will be for all customers. Neither of these beliefs is grounded in reality or can be proven but that does not matter. What matters is perception.

When LightSquared went to the FCC, as you can see, the FCC was already predisposed to approve both more broadband spectrum availability and more competition, especially since LightSquared is to be a wholesaler of broadband services. This will permit smaller competitors to compete against the larger nationwide network operators, at least on a local basis. It is no wonder that the FCC issued LightSquared its waiver to convert satellite spectrum into a full-blown terrestrial network as opposed to a previous ruling that would permit LightSquared to augment its satellite coverage with a few terrestrial sites. The warnings from the experts, once again, went unanswered, and the FCC issued the waiver.

Now, after the fact, the FCC and LightSquared are facing opposition from Congress, a collation of GPS companies, the aircraft industry, the Department of Justice, the Public Safety community and even the farmers who use precision GPS in growing their crops. Perhaps the good news is that the network has not been built yet and the hue and cry is growing louder by the day. The results of the first set of tests proved, beyond a doubt, that the first portion of the spectrum LightSquared intended to use for terrestrial broadband would have wiped out GPS reception within a few miles of each cell site and in the air as well.

Now LightSquared has filed its report, all 1200+ pages of it, and working my way through it is a challenge. It has said it will move to a different portion of its spectrum that is further away from the GPS band, and claims that in this portion of the band it will not interfere with 99.5% of the GPS devices. The fact that it WILL interfere with even, as they state, 0.5% of the devices should be reason enough not to grant permits to build the network. The other item of interest to me is that LightSquared now claims that it is the fault of the GPS vendors for building GPS receivers without the proper type of filtering in them.

I take issue with this point because the GPS band is in the satellite band and the GPS system was never intended to have to worry about interference from terrestrial transmitters. Because the GPS band is 20 MHz wide and uses spread spectrum technology, filtering these devices will take a lot of engineering if it is even possible. Meanwhile, if LightSquared is permitted to build out on this spectrum there is no guarantee there will not be interference to more than the 0.5% of devices it claims. The statistics it cites are not based on extensive testing, nor is this number substantiated properly in anything I can find.

The bottom line, to me, is that when non-technical people make decisions about wireless technology, which they do not understand or comprehend, and for the reasons stated above, if LightSquared is permitted to build out its system it can only result in a potential disaster that will take years and cost billions of dollars to resolve if it can be resolved at all.

There is no doubt that we need additional broadband spectrum to meet the growing demand for broadband services. However, endangering GPS, which is used for many different things including locating us when we call 911, is not the right way to go about this. We must take all the time needed to make 100% sure that there is no potential of interference, not 99.5%. I believe the best option is to deny LightSquared’s waiver, withdraw it, and let it find or purchase other spectrum that might be available at auction.

Clearwire is trying to raise money and has expressed a willingness to sell some of its spectrum holdings. I suggest that LightSquared make a deal with Clearwire and see if it can make its business model work on spectrum that does not have the potential to disrupt so many things we do each day, including saving lives. If you were the FCC Chairman, would you permit LightSquared to continue? I, for one, would not.

Andrew M. Seybold

No comments:

Post a Comment