WiWavelength

by Andrew J. Shepherd Sprint 4G Rollout Updates Wednesday, February 15, 2012 - 4:45 PM MST Clearwire released its fourth quarter and full year 2011 results in a conference call with investors, analysts, and the media this afternoon. S4GRU was on the call to bring you this report. Clearwire highlighted its 8-K report with the following statistics: Record Fourth Quarter 2011 Revenue of $361.9 Million, Up 107% Year Over Year From $175.2 Million Full Year Revenues of $1.25 Billion, Up 134% Year Over Year From $535.1 Million Full Year Wholesale Revenues Up 876% Year Over Year to $493.7 million 2011 Total Ending Subscribers of 10.4 Million, Up 140% Year Over Year from 4.3 Million Achieves Positive Quarterly Adjusted EBITDA For the First Time of $22.5 Million Average Smartphone 4G Usage Increased 88% Year Over Year in Fourth Quarter 2011 Much of the rest of the report is focused on business metrics that may not be of particular use to anyone without an investment in Clearwire. But we did pore over the report to glean the following numbers of interest to S4GRU readers: BRS 2500-2600 MHz licensed spectrum valuation remained steady at $4.3 billion EBS 2500-2600 MHz spectrum lease costs totaled $309 million for 2011 WiMAX covered POPs increased year over year from 112 million to 132 million but plateaued at that level by the end of the second quarter Wholesale (e.g. Sprint) churn almost doubled from 1.5 percent to 2.9 percent during the fourth quarter To provide some analysis of the four points above, first and second, Clearwire holds an average of ~160 MHz of BRS/EBS spectrum bandwidth in the top 100 markets. However, as noted above, some of this spectrum (EBS) is leased from educational institutions, not licensed directly to Clearwire. Additionally, higher frequency spectrum is generally less valuable than is lower frequency spectrum. Otherwise, Clearwire's ~160 MHz of spectrum would be valued in the tens of billions of dollars. Third, as Robert has detailed in a forum post about "protection sites," Clearwire faced a May 1, 2011 FCC deadline to demonstrate at least minimum coverage in numerous Basic Trading Areas (BTA) across the country. As a result, Clearwire lit up numerous license "protection sites" around the country during the first few months of last year, leading to the 20 million POPs increase that then stalled for the remainder of the year, as Clearwire made the decision to cease WiMAX deployment and switch to LTE. Fourth, Sprint is Clearwire's largest wholesale partner. Any Sprint retail subscriber who has a WiMAX capable device is technically also counted as a Clearwire wholesale subscriber. While Clearwire churn remained relatively flat through the first three quarters, it spiked in the fourth quarter. Clearwire attributed the increase in wholesale churn in large part to Sprint offering the iPhone 4S, which is not WiMAX capable. Lastly, Clearwire addressed some of its plans for its TD-LTE 2500-2600 overlay. Clearwire reiterated its commitment to the TDD "ecosystem," alongside strategic partners China Mobile, et al., and to TDD/FDD interoperability that will allow for seamless roaming on both types of LTE networks. Clearwire expects to start build out on its LTE Advanced ready TD-LTE overlay in the second quarter, spending $400 million this year and $200 million next year, keeping costs low because much of the WiMAX infrastructure can be reused for TD-LTE. Build out goals in phase one include 8000 TD-LTE sites, at least 5000 of which are to be live by June 2013. In the WiMAX build out, Clearwire selected its own independent site locations, and this led to great inconsistencies between Clearwire and Sprint coverage. But in the TD-LTE build out, Clearwire and Sprint will work together to identify sites within the Sprint portfolio that exhibit the "highest 4G data usage potential" with fallback to the Sprint FDD LTE 800/1900 network outside of those Clearwire data "hotspots." Finally, both Clearwire and Sprint project multi-band, multi-mode TDD/FDD LTE devices that can utilize the Clearwire TD-LTE overlay to be available by June 2013, by the same time that the first 5000 sites should be online. Source: Clearwire

I will stand to be corrected, but Sprint has yet to deploy any LightSquared LTE 1500-1600 infrastructure, as LightSquared has yet to garner FCC approval. Sure, Sprint has included space for LightSquared in its Network Vision plans. However, Sprint does not have to execute that element of the plan if LightSquared fails to hold up its end of the bargain. So, most likely, the mounting poles set aside for LightSquared will go empty for the time being. Similarly, you will note that the legacy CDMA1X 1900 antennas will remain mounted only during a transition phase, then they will be retired. And those mounting poles will become empty. No big deal. AJ

Indeed. And according to some posted Network Vision schematics, the LightSquared L-band 1500-1600 MHz antennas had to be additional and separate from the Sprint SMR/PCS 800/1900 MHz antennas. But if T-Mobile were to contribute PCS 1900 MHz spectrum to a joint venture with Sprint, that would not require much if any additional or separate infrastructure, as the Network Vision dual band antennas would already be in place. Thus, T-Mobile spectrum would be an easy fit. Ah, you are going to have to wait for part three of the article series next week. As part of my discussion "why" Sprint-T-Mobile could/should pursue an LTE network sharing agreement, I will address how Sprint and T-Mobile could maintain differentiation while still selling access to the same LTE network. And, yes, differential access to LTE 800 and/or TD-LTE 2500-2600 is one potentiality that I plan to cover. AJ

At the very least, I can tell you that Sprint's PCS 1900 MHz licenses in Raleigh-Durham have not changed. That market I know off the top of my head. Sprint still holds its original PCS D block 10 MHz license won at FCC auction and a PCS A5 block 10 MHz partition/disaggregation acquired from the old AT&TWS. Since both the PCS A5 and D blocks are contiguous, Sprint in Raleigh-Durham controls a solid 20 MHz block of spectrum, in which it can deploy a max of seven CDMA2000 carriers. AJ

Sgt., I did that legwork on Sprint's PCS 1900 MHz spectrum holdings about nine years ago. You may have already come across my old spectrum catalog and map at wirelesswavelength.com. Unfortunately, I do not have more recent info in finished form (only the raw data that I have compiled for this article and others). That said, my nine year old info is still ~90 percent accurate. AJ

Just as LTE is 5 MHz x 5 MHz, 10 MHz x 10 MHz, etc., keep in mind that CDMA1X and EV-DO are 1.25 MHz x 1.25 MHz. In other words, do not forget to count both uplink and downlink. Thus, each carrier channel is 2.5 MHz bandwidth in total. Then, do the math (e.g. 2.5 MHz x 6 = 15 MHz). And a good rule of thumb is that a CDMA2000 deployment requires guard bands equivalent to the bandwidth of one carrier channel (i.e. 2.5 MHz, split 625 kHz at the top/bottom of the uplink and top/bottom of the downlink). Do the math again (e.g. 15 MHz + 2.5 MHz = 17.5 MHz), and there you have it. The uplink/downlink calculations above are equally relevant here. If Sprint's final rebanded SMR 800 MHz allocation is 14 MHz (i.e. 7 MHz x 7 MHz), and Sprint deploys one 5 MHz x 5 MHz LTE channel, then Sprint will be able to support concurrently just one 1.25 MHz x 1.25 MHz CDMA1X Advanced carrier. AJ

Migrated from Original Forum. Originally Posted 5 February 2012 I am curious why everyone is so agog over VoLTE. Certainly, the prospect of ubiquitous mobile VoIP is, at least technologically speaking, an interesting paradigm shift. But do we really expect an enhanced end user experience from VoLTE? CDMA1X spreading gain, soft handoff, and noise suppression make it almost ideally suited to moderate bit rate voice transmission. So, VoLTE has a high standard to follow. AJ

Migrated from Original Forum. Originally Posted 28 January 2012 I could make a cogent argument that the iPhone is one of the worst things (as contrarian as that might sound) to ever happen to the overall well being of the domestic wireless industry. It would likely make my worst five (in no particular order): iPhone VZW-Alltel merger Cingular-AT&TWS merger sunset of the spectrum cap elimination of the Cellular 850 MHz cross interest rule AJ

Migrated from Original Forum. Originally Posted 18 January 2012 No, I think that any conspiracy theories are unfounded. GPS manufacturers have nothing to gain nor lose if LightSquared truly can ameliorate the ATC interference issues. But this is a legitimate and serious technical problem. And if LightSquared cannot prevent interference beyond a shadow of a doubt, then LightSquared has no choice but to abide by the original regulations of its L band satellite spectrum. AJ

Migrated from Original Forum. Originally Posted 28 January 2012 I would not be surprised if the remaining iDEN traffic is significantly shunted to SMR 900 MHz now, though I would love to know the current ratio of SMR 800 MHz to SMR 900 MHz iDEN channels on a market to market basis. AJ

Migrated from Original Forum. Originally Posted 15 January 2012 I would be careful not to extrapolate too much from the iPad 3. Chipset support for multiple band classes is not really the issue; most Qualcomm MSMs, MDMs, or SoCs already support far more numerous bands than are ever enabled in any one device. Rather, antenna support seems to be the sticking point. Handheld devices that strive to be compact have only so much space available for cellular antennas, which grow in size and/or number to support additional bands. LTE further complicates matters, since it is MIMO enabled. And, as far as I know, all LTE devices released on VZW or AT&T thus far support 2x1 or 2x2 MIMO, thus they require two Rx diversity antennas. As this pertains to the iPad 3, it is a tablet, so it should have far more space available for multiple antennas than will the eventual iPhone 4S successor. Additionally, if the iPhone 4S replacement does gain LTE capability, I would not be surprised if the current convergence to a single inventory iPhone 4S global model does not last, and Apple is forced to return to a multi inventory system just to support the innumerable LTE band classes being deployed worldwide. AJ

Migrated from Original Forum. Originally Posted 13 January 2012 Yes, I have long wondered this myself. Many think that Sprint overpaid for access to the iPhone for the next several years. So, for that huge outlay, what does Sprint get in return? As unlikely as it seems, an exclusive iPhone could be the return on that investment, could be the card up Sprint's sleeve. AJ

Migrated from Original Forum. Originally Posted 10 January 2012 I suspect that we will know for certain the supported LTE band class(es) only when Sprint's first two Samsung and LG LTE handsets hit the FCC OET authorization database. Hopefully, those authorizations will come through in the next several weeks. AJ

Migrated from Original Forum. Originally Posted 10 January 2012 I am uncertain how much market "pre-seeding" we will see with LTE devices. Keep in mind that current VZW LTE devices support only LTE band class 13 (Upper 700 MHz C block 22 MHz). So, if the VZW-SpectrumCo-Cox AWS 2100+1700 MHz spectrum transactions gain approval, even VZW will require a whole new set of LTE band class 4 capable devices. Also, Sprint did "pre-seed" the market with CDMA band class 14 (PCS "G" block 10 MHz) devices. Of course, CDMA band class 14 now will never see the light of day in the US, as Sprint has decided to use that spectrum for LTE instead. Hence, all of that "pre-seeding" proved pointless. AJ

Migrated from Original Forum. Originally Posted 10 January 2012 In an interview with Phonescoop.com, Sprint spokesperson David Owen confirms that Sprint will launch LTE this year in only its PCS 1900 MHz spectrum. So, do not expect anything beyond CDMA1X Advanced and/or EV-DO Rev B in SMR 800 MHz spectrum until at least 2013. The interview also seems to suggest that initial Sprint LTE devices will be LTE 1900 only, though that point is somewhat vague. But future support for LTE 800 and/or TD-LTE 2500-2600 in the first go round of Sprint LTE devices may be doubtful. http://www.phonescoo...icle.php?a=9639 AJ

Migrated from Original Forum. Originally Posted 28 January 2012 pyroscott, an interesting companion to this thread would be the obvious "Where would Sprint be today if it had not merged with Nextel?" Many seem to feel that the Nextel merger and its negative fallout caused Sprint to miss a prime link up opportunity with Alltel (after Alltel had solidified itself as the rural roaming carrier with its 2005 acquisition of WWC). Had Sprint been able to hook up with Alltel, then both VZW and AT&T (because of the WWC GSM roamer network) would have had to come to Sprint for roaming agreements, giving Sprint an advantageous negotiating position that is has rarely had. And, anecdotally, some say that Alltel CEO Scott Ford was practically begging then Sprint CEO Gary Forsee to make an offer before private equity first, then VZW second came in and took Alltel off the market. Care to start another thread? Or keep the discussion here? AJ

Migrated from Original Forum. Originally Posted 28 January 2012 Robert, hindsight is always 20/20, of course. But if Sprint had fully known earlier to what extent 800 MHz reconfiguration was going to be such an albatross and how ungrateful Nextel subs would unfairly blame Sprint for all ills and defect en masse, then Sprint should have just shut down iDEN within two years of the closing of the merger. Sprint could have given the iDEN subs an ultimatum (since most of them ended up leaving anyway) and used the iDEN retirement by 2008 as a means to expedite the 800 MHz rebanding process, followed quickly by a CDMA band class 10 roll out. Had Sprint done all of above, we could have had CDMA1X 800 overlay on the national network by 2010. AJ

Migrated from Original Forum. Originally Posted 27 January 2012 My guess is that VZW would have been enticed by Nextel's then strong class of business users plus PTT technology and acquired Nextel not long after Sprint did, say by 2006. In fact, though I can now find little support for this, so it may be just a figment of my memory, I recall hearing rumors circa 2004 that VZW was readying a bid for either Nextel or Sprint. Hence, Sprint had to beat VZW to the punch because 1) Sprint wanted to acquire Nextel and 2) Sprint feared for its own independence. Regardless, I strongly believe that Nextel would not have remained a standalone carrier for very long. No WiDEN, FLASH-OFDM, nor 6:1 vocoder was going to be enough to solve Nextel's coming 800 MHz reconfiguration, capacity, coverage, and obsolescence issues. Those problems were already baked into Nextel when Sprint arrived. Unfortunately, most subscribers were unaware of the impending storm. And, in classic post hoc fallacy fashion, they blamed Sprint for everything bad that happened to Nextel and iDEN after the merger. For that reason, I have long thought of many (but not all) Nextel subs as a bunch of ingrates who did not understand that Sprint tried to throw them a lifeline to get off their sinking ship. That so many Nextel subs, largely out of spite, have since churned from Sprint without realizing that it was really their former Nextel leadership and their beloved iDEN technology that put Nextel on the path to ruin, well, that is just sad. AJ

Migrated from Original Forum. Originally Posted 28 January 2012 If the two options for Sprint were 1) to use its BRS 2500-2600 MHz spectrum (as the FCC merger consent required) and WiMAX was the only viable option at the time or 2) to hold out for LTE (or UMB) and risk having its BRS licenses terminated, then, yes, WiMAX was not only a good decision but also the right one. AJ