by Andrew J. Shepherd

Sprint 4G Rollout Updates

Friday, February 10, 2012 - 12:22 PM MST

In part one of this series, we discussed how Sprint could theoretically leverage its long held PCS A-F block spectrum to deploy a second 5 MHz x 5 MHz LTE channel in many markets alongside the 5 MHz x 5 MHz LTE channel that Sprint plans to roll out in its unused, nationwide PCS G block 10 MHz spectrum. Here is the rundown:

.

• Verizon Wireless (VZW) and AT&T Mobility (AT&T) have the Upper/Lower 700 MHz spectrum right now to deploy 10 MHz x 10 MHz LTE in most/all markets and gain a significant leg up on the LTE competition.
  
• Sprint cannot match that LTE bandwidth in its PCS G 10 MHz license alone.
  
• Sprint’s hosting agreement with and wholesale access to LightSquared LTE bandwidth is all but dead.
  
• Sprint will not likely be able to utilize its SMR 800 MHz spectrum for LTE nor Clearwire’s BRS/EBS 2500-2600 MHz spectrum for TD-LTE until 2013-2014.
  
• Sprint can and does operate many of its CDMA2000 markets (e.g. Chicago, San Francisco, Houston, Atlanta, etc.) in only 20 MHz of licensed PCS A-F block spectrum.
  
• Sprint has numerous markets in which it holds 30 MHz of licensed PCS A-F spectrum, thus could potentially set aside that “surplus” 10 MHz for a second 5 MHz x 5 MHz LTE channel in those 30 MHz markets.
  

And that is just about where we left off at the end of part one.

To continue, in the top 100 markets, Sprint holds 30 MHz of PCS A-F block spectrum in 54 of those markets. Those are the markets in which Sprint could most readily utilize a 10 MHz “surplus” for a second LTE channel. See the graph of Sprint markets with “surplus” spectrum and note those markets which hit 10 MHz:

So, Sprint ostensibly could augment capacity with a second LTE channel in only about half of all top 100 markets (including just 11 of the top 20 markets). That could be a problem for those other major markets necessarily left out. But, as hinted at the end of part one, that problem would be solvable through a spectrum and LTE network sharing agreement with another carrier. And, yes, that other carrier would be T-Mobile. To be clear, it would not be a merger. Both carriers would continue to operate separately, but they would share the same LTE network on Sprint’s Network Vision platform. So, let us focus on “how” such an arrangement could be possible through Sprint’s and T-Mobile’s compatible spectrum holdings.

First, T-Mobile, too, holds “surplus” PCS A-F block spectrum. T-Mobile operates its 2G GSM network exclusively in its PCS spectrum, reserving its AWS 2100+1700 MHz licenses for its 3G/4G W-CDMA network overlay. Similar to Sprint, T-Mobile can and does operate its (GSM) network in only 20 MHz of spectrum in many very large markets (e.g. New York, Los Angeles, Boston, etc.). Also similar to Sprint, T-Mobile is licensed 30 MHz (or more) of PCS spectrum in many top 100 markets. See the graph of T-Mobile “surplus” PCS A-F spectrum; again, note the 10 MHz (or more) markets:

Something interesting happens when we combine the first two graphs. If Sprint and T-Mobile were to have joint access to their “surplus” PCS A-F spectrum, the number of markets in which a prospective LTE network partnership would have at least 10 MHz of spectrum for a second LTE channel rises to 81 markets (including all top 10 markets and 18 of the top 20). See the combined graph:

Better yet, because of the particular combinations of PCS A-F licenses that Sprint and T-Mobile hold and the oft symbiotic ways that those licenses line up, a Sprint-T-Mobile partnership could actually roll out a 10 MHz x 10 MHz LTE channel in top 20 markets Chicago, Philadelphia, Detroit, Dallas, St. Louis, Minneapolis, Denver, and Seattle – matching or even exceeding the LTE deployments of VZW and AT&T in those markets.

Furthermore, T-Mobile subscribers are increasingly adopting W-CDMA capable devices, thereby shifting much of their usage to the W-CDMA overlay and reducing the load on the GSM network. As a result, if T-Mobile could pare down its GSM spectrum utilization to 10 MHz per market, then T-Mobile could contribute 20 MHz of “surplus” spectrum in its 30 MHz (or more) PCS markets. And the Sprint-T-Mobile spectrum complement would go through the roof. The LTE partnership would then have sufficient spectrum to launch a second 5 MHz x 5 MHz LTE channel in all top 100 markets. And the markets in which it could deploy a 10 MHz x 10 MHz LTE channel would swell to include the eight aforementioned top 20 markets plus Houston, Atlanta, Pittsburgh, Kansas City, Buffalo, New Orleans, Portland, Salt Lake City, Austin, et al. Again, see the combined graph (now with T-Mobile “surplus” at greater than 10 MHz):

The numbers largely speak for themselves. Together, Sprint and T-Mobile could take advantage of their complementary spectrum to build a broader, higher capacity LTE network than either could alone. So, that is “how” a partnership could materialize. Next time, we will delve into “why” a Sprint-T-Mobile spectrum and LTE network sharing agreement would make a lot of sense for both carriers. A preview of one very big reason “why” – yes, the impending LTE iPhone.

Sources: FCC, author’s notes