Finally Clearwire has publicly announced its commitment to move to LTE. The decision to adopt LTE was a widely expected, and probably an unavoidable one, but there were (and still are) many open questions on how the transition will take place, in terms of timeline, technology, the relationship with Sprint, and, perhaps more crucially, of funding.
Which LTE?
True to form, Clearwire did not choose to go for the mainstream version of LTE--the frequency-division duplexing, or FDD, version--the one adopted by Verizon and that will also be adopted by Sprint and most mobile operators in Western Europe, Japan and South Korea.
Instead, Clearwire will roll out the time-division duplexing version, TDD LTE (also referred to as TD LTE), the adoption of which has been initially driven by mobile operators in China and India; but interest is quickly spreading beyond emerging markets, as shown by the fact that three of the five steering committee members of Global TD-LTE Initiative (GTI) are Clearwire, Softbank and Vodafone, all from developed markets with strong FDD commitments (the other two steering committee members are China Mobile and Airtel).
Why TD LTE?
For most operators, there is no choice between TD and FDD LTE, as the selection is dictated by the spectrum they have. Most cellular spectrum comes in paired channels (one for the uplink, and one for the downlink) with the requirement to use with FDD technologies. There are also substantial TDD spectrum allocations, which are to a large extent underutilized and are less expensive. In many markets, these bands were targeted for WiMAX--also a TDD technology--deployments. Clearwire finds itself in the most privileged position of not only having a huge amount of spectrum, but also of being able to decide whether to use TD or FDD interfaces.
From a technology perspective, TD LTE is an easy call for Clearwire. TDD technologies are better suited to accommodate asymmetric data traffic, which has a much heavier downlink load, increasingly driven by streaming video and audio applications. FDD interfaces have dominated in the cellular network because they are designed to carry the inherently symmetric voice traffic.
Coexistence and spectrum efficiency
Clearwire plans to establish the LTE network as an overlay over the existing WiMAX network, initially targeting the parts of its current footprint where demand is higher and keeping both networks active side-by-side (see here for more information on migration paths from WiMAX to LTE). This is an eminently sensible approach that enables Clearwire to continue to leverage the infrastructure it has on the ground, while gradually deploying LTE as needed to meet the growing demand from existing subscribers and the new demand from new, LTE-only subscribers. A fast switch to LTE would require a huge operational and funding effort for a network of Clearwire's size, even if all the deployed equipment was upgradeable (and it is not).
The need to coexist, at least initially, with WiMAX makes TD LTE more attractive than FDD LTE, as it makes the upgrade to LTE easier (TD LTE and WiMAX use a very similar wireless interface) and spectrum can be used more efficiently. Mixing TDD and FDD technologies in contiguous channels requires the creation of guard bands of spectrum that cannot be used otherwise to avoid interference.
The TD LTE ecosystem and market
There are however some possible downsides for an operator--especially in the U.S.--to deploy TD LTE. To date, Clearwire is the only major operator in the U.S. committed to the non-FDD version, and this may limit the market appeal of the network. Globally, TD LTE is lagging a bit behind FDD LTE, although TD LTE may pick up speed more quickly than LTE as its prime markets--China and India--are also the largest ones worldwide. Nevertheless, this may initially limit the availability of devices, especially smartphones and tablets--and this could be especially an issue if competing operators in the U.S. market were to choose to adopt devices that did not support TD LTE, either because they see no need for it, or to fend off competition from Clearwire and Sprint.
On the positive side, support for TD LTE basically comes for free in devices that support FDD LTE, as the same chipset can support both. As there will be a large market for both technologies and in many markets the two technologies will coexist, manufacturers are likely to support both TD and FDD LTE in their devices as this widens their addressable market. If this is the case, Clearwire should not be negatively affected by the adoption of TD versus FDD LTE--and, in fact, subscribers do not even need to be aware of it of the fact that two versions of LTE even exist.
Upgrade or extend?
One of the large gray areas in the announcement is how WiMAX and LTE will coexist and how the transition will be managed. To some extent, this is due to the funding uncertainty, which obviously precludes Clearwire management from committing to a timeline and transition plan.
Another area that is not clear is the expected balance between upgrading the current network and the deployment of new infrastructure. Both will be needed. Part of the equipment cannot be upgraded, and this may lead to sites where WiMAX and LTE will coexist as two independent interfaces. It is not clear, though, whether all the upgradeable equipment will be upgraded or not. To meet its commitment to support existing WiMAX subscribers, Clearwire will have to support at least for some time (i.e., until WiMAX subscribers have LTE devices) both WiMAX and LTE. While it is possible to support both WiMAX and TD LTE with the same equipment, this option reduces the WiMAX capacity to make room for LTE. If LTE is needed to increase capacity in high-traffic areas to create what John Stanton, Clearwire's chairman of the board, calls the "Switzerland of 4G," wouldn't Clearwire choose to install additional base stations co-located with the WiMAX ones? Clearwire's price tag of $600 million for the transition to LTE seems to depend on equipment upgrades, but the scope for upgrades may be limited if an increase in capacity is main driver behind the LTE rollout.
Of course, there will be synergies and cost efficiencies in deploying an overlay network over building a new market. Site locations can be shared, along with some hardware and backhaul. But installing an overlay network is a more expensive proposition than upgrading an existing one--both in terms of capex and opex.
The transition from the network and device side
The uncertain balance between upgrades and expansion also raises question on how the transition to LTE will take place from the subscriber perspective. Clearwire's announcement focused on the transition to LTE from the network side--the device side received much less attention, even though it also plays a crucial role (see here). If the WiMAX and LTE networks are treated as targeting non-overlapping customer segments, Clearwire will need more combined network capacity to accommodate traffic and to allow an organic migration to LTE. Alternatively, Clearwire and its partners could accelerate the transition to LTE, by pushing LTE devices to their current WiMAX subscribers. A third possibility is to start seeding the market with multimode devices--as UQ is doing in Japan as it plans to transition to WiMAX 2 from WiMAX--starting ahead of new network launch to make the transition to WiMAX 2 seamless and easier to manage. It would be interesting to know what Clearwire's (and Sprint's) plans are in this regard, as the device strategy has an effect on the network plans due to the fact that the devices that subscribers use determine the relative traffic loads of the WiMAX and LTE networks.
Funding
Of course this is all academic, since Clearwire does not currently have sufficient funding for the LTE rollout. While funding is not causing an immediate delay to the project--it is still early days for TD LTE--it may delay it or alter its scope. The rumored acquisition by Sprint or a change in ownership role may also cause a change in plans. Even so, the announcement brings some clarity into the long term plans of an operator that in the past months had to focus on short-term cost-cutting measures and it is refreshing to get a glimpse of the long-term vision.
Monica Paolini, PhD, is the founder and president of Senza Fili Consulting and can be contacted at [email protected]. Senza Fili Consulting provides expert advisory services on wireless data technologies and services.
A site at Clearwire's trial in Phoenix conducted to test coexistence of WiMAX, and TDD and FDD LTE. Source: Clearwire.
TD LTE trials and planned networks. Source: GTI.