Paolini: Planning the transition from WiMAX to LTE - the role of dual-mode devices

By Monica Paolini, Senza Fili ConsultingSince Yota, the Russian WiMAX operator, announced a year ago its plans to move to FDD LTE to everybody's surprise, there have been many more developments from operators and vendors that cast some light on the timeline and process of the transition to LTE and on the operators committed to it.

There are considerable differences among operators. Yota is still the most aggressive WiMAX operator planning a transition to LTE. Not only Yota is committed to move to FDD LTE, which is a more difficult transition as it requires paired spectrum, unlike WiMAX which uses TDD, but it has also entered an ambitious wholesale agreement with main Russian cellular operators, which will use Yota's network to host their services.

In Australia, VividWireless has completed an LTE trial with Huawei and in Malaysia P1plans for one later this year, also with Huawei. UQ in Japan is the strongest supporter of WiMAX 2 (i.e., IEEE 802.16m) along with its vendor Samsung.

Many operators are still in a wait-and-see phase, especially since TD LTE is not ready to be deployed yet, and they are still assessing the pros and cons of a transition, and the optimal timeline for the transition to WiMAX 2 or TD/FDD LTE.

In the U.S., Clearwire is conducting trials for both TD LTE and FDD LTE, but neither Clearwire nor its partner Sprint have announced their LTE plans, but it is widely expected that they will indeed move to LTE-the main question is whether it will be the TD or the FDD version.

In Ireland, Imagine is assessing different transition models, and may decide to move to either WiMAX 2 or TD LTE depending on how the ecosystem evolves and on market demand. Imagine is an interesting case because during the last decade it has accumulated a huge experience in operating multi-RAN networks and managing the transition from a RAN and subscriber perspective. It started with proprietary TDD and FDD wireless equipment from Navini (later acquired by Cisco) and Alvarion, and then transitioning to IEEE 802.16d (Fixed WiMAX) and finally to IEEE 802.16e (Mobile WiMAX, the dominating interface today). Imagine's acquisition of Clearwire Ireland brought into the network some proprietary NextNet (later acquired by Motorola) equipment and Motorola Mobile WiMAX equipment. Even though it has not yet committed to either wireless interface, Imagine is getting ready for a transition. Leo Lundy, Imagine Group CTO, has learned from experience that it is crucial to plan any transition very carefully to avoid churn and avoid service disruption. "Migration of the RAN is the easy bit. It is crucial to carefully prepare subscribers to the transition, provide them the information and support they need, in addition to the new devices". In previous transitions, Imagine did not see increased churn, as subscribers saw it as an upgrade and benefitted from higher speeds. At the same time, Imagine is already working to ensure that all the options will be available to them. "We are actively deploying  WiMAX base stations today, but we want to make sure that all the new equipment will allow us for a smooth transition to LTE, which will give us the ability to run WiMAX and LTE concurrently," Mike Stacey, Imagine CTO, said.

Imagine's case is also interesting because they use the 3.5 GHz spectrum, a band that was not initially included among the TD LTE specifications and that is not a top priority for vendors. The substantial amounts of spectrum available worldwide-underused and cheaper than cellular FDD bands-has created an interesting buzz in the latest months about the opportunities for TD LTE in this band. LTE in the 3.5 GHz band can be used to provide additional capacity in urban environments both in outdoor or indoor environment, as well as fixed services.

Multimode devices will facilitate the transition to LTE

From the RAN perspective, the adoption of base stations upgradeable to LTE gives operators which are committed to move to LTE the flexibility they need to plan the transition on their own terms, deciding which type of overlay network to build, and for how long to operate the two networks concurrently.

What can they do on the device front? WiMAX operators need to ensure service continuity to avoid churn. They also need to make sure that subscribers' terminal devices (either desktop modems, dongles or phones) support the right interfaces-WiMAX today, WiMAX and LTE during the transition period, and eventually LTE only.  This is particularly important for mobile devices which connect to the network at multiple locations, where different interfaces are available.

If using devices that support only WiMAX or LTE, the operator at some point has to provide a new device that supports LTE to subscribers. In many cases, they will need to shoulder the cost of the new device and need to coordinate tightly the device swap.

The alternative is to use multimode devices that support both WiMAX and LTE, which can be introduced in the market organically ahead of the LTE rollout. When LTE becomes available, the new devices see the new network and connect to it, without requiring a hard swap. From a subscriber-and a subscriber support-perspective, multimode devices can simplify the transition process. But is this a cost effective solution from a financial perspective?

We looked at this issue in a financial model described here, sponsored by Sequans, and found out that dual mode devices reduce the cost of device subsidies and provisioning, as operators can introduce LTE devices as part of the organic device replacement (Figure 1). Assuming an additional cost of $10 to support LTE in addition to WiMAX, the higher cost of multi-mode devices is easily recouped by avoiding the forced replacement of devices when the WiMAX networks are turned off.

More interestingly however, the availability of dual-mode devices has the strongest impact on the transition in the RAN. As subscribers are able to use both interfaces for a long time, operators can plan for a slower transition, which keeps risks down and required less resources, and which, even more importantly, distributes the cash flow over a longer period of time. As a result, the upfront financial requirements for the upgrade to LTE are much more limited. This is important because even if operators currently deploy only software-upgradeable base stations, most base stations already deployed are not upgradeable. During the first year of the transition, WiMAX operators can save up to 19% of capex and opex when using multimode devices (Figure 2).

But there is a caveat. To maximize the benefits of multimode devices, operators need to start introducing them before they roll out the LTE infrastructure, to get multimode devices in their hands when they sign up or when they are ready to upgrade to a new device.

Figure 1. Comparison of the device types in use among subscribers in a scenario where the operator uses dual-mode (WiMAX and LTE) devices and in another where all devices are single-mode (WiMAX or LTE). During Year 1, the operator introduces dual-mode devices, but does not deploy LTE in the RAN. During Year 2 and Year 3, both WiMAX and LTE networks are concurrently operational. In Year 4, the WiMAX network is switched off.  Source: Senza Fili Consulting

Figure 2. Comparison of the single-mode and dual-mode cases. Positive cost savings values denote lower costs for the dual-mode case. Source: Senza Fili Consulting

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.