According to the GSMA, there are now nearly 500 commercial LTE networks around the world and more than one billion active subscriptions. While most of these are data-only, things are changing rapidly because operators are also busily adopting Voice over LTE (VoLTE). At the same time, they are also beginning to invest in a new, virtualised network architecture, which promises fundamental changes to their systems, operations and processes. The two issues are intimately related and need to be carefully aligned, particularly when it comes to the validation of new services within the emerging architectural framework. This article explores a number of challenges associated with both initiatives and proposes a common solution that will help operators through this challenge.

Today, operators are compelled to migrate to VoLTE, as they need to activate dedicated voice services in their LTE networks, rather than defaulting to circuit-switched fall-back (CSFB) mode. VoLTE offers superior quality when compared with earlier generations of voice services and, as many subscribers are familiar with highly optimised OTT voice services that are based on free (or freemium) applications, they are likely to be disappointed by the degraded experience that CSFB offers.

The only way to solve this problem is to move to VoLTE. However, this presents a number of challenges. Operators have to introduce a complex new service but, at the same time, they also have to manage yet another network layer in addition to existing LTE, 3G and 2G deployments. VoLTE will deliver the expected performance improvements but this also raises questions regarding subscriber expectations.

First, unless VoLTE coverage is ubiquitous, user mobility means that individual subscribers will move between coverage areas. VoLTE users must be able to call any other user, regardless of the network to which they are connected. Voice services cannot exist in isolation and there must be a means of ensuring seamless continuity as users move beyond coverage areas.

Second, operators offering VoLTE services must ensure that they can interconnect and support call delivery between their networks, which means that operators must be able to manage sessions across multiple domains, both VoIP and circuit switched, with no disruption as subscribers move from one to another. If operators cannot offer such continuity, they risk losing more customers to OTT voice providers delivering proven services over data connections.

Finally, users will remain on legacy networks for some time to come, which means that interworking is essential. While handover to legacy connections can be achieved by using a capability known as Single Radio Voice Call Continuity (SRVCC), operators need to be sure that it functions correctly and at scale. The resulting environment can be extremely complicated (see Figure 1).

The answer is comprehensive end-to-end testing to confirm interoperability between heterogeneous networks and the correct functioning of SRVCC. Operators need a solution that can perform testing of different scenarios, as well as load testing to evaluate performance under stress.

Meanwhile, operators are also undertaking a complex and lengthy migration towards NFV architecture and solutions. It is anticipated that this migration will yield many benefits, allowing operators to utilise a common IT infrastructure that supports all network functions. For example, this will dramatically reduce operators’ costs, while creating an infrastructure that offers greater flexibility and helps secure a more agile orientation. The question is, how to achieve such a migration while ensuring that current investment programmes, such as for VoLTE, are directed in the right direction and are tuned with NFV adoption plans?

Operators must carefully evaluate how and when to take the step towards NFV in parallel with existing service and network enhancements that are already underway. As many analysts have argued, NFV migration will not only be a long and complex journey, it will also be more of an evolutionary and iterative process than a wholesale revolution.  As a result, functions that are required to support the network, such as OSS and BSS platforms, as well as test and measurement solutions, are leading candidates for early migration towards NFV.

VoLTE is one such candidate, but the promised benefits of NFV also demand the ability to leverage common assets. Operators will want to implement complementary solutions that are compatible with their new investments. One consequence of this is that operators will need test solutions that enable them to validate and verify their VoLTE deployments in complex network environments and which are compatible with their NFV migration projects and which leverage the new, common IT infrastructure.

Fortunately, such solutions are already available and deliver significant benefits. The virtual model dramatically reduces delivery times of test solutions to operators and also allows remote installation to be facilitated for system extension. For example, if more capacity is required, it can simply be enabled by extending user licenses. Systems based on legacy or proprietary hardware cannot be extended in this manner.

A leading Scandinavian operator has already enjoyed the benefits of this approach to VoLTE deployment. It selected a solution that offered both comprehensive VoLTE and legacy capabilities and which was also suitable for deployment in a virtualised environment. This generated tangible results. Compared to operators of similar size and vendor complexity this operator saved at least six months by using this strategy.

The key is to choose virtualised solutions that combine VoLTE, LTE and legacy network technologies, such as 2G, 2,5G and 3G, into a single package. This significantly reduces troubleshooting effort and the time required to complete LTE and VoLTE deployments while ensuring full compatibility with NFV migration projects.

Alignment across strategic goals is essential. Advancing NFV programmes without paying attention to the needs of specific services and without maintaining a common approach to infrastructure evolution will be a costly mistake. Complementary tools to validate key services must be compatible with changing infrastructure. It’s hard enough getting services such as VoLTE right. Investing in technology that will soon be made obsolete by network virtualisation is a step in the wrong direction that will lead to serious cost and performance issues in the future.

Telecomlead.com