Last week saw the launch of the first commercial Voice-over-LTE (VoLTE) services, but most operators are likely to take a cautious approach as they face technical and business challenges.
U.S. operator MetroPCS as well as SK Telecom and LG Uplus in South Korea will be the first to offer VoLTE services.
"It is the crystallization of all of what we have been working on in the last two years," said Dan Warren, director of technology at industry organization GSM Association, which has lead the work to implement VoLTE.
Eric Ericsson, head of the Mobile Telephony Evolution Program at Ericsson Networks, agrees: "It is proof that VoLTE works. I find it difficult to believe that someone would make a commercial announcement without being fairly sure that it works from a technical standpoint," he said.
However, take-up of telephony will be slow compared to the roll-out of commercial LTE networks for mobile broadband, which is expected to reach 150 networks by the end of the year, according to industry organization Global Mobile Suppliers Association (GSA).
Expectations for VoLTE are much lower, according to market researcher Dell'Oro Group.
"There will be up to ten commercial launches next year around various parts of the world. The success or failure at MetroPCS and shortly thereafter Verizon Wireless and what happens in Korea will determine whether the other projects go ahead quickly or slowly," said Chris DePuy, analyst at Dell'Oro.
For operators, VoLTE has to make sense both technically and financially, and the latter may be a struggle in the short term, according to DePuy.
"The problem with VoLTE is that operators aren't able to charge more than for current voice services, and therefore carriers are spending money but not getting more revenue; possibly even getting less revenue because voice usage is going down, and it's very competitive," said DePuy.
Swedish TeliaSonera was the first to launch a commercial LTE network for mobile broadband in 2009. But when it comes to moving voice to 4G the operator won't be an early adopter, as it sees few advantages over existing systems, according to Tommy Ljunggren, vice president of system development at TeliaSonera.
"The drivers for us in Europe for implementing VoLTE for economic reasons are not strong. In the long term, and that is in the very long term, it will be easier for us to handle just IP. But producing VoLTE isn't automatically cheaper than circuit-switched voice," said Ljunggren.
TeliaSonera will instead continue to rely on CS (Circuit-Switched) Fallback, a mechanism that allows smartphones to access the Internet using LTE and then switch to GSM or 3G when there is an incoming call. So the radios for GSM and 3G as well as LTE don't have to be turned on at the same time, which increases battery life.
Warren, on the other hand, isn't a big proponent of CS Fallback.
"I am hoping [the first launches] will give a strong indication that within the next 12 to 18 months operators can take the step directly into VoLTE and not encumber themselves with a CS Fallback launch," said Warren.
Proponents say VoLTE will result in better voice quality thanks to lower latency and HD Voice becoming a defacto standard. There is nothing in the VoLTE specification that makes implementing HD Voice mandatory, but most operators are still expected to do that, according to Warren.
The improved quality HD Voice offers is possible thanks to AMR-WB (Adaptive Multi-Rate - Wideband), a speech-compression algorithm that doubles the range of voice frequencies transmitted.
To take full advantage of the improvement HD Voice can offer, users also need phones and headsets with good sound quality.
"More and more people are using headsets, and you can really tell that there is a difference in quality. There are lots that are really bad," said Ljunggren.
However, HD Voice isn't a VoLTE exclusive, and has already been implemented in over 40 3G networks, according to GSA. Lower connection times should give VoLTE an edge over CS Fallback, but the technology also introduces new potential problems, according to Ljunggren.
That's because the implementation of VoLTE also means telephony traffic is moved from a circuit-switched world to one where everything is IP-based.
"Previous mobile networks have been optimized for telephony and SMS, while LTE has been optimized for offering broadband connections that are as fast as possible," said Ericsson.
The challenge is to ensure the quality of the voice service on that data-centric network, including at the edge of a cell or network where the available bandwidth is limited, according to Ericsson. The key to doing that is prioritizing voice over data, he said.
"From the LTE device, telephony is sent using one radio bearer or pipe and data is sent using another pipe, and if a conflict arises, the telephony gets priority," said Ericsson.
Getting the quality right will be important to operators. The disaster scenario is that the quality and coverage is not as good as the old telephony systems, according to DePuy, who nonetheless is still optimistic about the outlook for the pioneering operators.
"I actually do think they will be successful. Technically it will work," said DePuy.
The use of traffic prioritization opens up discussions on network neutrality, but Ljunggren isn't too worried.
"As long as we offer other services the same possibilities to use QoS and priority, I don't see a problem with using it for VoLTE," he said.
VoLTE is powered by the IP Multimedia Subsystem framework, the implementation of which also allows operators to roll out over-the-top services such as instant messaging based in turn on a specification called Rich Communication Suite (RCS).
Work on the suite started in 2008, and the goal was to turn IMS into standardized services for operators. IMS had been around for a long time, but due to the framework's complexity it hadn't taken off, a situation which RCS was meant to change.
Even though it is off to a slow start, there is a lot of hope in the telecom industry that RCS will help operators develop more attractive offerings and compete with Web-based services, while at the same time potentially delaying the arrival of VoLTE.
"In the past month or so, bundling voice and messaging has become a priority for operators, and that is having somewhat of a delaying effect," said DePuy.
The combination of VoLTE and RCS will allow users to communicate in new ways, according to Warren. The vision is that they will be able to see who is available, and chat, share files across any device, on any network, with anyone in their address book.
"What I think will be really interesting is the step beyond that when operators expose those capabilities to application developers," said Warren.
But even if VoLTE works well, it won't take off until there are a number of great smartphones for users to choose between, including a future version of Apple's iPhone.
"Handsets will be shipping in small volumes initially, and that is not good for keeping prices low," said DePuy.
TeliaSonera's motivation to implement VoLTE isn't helped by the fact that Europe is behind the U.S. when it comes to rolling out 4G. That has resulted in fewer smartphones and tablets being adapted for the European spectrum bands.
The launches in South Korea and the U.S. are backed by LG Electronics and Samsung Electronics, with smartphones like the Galaxy S III available. It is good news that Samsung is onboard, according to Warren. But when it comes to Apple he isn't as hopeful.
"When it comes to this kind of technological evolution, Apple isn't on the bleeding edge. Instead it is relatively cautious," said Warren.
Together Samsung and Apple own the smartphone market with a 50 percent market share, according to IDC.
Still, the arrival of the first VoLTE services is a really positive step for 4G as a whole, and shows that LTE can become the preeminent technology for all services, according to Warren.
Send news tips and comments to firstname.lastname@example.org