Where is LTE Voice Service?

Details

Published on Monday, 27 August 2012 06:51

Written by Craig Sutherland

Despite the massive growth in broadband mobile, the average revenue per user (ARPU) has been on the decline. Network operators are reluctant to invest in existing 2G/3G infrastructure, which will become obsolete over time. For operators this becomes a catch 22 situation.

The total worldwide revenue for the cellular industry is expected to reach US$1.1 trillion by 2012, according to analyst firm Wireless Intelligence. However, the rapidly increasing number of global mobile connections combined with intensifying competition in the industry has, over time, inevitably forced down ARPU. There are now more than 800 mobile operators worldwide, an increase of more than 40% over the last ten years. And, the race by operators to build market share in some developing countries (such as those in the Indian subcontinent or East Africa) caused call prices to drop as low as US $0.01 per minute in 2011. The fact these rates are generally only available for on-net calls has also led to an abundance of multiple SIM users in some markets, thereby driving ARPU down further. 

A major reason for the downward trend is that users are just not prepared to reach into their pockets and pay a premium for mobile broadband, especially when other alternatives such as free Wi-Fi exist. Wireless Intelligence says the commoditization of messaging services, mainly through ‘unlimited’ tariffs and bundled packages, has limited their potential as a revenue generator, leaving data and particularly the promotion of smartphones and mobile broadband services to form the core of most major operators’ revenue growth strategies.

The analyst firm is not wrong there. It seems every mobile operator is playing a waiting game. Waiting to see how future revenues will be realised from mobile services. Over the top (OTT) services such as Skype and Google Talk have eroded into the traditional minutes based business operators have enjoyed for years. In addition, although the next generation LTE networks represent a huge technological advancement in providing a complete IP evolved packet core, challenges around interoperability, roaming and legacy fallback still exist.  

The Driver For LTE

Recently I was involved in a discussion with an operations team from a mobile operator who had just launched an enhanced broadband LTE package boasting speeds of up to 100Mbps. I was curious to understand what was the use case for this package, given  the demand for the operator’s 42 Mbps offering was still limited to the business community. The firm agreed the likelihood of being able to sustain this traffic volume for a large number of subscribers would be impossible due to the interconnection bottleneck with the rest of the world. The pool of bandwidth available would be shared with all users of their LTE service. It seemed the promotion was to create a mindset with the public its network was faster than the competition’s similar offering.

I consider myself technology savvy and for my daily activities, 3G broadband is perfectly adequate in most places of the city where I live. I just can’t think of a use case where I would pay a premium to use faster mobile broadband. The compelling argument which would make me upgrade my mobile to LTE would be better quality voice.

LTE Standards

Mobile operators share my sentiments as well. However, they are deploying LTE networks in readiness for voice, which when handsets become available, will see a transition from 2G circuit switched connections to all IP voice. The challenges around this are how to handle roaming, fallback to 2G, SMS, intercarrier billing and establishing the globally accepted standard for voice over LTE. The GSMA is promoting its VoLTE initiative to handle roaming and interconnect standards. It is also advocating the 3GPP MMTEL initiatives to handle quality of service, global reach and emergency call standards. Other groups promoting different methods to handle legacy fallback have also been established, although have lost traction to the GSMA in this regard.

OTT LTE

OTT services like Skype and Google talk are already the first voice services using LTE, but these services although free, do not guarantee a uniform user experience in terms of call quality and reliability. In addition these services cannot be used for emergency calls. Already we are seeing operators deploy their own OTT services as an interim step.

Handsets

It seems clear that a new generation of handsets ranging from entry level to premium smartphones needs to be established. These handsets should by default support the G.722.2 wideband HD voice codec, since clearer audio will be the compelling argument for most people to upgrade their current generation of handset. The IP SIP stack used for voice calls should be built into the hardware, rather than operate as a software application. This way, even if the operating system (OS) is totally messed up, the phone should still be able to make and receive basic calls.

Cold start times are becoming unacceptable for most smartphones. As OSs become more complex, the startup times are increasing. This could be a critical factor in an emergency situation. Battery life still needs to be addressed, since it has become acceptable practice that smartphones must be charged daily. There should be a certification program where next generation phones are rated based upon these fundamental areas and given a score.

Infrastructure

We are in a limbo situation right now. LTE networks are ready but faster mobile broadband attracts only a small segment of the market. Hardware vendors want to design and deploy handsets that can be used on any network, but are reluctant to design devices only for early adopters before globally accepted standards are established. Network operators are reluctant to invest heavily in existing 2G/3G infrastructure, which will become obsolete over time. For operators this becomes a catch 22 situation. Oversubscribed 2G/3G networks are requiring infrastructure upgrades to sustain the current levels of customer demand and operators feel reluctant to do so, resulting in poorer levels of services for customers across the board.

For many operators the return-on-investment (ROI) for any core infrastructure upgrade has decreased, resulting in network planning teams realising they will only receive budget for a single technology upgrade. Many initial LTE deployments are deliberately over dimensioned for the size and projected growth of the network for this reason. Planning teams know procurement are likely to decline expansion requests in year 2 and 3, if the revenue stream for the initial phase is not satisfactory. This ‘big bang’ approach can make market projections on growth appear misleading.

Although, the all IP evolved packet core solves many of the congestion, authentication and capacity bottleneck issues, which troubled earlier generation networks, the all IP evolved packet core is not without its problems. LTE demands harmonized radio spectrum to operate at its optimum performance levels. In competitive markets obtaining sufficient licences can be problematic. This has meant broadband LTE handsets are appearing only in selected markets. Refarming of GSM spectrum can be complicated. Operators need to evolve with the technology ecosystem. The multiplication of bands (700 MHz, 850 MHz, 900 MHz, 1800 MHz, 1900 MHz, 1700-2100 MHz, 2500 MHz), with different versions in different regions and technologies within bands (EDGE, UMTS, HSPA, HSPA+, LTE) complicates the technology choice.

Conclusion

Getting the right mix of investment in both existing 2G/3G technology and next generation LTE networks is a complicated balancing act, especially when trying to maintain network availability levels and expand for the future at the same time. Until the underlying standards of how planned LTE voice services operate and fully operational handsets appear, LTE adoption will remain a unique niche for a small percentage of the business community that demands faster broadband connectivity.