Mobile Broadband: Evolution or Revolution?

The Merriam-Webster Dictionary defines evolution as, among other things, “a process in which the whole universe is a progression of interrelated phenomena.” On the other hand, revolution is defined as “a sudden, radical, or complete change.” With so many different parts simultaneously in motion, the global communications industry is certainly undergoing radical change; whether change will be “sudden” or a more Darwinian “progression” that preserves the inter-relations of networks is not yet clear.

The mobile broadband (MBB) space, comprised of wireless high-speed internet access from a PC or mobile Internet device (MID) using standards such as GPRS, 3G, WiMAX, LTE UMTS/HSPA, and EV-DO, is particularly dynamic and seems to tilt toward revolution.

In a presentation at the GSMA World Congress in February of 2009, market research firm McKinsey & Company reported that countries with MBB penetration greater than 10 percent achieve incremental GDP growth of 0.6 to 0.7 percent. Growth comes through direct channels such as communications and technology products and services, as well as indirectly through productivity improvements driven by the new technologies and infrastructure.

Keen to drive MBB penetration, MNOs have been upgrading their networks to accommodate the anticipated broadband market and cater to an explosion of consumer and business applications. Telstra, the Australian MNO, has seen revenue from data grow as a percentage of its total retail revenue from 23 percent in 2007 to 38 percent in 2009.

The GSM networks have been evolving their radio access network (RAN) technologies and corresponding speeds as well. The following table lists the GSM RAN access speeds and timelines:

• 2G
  GPRS – 114 kbps – 2000
• 3G
  Wideband CDMA (WCDMA) – 384 kbps – 2001
• 3.5G
  High speed downlink packet access (HSDPA-1) – 3.6 Mbps – 2005
  HSPDA-2 – 14.4 Mbps – 2006
  High speed packet access (HSPA+) – 21-28 Mbps – 2008
  HSPA+ - 42 Mbps – 2009
• 4G
  Long-term evolution (LTE) – 100 Mbps – 2010
  LTE – 1Gbps peak – 2013

MNOs and their customers are realizing the power and benefits of MBB, which gives customers ubiquitous high-speed access to the Internet on the move. Communities and regulators are also realizing that MBB is one of the fastest and most cost-effective ways of rolling out broadband networks, in contrast to fixed broadband roll-outs that depend on customer density, truck-rolls, copper wire penetration and digging for rights-of-way.

MNOs such as Telstra are realizing that by increasing access speed, the newer generation of technology increases data usage and drives revenue growth. Also, the newer generation technologies drive down the cost of transporting the data for the MNO. A cost comparison of different data technologies from Telstra shows that long-term evolution (LTE) is 65 times more cost-effective than the older GPRS technology:

• GPRS = 100% of GPRS
• EDGE = 50% of GPRS
• WCDMA = 10% of GPRS
• HSDPA = 5% of GPRS
• HSPA = 2% of GPRS
• LTE = 1% of GPRS

As of this writing, there are approximately 90 million HSPA subscribers worldwide with projections for 400 million subscribers by 2012. The technology driving the MBB usage by 2012 will be HSPA/HSPA+. This will be followed by greater penetration globally of LTE. Both the GSM and the CDMA class of networks have an evolution path to faster data (e.g. mobile broadband) and both paths lead to LTE as the final technology choice.

In February 2009, Verizon, the US CDMA operator, announced the commercial rollout of its LTE network, beginning in mid-2010. The intended target is a new generation of MIDs with data speeds of 8 – 12 Mbps. AT&T currently plans to run LTE trials in 2010 in select U.S. markets. LTE has the benefit of reduced latency of less than 10 milliseconds across the radio access network compared with 70 milliseconds for older radio access technologies. This reduced latency can support direct voice over IP (VoIP), enabling Skype-like services directly on the handset communicating IP end-to-end with wideband codecs. However, operators deploying LTE do not yet have plans to migrate their voice away from their circuit switched infrastructure.

My calculations suggest this migration will take place once voice service is generating less than 50 percent of the MNOs’ revenue and the benefits of running a single all-IP network become overwhelming. At the rate of current trends, I project this will take place around 2013 for the major MNOs in the U.S.

The migration of MNO networks to offer these high-speed broadband services will put increasing pressure on the existing resources and businesses of fixed and mobile operators and, importantly, on the availability of interconnect services — hubs — that enable cross-network communication and the flow of payments (à la the IPX) for any service that requires global interconnectivity.

Regardless of semantics, change is coming, it’s going to affect every aspect of the industry and it’s moving fast.

Ajay Joseph is CTO of iBasis, responsible for the technical strategy, innovation and engineering of the company’s global telecommunications network and supporting systems. Prior to joining iBasis in 1999, he worked in engineering and design for companies including GTE Internetworking, DeskNet Systems and NYNEX Science & Technology.