Bumps In the Road
Although some operators have been successful with 3G, there are still significant speed bumps that are slowing widespread adoption. Many parts of the world still use older wireless infrastructure. There remain vast numbers of people who have never used any kind of mobile phone, not to mention a smartphone with high-speed data services. India , for example, is often thought of as a cradle of high-tech, yet less than five percent of the population has cellular service, according to SG Cowen & Co. For these areas and these future wireless users, cost is perhaps the most important factor.
Other roadblocks on the path to 3G have included regulatory issues, clunky and expensive handsets, patchy networks and differing technical standards. Competing network standards pose a problem, with technologies such as CDMA2000 1X EV-DO, W-CDMA, and China 's TD-SCDMA that do not interoperate. Even with operator subsidies, 3G handsets are still too expensive. And a few studies have indicated that consumers don't understand what 3G is and what it can do for them—or, if they do understand, find the services and the devices overly complicated.
Why 3G Matters
For years we have been searching for that “killer app” that will make everyone want to run out and buy a 3G phone and subscribe to costlier high-data-rate services. In the end, it is turning out that the same service that brought customers to wireless to begin with—voice—will still be the most important 3G application, especially in countries where wireless service is not available or not widespread.
This brings up a rather obvious question: Why do we need expensive 3G infrastructure just to support plain old voice phone calls? Yes, 3G will bring advanced services that people have been talking about, but one of the real benefits is network capacity enhancement, to get more voice channels on the network in a more efficient way. 3G networks can provide about three times the capacity of 2G, at a cost per-minute that is about a quarter of 2G. By increasing capacity, operators can accommodate more customers. This leads to increased revenue, a larger customer base to sell services to and greater return on investment.
By increasing capacity at lower costs, operators can not only make voice service more affordable for those who don't have it, but they can also provide phone service at a price point attractive enough for customers to give up their existing landlines entirely. The billions of people in the world who do not yet have wireless service could probably be accommodated by existing 2G networks. However, 3G is a better and more efficient pipe. And of course, 3G provides the bandwidth for the higher-priced packet services.
Average revenue per user or per unit (ARPU) can benefit from 3G technologies which increase the efficiency of networks for both voice and data services. The improved voice efficiency frees more of the network for enhanced services, which in turn offers the opportunity for increased ARPU. While ARPU dropped in 2005 for wireless services overall, TeleGeography Research found that 3G services offered by companies such as Hutchison 3G had better ARPU than their regional averages in 2005—more than triple in Hutchison's Asia Pacific region, and close to 70 percent in western Europe. That increase can help carriers offset the costs of subsidizing new 3G phones for their customers.
3G technology continues to improve, moving toward 3.5G and 4G speeds. For example, High Speed Packet Access (HSPA), which encompasses HSDPA and High Speed Uplink Packet Access (HSUPA), promises to make W-CDMA networks faster and smarter, improving spectral efficiency and latency especially at the network edge. It is estimated that delivering a 10 MB file with HSDPA will only be 20 percent of the cost of delivery with W-CDMA. HSUPA is the next step to do the same for the W-CDMA uplink.
Saving Money Is the Key
Controlling price and cost are two key factors in how enthusiastically the market will embrace 3G. Low-cost operators are already leading European carriers down this road, selling voice services on their W-CDMA network at lower prices and pushing their competition to do the same. Design centers for 3G phones are moving to India to meet the low-cost race. Competing technologies such as WiMAX have improved to the point that they can act as serious competition to 3G cellular, pushing bandwidth prices down across the board—especially in North America, where there are the greatest number of broadband wireless access and WiMAX license holders, according to research firm Maravedis.
As a semiconductor guy, I can't speak for the carriers. But I can tell you that there are a few things that 3G handsets will need in order to compete and succeed: in a nutshell, they need to work better, last longer, and they need to cost less. Greater form factor flexibility and power management technology will allow designers to create handsets that meet these criteria. Reduced electronic bill of materials will help manufacturers save money and time. We're doing well with the “work better” part; some of the latest smartphones have the computing power of old desktop PCs. It's the “cost less” part that needs work. There is a market void in the area of low to mid-tier 3G handsets.
To drive costs down with increased functionality, the industry must reduce the materials and the development effort required to deliver both mid- and high-tier mobile devices. Semiconductor companies can try to “glue” more chips together and cut costs through process technology, but that simply won't do it.
And the Answer Is …
A single core modem (SCM) architecture removes numerous costly components, shares memory and cleanly separates development environments. This can help manufacturers save up to 50 percent in development time and achieve a 20 percent to 30 percent reduction in electronic bill of materials for handsets, depending on the tier that they are targeting. A substantially integrated hardware platform can reduce part counts and lower the total system costs.
Power management is one of the top technical issues that manufacturers and designers face, and power conservation can benefit most machines, from MP3 players to fuel-cell cars. As I mentioned, newer smartphones can have amazing computing power—but the technical challenge is to squeeze that power into something that will fit in your pocket, and not have to be plugged into a wall outlet all day. The shared memory approach and the built-in hardware accelerators of an SCM architecture can achieve up to 40 percent lower power consumption compared to traditional architectures, increasing battery life for power-hungry 3G devices.
The space savings of such an SCM design allows designers literally more room to innovate. New markets are opening up for all kinds of uses. One phone design will be right for a African farmer who needs to make frequent calls and calculate market prices for crops, but that may not be the same design that is right for an American senior citizen who values ease-of-use, or a child who needs a phone that is hard to break and is programmed to only send or receive certain numbers.
So, are we there yet? We can see the destination from here. While the industry continues to pave the supporting infrastructure, the vehicle must become smaller, less expensive and more efficient.
About the Author:
Dr. Franz Fink is senior VP and general manager for Freescale Semiconductor's wireless and mobile systems group.
Since joining Motorola's Semiconductor business in 1991, Franz held positions of increasing responsibility within the company in the areas of design and operations management. Prior to his current role, he was general manager of the 32-bit Embedded Controller Division in Motorola's Transportation and Standard Products Group. In 1999, Franz served as operations director of the Powertrain Systems Operation within the Advanced Vehicle Systems Division. In this position, he played an integral role in defining a product roadmap for PowerPC(tm)-one of the world's most commercially successful microcontroller architectures.
From 1994 to 1996, Franz was European design technology manager for Motorola Semiconductor business, based in Munich . From 1991 to 1994, he was the Application Specific Integrated Circuit (ASIC) design automation manager for Europe .
Fink earned a master's degree in computer science and electronics from the Technical University of Munich. Later, he graduated summa cum laude with a doctorate from the department of computer-aided design.
