Archived - Wireless Technology Roadmap — Market Realities at the Outset of the Later Growth Period

This chapter outlines the broad market forces that will play a critical role in shaping the ongoing later growth period (2004–2024) of the global wireless industry.

• Regulation (Political)
• Technology
• Economic
• Social
• Environmental

---

Regulation (Political)

This is a major issue in meeting the growing demand for wireless services: under the current system, spectrum shortages are expected within about a decade.

The central problem is that the regulatory framework is largely a legacy, ²⁸ built around the technology of the first global wireless industry, namely: radio broadcasting. During that era, the only way of avoiding interference was to grant exclusive use of specific frequency bands. As a consequence, radio spectrum has been treated as a scarce resource that "must be allocated by governments or bought and sold like property." ²⁹

Figure 8: Spectrum Supply and Demand

Description Link  ²⁷

Since the 1930s, governments have decided on the best use of the airwaves. This central planning approach has allocated exclusive free licenses "in the public interest": from radio and television to local government, the military and educational broadcasters. A market-based approach to spectrum use is still seen as novel. Although governments in both North America and Europe began to auction frequency bands in 1995, central planning still accounts for 98 percent of spectrum use: even now, auctions represent only 2 percent of "prime spectrum."

Prime spectrum, the 1 percent of all frequencies below 3 GHz, is worth more than the other 99 percent of radio spectrum between 3 GHz and 300 GHz. The lower frequencies are prized because they are much better at penetrating rain, trees and buildings. However, the legacy of almost a century of regulation has been to vest huge swaths of frequency to underutilized purposes; new uses of desperately-needed prime spectrum are systematically starved as a result.

For example, TV broadcasters control 15 percent of prime spectrum but serve only 11 million U.S. homes (cable and satellite serve the other 88 percent — using no prime spectrum). Cellular carriers get only half as much spectrum to serve 137 million customers. Wi-Fi serves 20 million customers with only half the spectrum allocated (but rarely used) to distribute video to schools.

The resulting opportunity cost of services and technologies not offered because of legacy allocation of spectrum is estimated at $771 billion in the U.S. (2001). ³²

In summary, the problem of radio interference that long ago led to the current regulatory regime is in fact a technological issue. Moreover, developments in radio technology are obviating the need for central planning.

Technology

There are five broad developments in wireless technology that effectively increase the ability of users to share spectrum:

• Spread Spectrum: This technique distributes a signal across wide bands of frequencies. This is a departure from the old technology of radio broadcasting which concentrates the signal on a single frequency. Spread spectrum's advantage is that it can operate in frequency bands that are already assigned to other uses. One way is for the signal to rapidly "hop" between unused frequencies in its assigned band — constantly adjusting to noise and interference to make best use of available frequencies. Another way is to spread a low-power signal across a wide band of frequencies, "underlying" existing uses. Ultra wideband (UWB) applies this approach across a huge range of frequencies. "UWB is an interesting technology, but caution needs to be exercised because it contributes to the RF noise floor, and in so doing can make a previously workable solution for other users of the spectrum unworkable." ³³
• Software-Defined Radio: Most radios are "hard-wired" to transmit and receive one type of signal at one range of frequencies. However, a sufficiently powerful microprocessor can use software to reconfigure a chip's circuitry — within limits ³⁴ — to work with many types of signals and frequency ranges.
• Cognitive Radio: Powerful software programs give wireless systems the capability to sense and adapt to the EM environment, per example, changing transmission characteristics to make best use of available radio spectrum.
• Smart Antennas: These systems use multiple antennas both to aim signals in a particular direction and to pick out a specific signal from background noise. Smart antennas can retrieve individual signals that share the same frequencies.
• Mesh Networks: This is a recent development that was deliberately designed ³⁵ as a disruptive technology to replace commercial networks with a free, user-based user-owned service. These are random, unplanned networks, self-formed through the co-operation of individual transceivers. Each transceiver in the network acts as a repeater to transmit data from nearby neighbours to distant peers, allowing signals to "hop" across the network. As with spread spectrum, the advantage is that signals — with only hundreds of metres instead of kilometres to travel — can be sent at low power, allowing reuse of the same spectrum again and again.

Like all technological systems, the performance of wireless depends on developments in supporting devices, components and materials. While a detailed analysis of these areas is beyond the scope of this document, key issues are listed below:

• Batteries: These are the Achilles' Heel of mobile, handheld devices. They continue to pose a significant constraint on the use of today's features like colour screens. Tomorrow's handsets will include even more power-hungry features like video.

  Batteries will also be a factor in the market penetration of newer wireless technologies like WiMax ³⁶ that offer faster network speeds at the expense of higher power consumption.

  In summary: "growing processing capabilities and application demands are in conflict with the slow development of battery technology." ³⁷

• Chipsets: Advances in semiconductor technology have been instrumental in making handheld wireless devices possible. They are a critical plank in the platform technology on which electronic systems and their wireless applications rest. Shrinking transistor size has been the key to cost and performance improvements in chipsets. Intel, the world's largest semiconductor manufacturer, has been a leader in advancing chip-making technology. For example, at the beginning of 2006, most of Intel's production ³⁸ used 90-nanometer (nm) ³⁹ technology. By year's end, the bulk of Intel's chips were made using a more advanced 65-nm process. Later in 2007, Intel expects to produce the first devices based on its new 45-nm technology. ⁴⁰

  While chip fabrication technology will run into fundamental barriers ⁴¹ imposed by quantum physics as feature sizes decrease, this is not expected to create any significant bottlenecks for the development of wireless applications over the 2006–2016 roadmap timeframe.

• Photonics: It addresses transmission capacity, ⁴² a basic limitation of electronics systems. As chips shrink, the delay in electronic signal transmission from chip to chip and chip to board becomes the performance bottleneck in systems. This is because the huge number of transistors on a chip results in extensive interconnect ⁴³ length. Increasing integration density will result in 20km/cm2 of interconnect around 2010. ⁴⁴

  Consequently, photonics connections have increasingly replaced electronic ones where high bandwidth (bits/second) is required. In fact, electronic approaches to deliver 10 Gbps ⁴⁵ at motherboard level have proven (so far) problematic. ⁴⁶

• Software: For electronics, software is more and more instrumental in delivering the increased functionality that users expect. This is true from device level (per example, chips) to systems level (per example, cell phones).

  Initially, chips were very low-level building blocks of electronic systems: a data sheet was sufficient to support their application in systems. Today, advances in fabrication technology have made it possible to shrink entire systems onto a single chip. This is referred to as system-on-chip (SoC) technology. The result is that where there was once a clear separation between chip and systems design, these two activities are converging.

  For example, it used to be that chip makers supplied the lower-level software and third parties developed the operating systems. Once the wireless system was provided to users, they developed the specialized applications software. However, systems integration presents a whole new challenge: now, software all the way from the chip level to systems to subsequent applications has to work together seamlessly. The result is hardware/software co-design.

Economic

Economics will place significant limitations on the commercialization of wireless technology. The major factors are outlined below:

• Globalization: This has a major impact on industry structure and on Canada's ability to develop the skills needed to compete internationally.

  The core structural problem is that Canada's ICT sector is fragmented. SMEs dominate: fully 98 percent of the 32 000 ICT firms employ fewer than 100 people. The large, vertically-integrated companies that used to be an effective industry training ground for newly-graduated scientific and engineering talent are gone. They exposed young graduates to the whole range of doing business: taking products from ideas through to market; from R&D to engineering and production prototypes then on into manufacturing and distribution.

  In the new global economy, functions like R&D, manufacturing and marketing are each located separately. Head office integrates the results across many countries. Because of Canada's favourable environment for performing R&D, a significant portion of the research is often done here. However, SMEs have to cover the entire range of business activities. Their source for the needed experience (about five years) was the large vertically-integrated companies. Unfortunately, these are gone, bought by global giants and redirected to more specialized missions.

  Even billion-dollar Canadian companies have been bought out. A short list of ICT examples includes Newbridge Networks (by Alcatel), ATI (by Advanced Micro Devices), C-Mac (by Solectron), and Creo (by Kodak).

• Sunk Assets and Switching Costs: The established wireless industry is a US$500 billion (c.2005) global giant, dominated by large operators with a huge asset base of existing networks.

  Although third-generation (3G) networks are the current industry standard, there are already concerns that 3G networks may fall short of the demands of important new applications:

  "3G may not be able to meet the performance required for future multimedia, full motion video and wireless teleconferencing." ⁴⁷ Others put it more bluntly; "3G isn't good enough to meet wireless broadband demands." ⁴⁸

  Yet, at the outset of the industry's later growth stage — a time of slowing single-digit market expansion — wireless network operators are caught in the vise of high fixed costs and declining marginal revenues (see Chapter 3: The Wireless Industry). There is a need to invest in technologies beyond 3G, but the switchover costs are prohibitive. As illustrated in Figure 9, the cost of the new equipment may be only one-third the cost of the old equipment, but the cost of switch-over may exceed the cost of the existing equipment.

  Moreover, as major public companies, the adverse effects of asset write-downs compel wireless operators — the key customers for large infrastructure investments — to manage with their existing asset base.

Figure 9: Switching Costs

Description Link  ⁴⁹

Finally, Canada's Competition Bureau found that "there are very high barriers to facilities-based entry, including high capital costs to construct and run a network, regulatory requirements and foreign ownership restrictions." ⁵⁰

• Intellectual Property (IP) Costs: The fifty-year evolution of the semiconductor industry from vertically-integrated giants to layers of horizontal specialists has seen the rise of IP as a stand-alone business. For example, "fabless" companies design and market microchips. However, they outsource the complete manufacturing process to merchant wafer foundries, typically Southeast Asian companies that specialize in chip production. Canadian examples of fabless companies include Zarlink and Tundra Semiconductor.

• With the growing complexity of chips, the design business has produced companies that further specialize in particular parts of the chip circuitry. These designs are sold as IP to others that incorporate them into their chips. Canadian examples of these "chipless" companies include MOSAID Technologies in memory and Elliptic Semiconductor in security and verification.

• A related development has been the rise of "patent trolls" — companies that threaten legal action based on alleged infringement of their designs. With costs and the additional uncertainty that the courts can properly resolve complex technical issues, costly out-of-court settlements are often the result. The dysfunctional result is to increase the cost of innovation. In fact, "Patent trolls and related IPR issues are becoming an increasingly serious and costly problem, and have the potential to entirely stifle innovation."  ⁵¹

• A recent Canadian example ⁵² is RIM's $612.5-million payment to patent-holding company NTP to settle a long-running dispute that had threatened to shut down RIM's email service for its three million Blackberry users.

Social

Wireless access is now taken for granted. Users expect a minimum level of service; a normal development in the technology lifecycle when new products begin to be taken for granted, fading into the background as an everyday convenience.

Privacy concerns over the use of wireless are increasing as the technology makes it easy to intrude to the point of tracking the location of individuals. Security is a major issue as well, in particular as e-commerce grows.

Environmental

The disposal of obsolete computers and their accessories is becoming a major issue throughout the industrialized world. Many states in the U.S. (per example, California, Massachusetts, and Minnesota) have outlawed their disposal in landfill sites and incineration operators are apprehensive about placing certain items (particularly batteries) in their plants. As the waste products become smaller and smaller, many small businesses are discarding them in their household garbage because its disposal requirements are generally less stringent. It is estimated that 500 million personal computers were taken out of service between 2000 and 2007.

Waste management companies in the more regulated areas are offering specialized disposal services at costs that are in addition to their normal disposal costs. In less regulated areas, such equipment is finding its way into the hands of junk dealers who dismantle it, making it more difficult for authorities to track its disposal. In keeping with other environmental legislation, the manufacturers of such equipment will face more and more legislation that facilitates the safe disposal of their products. This could result in the application of new technologies (per example, sensors) at the manufacturing stage in order to monitor compliance.

---

²⁷  Source: IST-2003-507581 Winner D6.5 v1.0 Spectrum Requirements for "Further Developments of IMT-2000 and Systems Beyond IMT-2000" (Return to text)

²⁸  U.S. regulation dates from the Radio Act of 1912 and the Communications Act of 1934. However, Canada has closely followed American practice. (Return to text)

²⁹  "On the same Wavelength," The Economist, August 12, 2004. (Return to text)

³⁰  "Dead Air," Forbes Magazine, Nov. 25, 2002. (Return to text)

³¹  Ibid.. (Return to text)

³²  "On the same Wavelength," The Economist, August 12, 2004. (Return to text)

³³  John Visser, P.Eng. International Wireless Standards, Nortel (May 18, 2007). (Return to text)

³⁴  "For example, a radio designed for, say 3.7 GHz, cannot be easily adapted to 3.5 GHz because it strays too far from the optimal performance of its RF components governed by the laws of physics" John Visser, P.Eng. International Wireless Standards, Nortel (May 18, 2007). (Return to text)

³⁵  1995, the massive array cellular system (MACS), a Canadian patent. (Return to text)

³⁶  "WhyMax?" The Economist, Feb. 24, 2007. (Return to text)

³⁷  "The Future of Wireless" Mark Pecen, VP Advanced Technology, RIM (January 11, 2007). (Return to text)

³⁸  "Intel to produce smaller and less power-consumptive chips" (March 16, 2007). (Return to text)

³⁹  One nanometer is one-billionth of a meter. (Return to text)

⁴⁰  "Intel UltraMobile PC chip nears release" (March 21, 2007). (Return to text)

⁴¹  For example "The Red Brick Wall: Computing faces the end of a road," Pile Systems Inc. (2004). (Return to text)

⁴²  Transmission capacity is measured by the product of bit rate (B) x distance (L). B is the transmission speed (bits/sec.) and L is the distance signals can travel without regeneration. (Return to text)

⁴³  Interconnect is the "wiring" between transistors. (Return to text)

⁴⁴  "Will Silicon be the Photonic Material of the Third Millennium?" Journal of Physics: Condensed Matter 15(2003) R1169-R1196. (Return to text)

⁴⁵  Gigabits per second, a data transfer speed measurement (When used to describe data transfer rates, a gigabit equals 1 000 000 000 bits per second). (Return to text)

⁴⁶  "Silicon Photonics Poised to Invade Local Area Networks," Photonics Spectra, March 2006. (Return to text)

⁴⁷  "3G — Beyond 2.5G and 3G Wireless Networks" (Oct. 2, 2006). (Return to text)

⁴⁸  "Nortel CEO: 3G can't cut it" http://www.lightreading.com/document.asp?doc_id=117486 (Feb. 15, 2007). (Return to text)

⁴⁹  "The Future of Wireless" Mark Pecen, VP Advanced Technology, RIM (January 11, 2007). (Return to text)

⁵⁰  "Competition Bureau's Clearance of Rogers-Microcell Wireless Merger Explained," The Competitor, Stikeman Elliott (June 2005). (Return to text)

⁵¹  John Visser, P.Eng. International Wireless Standards, Nortel (May 18, 2007). (Return to text)

⁵²  "Blackberry Maker, NTP Ink $612 Million Settlement" (March 3, 2006). (Return to text)

---

Information on Downloading a PDF Reader
To access the Portable Document Format (PDF) version you must have a PDF reader installed. If you do not already have such a reader, there are numerous PDF readers available for free download or for purchase on the Internet:

• Adobe Reader
• Foxit Reader
• Xpdf
• eXPert PDF Reader