Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : The New Qualcomm - a S&P500 company -- Ignore unavailable to you. Want to Upgrade?

To: Eric L who wrote (6265)	2/5/2000 11:22:00 AM
From: Ruffian	Respond to of 13582

2/05/00 - SiGe upgrades wireless, soft radio Feb. 04, 2000 (Electronic Engineering Times - CMP via COMTEX) -- In the past year, several companies, including Atmel, Conexant, Infineon, Lucent, Motorola and STMicroelectronics, have announced that they intend to join IBM in offering highly integrated silicon-germanium (SiGe) BiCMOS technologies. A similar number of SiGe discrete (HBT-only) technology offerings are on their way as well. This transition to a silicon-based heterojunction technology, focused on the communications segment of the market, bears further examination. In late 1999, silicon-based transistor technology began the first of several fundamental shifts in its traditional road map. Until recently, scaling had been the focus of improved transistor performance, both in bipolar and FET devices. As transistor dimensions scaled downward, so too did parasitic capacitances, carrier transit times and chip dimensions for given functional blocks. This translated into a Moore's Law progression toward higher levels of circuit performance and lower overall technology costs. There are limits to such device scaling for silicon-based FETs, however, so alternatives must be developed for further improvement. The oldest class of silicon-based transistors, bipolar devices, had some time ago reached the point where further scaling became an exercise in diminishing returns. IBM developed silicon-germanium-based HBT technology in the mid-1980s to circumvent the limits of bipolar transistor scaling. SiGe HBT relies on "bandgap engineering" for its extraordinary performance, where low (10 percent to 20 percent) Ge content in the transistor base is used to enhance the speed, gain and noise properties of such devices. Papers at the 1999 International Electron Devices Meeting dealing with the upcoming generation of 0.18-micron SiGe HBTs demonstrated fT and fMAX values bracketing 100 GHz for 1-V operation, with the Nyquist frequency less than or equal to 0.4 dB at 2 GHz. Perhaps of greater importance was a report of high-yielding 0.25-micron SiGe BiCMOS production, with achievable integration levels of tens of thousands of SiGe HBTs and millions of FETs. Yet even these indicators of superb performance and integratability for both HBTs and CMOS do not explain the recent flood of SiGe technology announcements. That trend is driven by other factors. IBM took SiGe technology into volume manufacturing in 1996 with early product application focused on demanding circuits that relied on both performance and integration. In 1998 Alcatel (Paris) announced the commercial availability of its newest OC-192 (10-Gbit/second) Sonet system, an optical-fiber transmission system based upon the earliest generation of SiGe technology. Exploiting SiGe's integration capabilities, the OC-192 chip set self-optimized optical-system performance and produced extraordinarily low bit error rates: 1 bit error in 6 billion years. Reporting yields that exceeded 90 percent for the chip set, Alcatel had created a crucial precedent for SiGe's commercial viability. Alcatel also reported success in two OC-768 (40-Gbit/s) field trials in Germany and Portugal, where the company sustained long-haul 40-Gbit/s transmission over existing single-mode fiber using several SiGe test chips built alongside the 10-Gbit/s production chip sets. It's important to note this was accomplished using SiGe technology derived from IBM's long-established 0.5-micron CMOS fabricator, not a unique custom fabricator operating at the technological and financial bleeding edge. That is, the investments, learning and road map of silicon technology had been successfully used to operate at a performance level never before associated with silicon-based devices. Similarly, by embracing conventional silicon as the process technology, the economy of scale of the base CMOS process is shared, with greater than 1,000 wafers per day throughput in the shared 200-mm wafer facility. These financial considerations have proven to be at least as important to the industry's adoption of SiGe technology as has SiGe's performance metrics. At another level of integration, components for OC-48 and OC-192 systems are marketed in SiGe technology by Applied Micro Circuits Corp. (San Diego). In the wireless arena, a highly integrated entrant into SiGe-based electronics debuted in early 1999. The Prism II chip set from Harris Semiconductor (Melbourne, Fla.) conforms to the IEEE-802.11 wireless LAN standard and supports data rates of 11 Mbits/s in the 2.4-GHz ISM band. Compared with its predecessor, the chip set reduced power and size by 50 percent and cut material costs. Despite the reductions in cost and power, this chip set operates at a 550 percent higher data rate than the original 2-Mbit/s chip set it replaced. Cellular telephony has also embraced SiGe technology, as evidenced by Qualcomm Inc.'s (San Diego) February 1999 announcement of the RFR3100, an RF-to-IF receive processor deployed at the front end of its CDMA handsets. Similarly, RF Micro Devices Inc. (Greenboro, N.C.) has announced it intends to adopt SiGe-based device technology for a spectrum of its offerings in the cellular market. These developments indicate that any new technology in the cost-sensitive handset industry must drive down material costs if it is to become pervasive. One challenge to reducing costs is the need to address passive devices as well as active ones. With wired and wireless applications migrating to ever higher frequencies, it becomes increasingly difficult to ignore interconnect losses and passive element degradation that rise from the conductivity of silicon substrates. Integrated inductors on silicon historically have had poor quality factors, Q's of less than 5 being typical at frequencies of interest. With the advent of analog-optimized interconnect technologies for silicon, values exceeding 20 are achievable, and they enable the gathering of additional functions, such as VCOs, onto monolithic dice. The elimination of off-chip high-frequency interconnects and passive elements does help cut costs, but much work remains to integrate more challenging functions such as high-quality filters. In September 1999, Leica and IBM announced and began distribution of a global positioning system (GPS) employing the first digital radio architecture found in the consumer market. The GPS radio, a single die only a few millimeters square, consists of an amplifier followed by a high-performance analog-to-digital converter. Taking the signal from the antenna directly to bits at the carrier frequency, without mixing or downconversion, allows extraordinary miniaturization. An ASIC matched to the radio completes the receiver and converts the high-speed bit stream to positioning data. This novel system architecture produced dramatic savings in both system size and cost. With the advent of a highly integrated mixed-signal technology that supports fT and fMAX values well above 100 GHz, interesting possibilities abound. The GPS unit noted above employs a true digital architecture, taking RF directly to bits. Going forward, it is logical to contemplate dealing with more-complex voice and data protocols, such as GSM or CDMA, in the same way. Assuming digitization at an adequate rate and precision to preserve the information integrity of the arriving signal, it's possible to implement what amounts to a "universal radio" or "soft radio." The combination of SiGe technology's capabilities for high performance, low power and low noise have enabled the first step in the direction of soft radio, and more will be possible as performance increases with each generation. It is reasonable to forecast that the first soft-radio implementation will occur in the basestation arena, where power consumption in high-performance data converters is less of an issue, while handset or other mobile implementations will lag well behind. -0- By: Bernard Meyerson, Director, Telecommunication Technologies, IBM T.J. Watson Rese Copyright 2000 CMP Media Inc.

To: Eric L who wrote (6265)	2/5/2000 12:47:00 PM
From: uel_Dave	Read Replies (2) \| Respond to of 13582

Eric, Another article on CDN Wireless Companies from the G&M of Feb. 5, 2000 Wireless data companies ride tidal wave of investor interest BUYING FRENZY Share prices of a handful of Canadian companies are skyrocketing in the sector, which is creating an entirely new market for equipment, software and services. TYLER HAMILTON Technology Reporter Saturday, February 5, 2000 More than a dozen Canadian companies, small and large, have been riding a tidal wave of investor interest in the wireless data market -- also known as "Internet Wave II." It's a frenzy that has given Research In Motion Ltd. a market value greater than that of Newbridge Networks Corp., suggesting investors see more potential in wireless data devices than the traditional networking equipment business. It's also the same investor enthusiasm that has more than tripled the share values of companies such as Sierra Wireless Inc., Wi-LAN Inc., Infowave Wireless Inc. and eDispatch.com Wireless Data Inc. over the past few months. The reason for this excitement is simple: Wireless is hot. The Internet is sexy. Bring them together and you expand the mobility and reach of the dot-com economy, creating an entirely new market for equipment, software and services. "We're very, very bullish on this sector," said Barry Richard, a technology analyst from Sprott Securities Ltd. in Toronto. "All of these stocks are just running wild. Canadian companies are doing great in wireless data, because it's an area where they have tons of strength." Sierra Wireless, for example, has seen its stock jump 800 per cent in less than three months, from a low of $16.15 in November to its close yesterday of $128 on the Toronto Stock Exchange. The Richmond, B.C.-based company, which was added to the TSE 300 last week, makes high-speed wireless modems that can be used in laptop computers and mobile devices. In three months, Sierra has signed supply contracts ranging between $24-million and $30-million (U.S.) with three major U.S. companies, including AT&T Corp. and Sprint Corp. "We see Sierra Wireless as one of the leaders in the industry," said Mark Pavan, an analyst with Yorkton Securities Inc. in Toronto. The same rise to glory holds true for Research In Motion, the Waterloo, Ont.-based maker of interactive paging devices. The company's Blackberry product, which lets users send and receive e-mail while on the go, has been endorsed by the who's who of the high-tech industry. Michael Dell swears by it. Intel wants all of its employees to use it. And Nortel Networks Corp. wants to co-develop future networks around it. The stock, as a result, has skyrocketed 300 per cent since October and the company now has a market value of more than $9-billion (Canadian). "The reality is that the number of cellular phones and pagers dwarf the number of personal computers," Mr. Pavan explained. "As the Internet becomes more utilitarian, and is used less for entertainment and more for practical applications, the environment will migrate to wireless." In recent months, wireless data software and equipment developers have been signing deals and striking alliances with service providers. On the consumer side, portal players such as Yahoo Inc. have embraced the mobile environment while financial institutions have taken the lead by rolling out wireless trading and banking services. Bank of Montreal, for example, introduced wireless stock trading last week, using software developed by Toronto-based 724 Solutions Inc., whose market value has jumped five-fold since its initial public offering on the TSE Jan. 29. In the past week alone, a number of events and announcements have bolstered the wireless data agenda, including: Plans by Canadian stalwart Nortel Networks Inc. and Phone.com Inc., a U.S. provider of software and services for the wireless delivery of Internet content, to co-develop wireless data technologies; The formation in Cannes, France, this week of the Mobile Wireless Internet Forum, consisting of industry leaders including Nortel and Cisco Systems Inc. The group will co-operate to develop standards for wireless Internet applications and equipment. The creation of a $300-million (U.S.) wireless Internet fund, set up by Argo Global Capital to help finance new wireless data ventures. The list goes on, supporting research that indicates the wireless data sector is about to explode. In the United States alone, San Jose, Calif.-based research firm Dataquest predicts that more than 36 million people will subscribe to a wireless data service by 2003, making the market worth about $3-billion. That's why companies such as Calgary-based Wi-LAN have emerged from relative obscurity to become a leading contender for the future development of mobile and fixed-wireless networks. Many industry observers say Wi-LAN's wideband orthogonal frequency division multiplexing (W-OFDM) technology could become a standard for future wireless networks, and the company is currently building support for that vision through an alliance of major industry players. It's tweaked the interest of high-tech giants such as Cisco. And like many of its Canadian peers, Wi-LAN's stock has more than quadrupled in four months, rising $1.15 (Canadian) yesterday to $45.85 on the TSE. Mark Zohar, wireless data analyst for Cambridge, Mass.-based Forrester Research, said there's no doubt wireless Internet will be the market to watch this year. But he said there's a danger in confusing the capabilities of wireless Internet technologies with what people will actually use. "Do customers really want to have applications to buy lipstick on a mobile phone?" Mr. Zohar said. "What is the revenue model? What are the winning devices? There's huge uncertainty in this market, and as a result everyone is jumping in." He said there's too much focus right now on trying to deliver to mobile phones the same content that a person could get through a personal computer. The mistake, he added, is to believe that just because people shop on computers means they'll want to shop while sitting in an airport or walking to a meeting. "The information you receive has to be extremely timely and time-sensitive. It has to be very simple to execute or transact, meaning 'one click' is absolutely critical. It also has to be easy to personalize," said Mr. Zohar, explaining that wireless location-detection technologies will become increasingly important. That puts Calgary-based Cell-Loc Inc. in a sweet position. The maker of wireless device tracking technologies has seen its stock jump from $5 in September to above $20 on the Canadian Venture Exchange. Mr. Pavan said investors are betting on the sector rather than individual companies, essentially because nobody knows for sure who the winners are going to be. "That's why you're seeing a huge runup in these stocks." It's also why a growing number of lesser-known wireless data players -- including eDispatch.com, Infowave Wireless and Unique Broadband Systems Inc. -- have done well by the market. It's also why a new undercurrent of private firms, such as Toronto-based Wydsom Inc., are poised to make waves in the public markets. WIRELESS TIDAL WAVE Research In Motion Daily share price, RIM-TSE Closed yesterday at $135.50, up $4.20 Sierra Wireless Daily share price, SW-TSE Closed yesterday at $128, down $4 eDispatch.com Wireless Data Daily share price, EWD-CDNX Closed yesterday at $11.50, down $1.75