Chair: John Trnka , IBM Corp., Rochester, MN

ISSCC Executive Committee Chair

Associate Chair: Richard Crisp , Rambus, Inc., Mountain View, CA

ISSCC Program Committee Chair

Naoki Noda, Toyota Motor Corp., Toyota-cho, Toyota, Japan

Worldwide annual automobile production currently totals more than 50M units and continues to grow. Automobiles are mass-market products comparable in scale to the production of color TVs and personal computers. Extensive innovations are occuring as witnessed by the widespread use of electronics to improve the environment and safety, and also in more recent applications based on the use of information technology to provide greater convenience and comfort. This can be described as a revolution of automotive design with electronics and semiconductors as enabling technologies.

One of the two main streams is electric and hybrid electric vehicles that have revolutionized automotive power sources. The other stream is the rapid evolution of intelligent transport systems (ITS) that can be considered as the information revolution for cars. As cars, hitherto standing alone, are integrated into highly-developed information networks, ITS will undergo rapid evolution for next-generation cars. Technologies supporting developments in automotive electronic systems are semiconductor devices and ICs. They provide high-speed data communications between cars and the infrastructure as well as high-speed data processing, high current power electronics, and analog/digital signal processing inside the car.

The technological innovations taking place in cars are reviewed and the increasing expectations for semiconductors are described from the viewpoint of automakers.

Mobile Society 10:20 AM
Guenter Weinberger , Infineon Technologies, Munich, Germany

Universal access to a wide range of telecommunication services, from speech over multimedia to navigation and mobile computing, must be provided for the mobile society of the next millennium. The backbone of this scenario is a worldwide high-bandwidth wireless system. The wireless terminals of the next millennium, comprising all conceivable services in small size with long operating time require significant improvements in all underlying technologies as well as new architectures and new development strategies and design tools. The computing power needed from several tens of millions transistors at high clock rates requires new approaches simultaneously to allow maximum performance and minimum leakage current.

Revived conversion principles like direct downconversion start to be seen in volume production. These concepts put most selection into baseband, avoiding a SAW filter and drasticly reducing the number of RF blocks. Power amplifiers still use expensive GaAs HBTs. New concepts will improve power-added efficiency and give GaAs a push. For low-transmit-power applications silicon PAs with higher performance and lower cost soon will be feasible. MEMS technologies will allow integration of mechanical antenna switch as well as other discrete RF circuitry. The remaining discretes will be integrated by advanced packaging technologies such as subsystem assembly, multi-chip or system module packages.

With present growth rates, the sales of wireless terminals will exceed those of computers and internet nodes. These terminals will drive CMOS technology.

Mark R. Pinto, Lucent Technologies, Microelectronics Group,

Allentown, PA

Enabled by the seemingly never-ending progress of silicon technology, the scope of the task of developing an IC has been broadening - orders of magnitude beyond that which worried industry leaders in the early sixties. As we enter the new millennium, devices are being fabricated using processes managed at atomic levels while IC design involves detailed systems engineering, including direct consideration or even incorporation of application content. By attempting to leverage that entire chain of associated technologies, ever more powerful end-user functionality is enabled, e.g. mobile information access at the most practical costs, power levels, and form factors. Underlying this revolution is that on-chip interconnect is cheaper than going off-chip and the cost of nearly any component on a chip has been consistently reduced 20-30% per year. Additionally, the corresponding improvements in overall system power dissipation, miniaturization, and I/O bandwidth have driven integration of what once were considered distinct technologies, e.g. logic, analog, memory, and RF. While the vision of completely monolithic systems is practical, a level of analysis much more complex than that for the homogeneous CMOS IC is required to justify appropriate directions. Looking further forward, exponential improvements clearly cannot continue forever, and limitations may arise from a variety of interrelated sources - physics, economics, complexity of the task, or lack of applications.

This presentation focuses on the trends of system-level integration in ICs - the market-driving forces, current and emerging technical solutions, and predictions as to where the underlying process technology capability will saturate, at which point continued system level improvements will rely even more on the level of efficiency in mapping transformations of "atoms to applets".