Speaker:

Tohru Shimizu, Renesas Electronics Corporation, Japan

Title:

Normally-off Computing and Application to the Sensor-net

Abstract:

Global optimization of the power management is an important issue for balancing service comfort and energy saving. In order to achieve the optimization, real-time monitoring of the social and the natural environmental status information is necessary. The sensor network is one of the technologies to satisfy the requirement. Considering the configuration of the sensor network for monitoring the environmental information, the number of the sensor nodes is increasing, the more precisely and more widely the environment can be covered. However, increase of the number of sensor nodes means increase of the power for the sensor network operation. The normally-off computing technology is the way to reduce the total power of the sensor network as well as the low power design of LSIs for the sensor nodes. I would like to discuss the concept and the target of the normally-off sensor node development.

Bio:

Dr. Shimizu received B.S., M.S., and Ph.D. degrees of Information Science from The University of Tokyo, Japan. Since 1986, he has been involved in microprocessor, microcontroller and SoC design R&D, working in Mitsubishi Electric, Renesas Technology, and Renesas Electronics. He lead many design projects of embedded microprocessor LSIs. Leading edge RISC microprocessors with embedded DRAM, microcontrollers with embedded flash memory, multi-core SoCs for embedded processing are some of the project outputs. Most of the designs were presented in the ISSCC and the A-SSCC, and have been applied to products of Mitsubishi and Renesas. His R&D domain covers not only LSI design and architecture but embedded software and application systems. He was a TPC member of the ISSCC and he is a steering committee member of the A-SSCC. He is a director of the executive committee of the Embedded Technology, the largest embedded system show in Japan.