A new class of silicon systems for AI connected devices

Researchers from the National University of Singapore have demonstrated a new class of silicon systems that could enhance the energy efficiency of AI connected devices. These technological breakthroughs could also enhance the capabilities of the semiconductor industry in Singapore.

The innovation was demonstrated in a fully-depleted silicon-on-insulator (FD-SOI) and can be applied to the design and fabrication of advanced semiconductor components for AI applications. The new chip has the potential to extend the battery life of wearables and smart objects by a factor of 10, support intense computational workloads for use in Internet of Things applications and reduce the power consumption associated with wireless communications with the cloud.

Professor Massimo Alioto from the NUS College of Design and Engineering said IoT devices often operate on a limited power budget and therefore require low average power to efficiently perform regular tasks such as physical signal monitoring. At the same time, high performance is needed to process occasional signal events with computationally intensive AI algorithms. “Our research allows us to simultaneously reduce the average power and improve the peak performance. The applications are wide-ranging and include smart cities, smart buildings, Industry 4.0, wearables and smart logistics,” Alioto said.

The research findings could also allow researchers to enhance the area of battery-powered AI devices, as they could move intelligence from conventional cloud to smart miniaturised devices. The research conducted by the NUS FD-SOI Always-on Intelligent & Connected Systems (FD-fAbrICS) joint lab revealed that the researchers’ FD-SOI chip technology can be deployed at scale with enhanced design and system integration productivity for faster market reach and rapid industry adoption.

“This innovation has the potential to accelerate the time to market for key players in Singapore’s semiconductor ecosystem,” Alioto said.

Going forward, the researchers hope to facilitate the adoption and deployment of their design technologies at scale. This will enable the AI and semiconductor industry in Singapore to have a competitive advantage while reducing the overall deployment cost of FD-SOI systems. The NUS research team will now look into developing new classes of intelligent and connected silicon systems that could support larger AI model sizes (“large models”) for generative AI applications. The resulting decentralisation of AI computation from cloud to distributed devices will preserve privacy and avoid wireless data deluge under the simultaneous presence of a range of devices.

Image credit: iStock.com/luismmolina