This  project will address the challenges of implementing complex DNN models for energy efficient and performance driven IoT edge devices. The research will develop hardware architectures that are scalable, programmable, and easily adaptable to different DNN models and applications. The DNN architectures developed will be demonstrated in hardware for a wearable healthcare application such as heartbeat identification.

Deep neural networks (DNNs), which are loosely modelled after the human brain, have been shown to be remarkably successful in recognising and interpreting patterns in many applications such as image and video processing. However, due to high computational complexity, power, and resource requirements, DNNs are not well explored for low power applications such as Internet of Things (IoT) sensors, wearable healthcare etc. IoT devices have stringent energy and resource constraints and, in many cases, deal with one-dimensional time series data.This project aims to develop energy efficient IoT sensors that can perform deep learning and pattern recognition at the edge of the network. The proposed research can potentially transform various industries, such as connected health, weather forecasting, autonomous manufacturing  by providing improved turnaround times and lower costs, and thus will enhance the economic opportunities and competitiveness of these industries.

This project was allocated funding following the MCCI call for proposals for research into innovative future technology solutions in the area of microelectronics. The centre awarded over €5 million in funding to eight MCCI researchers for research into deeptech microelectronic solutions such as beyond 5G wireless communications, implantable biomedical devices, IoT, space and satellite electronics, and sustainable electronics.

To find out more please click here