Information for Paper ID 7052
Paper Information:
Paper Title: Low-Power HW Accelerator for AI Edge-Computing in Human Activity Recognition Systems 
Student Contest: Yes 
Affiliation Type: Academia 
Keywords: artificial intelligence, edge computing, hardware, low-power, human activity recognition 
Abstract: In this paper, an energy efficient HW accelerator for AI edge-computing in Human Activity Recognition is proposed. The system processes samples from a tri-axial accelerometer and classifies the human activities by using a novel Hybrid Neural Network (HNN) topology, which has been designed to reduce the computational complexity of the system while preserving its accuracy. The HW design improves the characteristics of the HNN by means of an architecture that is aimed to reduce the allocated physical resources and the memory accesses. While accuracy measured on ad-hoc dataset is 97.5 %, measurements from synthesis with CMOS 65 nm standard cells report power consumption of 6.3 μW when the sensor ouput data rate is 25 Hz, normally used for HAR. 
Track ID: 10 
Track Name: AICAS in IoT, HCI, Healthcare, Autonomous Systems, Homes/Factories/Cities/Nature 
Final Decision: Accept as Poster 
Session Name: AICAS Applications II (Poster)