The BioNetExplorer framework systematically generates and explores multiple DNN architectures for bio-signal processing in wearable devices. Our framework varies key neural architecture parameters to search for an embedded DNN with low hardware overhead that can be deployed in wearable edge devices to analyze the bio-signal data and to extract the relevant information, such as arrhythmia and seizure. Furthermore, BioNetExplorer reduces the exploration time by deploying genetic algorithms, such as NSGA-II, SPEA-2, etc. Our framework also enables hardware-aware DNN architecture search by imposing user requirements and hardware constraints (storage, FLOPs, etc.) during the exploration stage, thereby limiting the number of networks explored.
In case of usage, please refer to:
B. S. Prabakaran, A. Akhtar, S. Rehman, O. Hasan, M. Shafique, “BioNetExplorer: Architecture-Space Exploration of Bio-Signal Processing Deep Neural Networks for Wearables,” IEEE Internet of Things Journal, 2021.
BioNetExplorer
Framework for Exploration of Bio-signal Processing DNNs in Wearables
Brought to you by:
crossbow
Downloads:
0 This Week