2019/02/02 by Ganapati Bhat, Bhat, Ganapati, Kunal Bagewadi +5
Engineering · #Advanced Sensor and Energy Harvesting Materials #Energy Harvesting in Wireless Networks #FOS: Electrical engineering #Innovative Energy Harvesting Technologies #Signal Processing (eess.SP) #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1902.02639
openalex publication_date 2019/02/02 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
The use of wearable and mobile devices for health monitoring and activity\nrecognition applications is increasing rapidly. These devices need to maximize\ntheir accuracy and active time under a tight energy budget imposed by battery\nand small form-factor constraints. This paper considers energy harvesting\ndevices that run on a limited energy budget to recognize user activities over a\ngiven period. We propose a technique to co-optimize the accuracy and active\ntime by utilizing multiple design points with different energy-accuracy\ntrade-offs. The proposed technique switches between these design points at\nruntime to maximize a generalized objective function under tight harvested\nenergy budget constraints. We evaluate the proposed approach experimentally\nusing a custom hardware prototype and fourteen user studies. The proposed\napproach achieves both 46% higher expected accuracy and 66% longer active time\ncompared to the highest performance design point.\n