2 Hits in 9.7 sec

Optimized Biosignals Processing Algorithms for New Designs of Human Machine Interfaces on Parallel Ultra-Low Power Architectures

Fabio Montagna
This is followed by several case studies in the biomedical field, starting with the analysis of a Hand Gesture Recognition, based on the Hyperdimensional Computing algorithm, which allows performing a  ...  fast on-chip re-training, and a comparison with the state-of-the-art Support Vector Machine (SVM); then a Brain Machine Interface (BCI) to detect the respond of the brain to a visual stimulus follows in  ...  The amount of power used in sleep mode is 72 µW for the RTC and 3x4.5=13.5 µW for memory retention,which leads to a total power of 85.5 µW.  ... 
doi:10.6092/unibo/amsdottorato/9381 fatcat:kjpnbovoszbdvaipuakhzouzf4

Tuning the Computational Effort: An Adaptive Accuracy-aware Approach Across System Layers

Michael Bromberger
The required tuning of the accuracy-aware methods is integrated into a configuration layer that tunes the available knobs of the accuracy-aware methods integrated into a system.  ...  It proposes an adaptive accuracy-aware approach across system layers that addresses current challenges in that domain, combining and tuning accuracy-aware methods on different system layers.  ...  Acknowledgments Doing a PhD is a long way full of ups and downs. It is not successfully viable without the great support of people and thus I want to express my heartfelt thanks to these people.  ... 
doi:10.5445/ir/1000085787 fatcat:d2aqtv6mmvh7jixw2crnrccmnu