{"title":"A RISC-V Based Open Hardware Platform for Always-On Wearable Smart Sensing","authors":"M. Eggimann, Stefan Mach, M. Magno, L. Benini","doi":"10.1109/IWASI.2019.8791364","DOIUrl":null,"url":null,"abstract":"We present a fully programmable ultra-low-power embedded platform that hosts an \"electronic skin\" (E-skin) arrays of tactile sensors with up to 64 channels, ECG/EMG sensors up to 8 channels, inertial sensors, and a Bluetooth Low Energy 5.0 module. The platform’s compute engine is a heterogeneous multi-core parallel ultra-low power (PULP) processor based on RISC-V, capable of delivering up to 2.5 GOPS, within a 55 mW power consumption envelope, which makes the platform ideal for battery-powered always-on operation. Experimental results show a peak of 38.3x energy efficiency increase (0.7 V, 85 MHz) compared to ARM-Cortex-M microcontrollers with similar power budgets.","PeriodicalId":330672,"journal":{"name":"2019 IEEE 8th International Workshop on Advances in Sensors and Interfaces (IWASI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"13","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 IEEE 8th International Workshop on Advances in Sensors and Interfaces (IWASI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IWASI.2019.8791364","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 13
Abstract
We present a fully programmable ultra-low-power embedded platform that hosts an "electronic skin" (E-skin) arrays of tactile sensors with up to 64 channels, ECG/EMG sensors up to 8 channels, inertial sensors, and a Bluetooth Low Energy 5.0 module. The platform’s compute engine is a heterogeneous multi-core parallel ultra-low power (PULP) processor based on RISC-V, capable of delivering up to 2.5 GOPS, within a 55 mW power consumption envelope, which makes the platform ideal for battery-powered always-on operation. Experimental results show a peak of 38.3x energy efficiency increase (0.7 V, 85 MHz) compared to ARM-Cortex-M microcontrollers with similar power budgets.