FAR: A 4.12μW ferro-electric auto-recovery for battery-less BSN SoCs
| Autor: | Christopher J. Lukas, Benton H. Calhoun, Farah B. Yahya, Steven Bartling |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Battery (electricity)
Source code Computer science business.industry media_common.quotation_subject 020208 electrical & electronic engineering 02 engineering and technology Non-volatile memory Memory management Backup Transfer (computing) Sensor node Embedded system 0202 electrical engineering electronic engineering information engineering business Energy harvesting media_common |
| Zdroj: | BioCAS |
| DOI: | 10.1109/biocas.2017.8325082 |
| Popis: | This paper presents a Ferro-electric Auto-Recovery (FAR) sub-system designed to support a battery-less body sensor node (BSN) system-on-chip (SoC). FAR enables this type of SoCs to recover from power loss when energy harvesting conditions are poor. It is designed to minimize the power consumption during recovery through architectural optimizations to reduce the number of required accesses to the memory and by introducing a specialized cold-boot bus that speeds-up the transfer of information to the SoC. The bus enables the integration of FAR and the SoC in a compact system-in-package (SiP) to reduce the overall system's form-factor and enable its use in BSNs. FAR has a 2KB boot-up memory to hold the programming code and a 16B backup memory to hold critical data (such as average heart rate and afib events). The proposed architecture allows recovery within a maximum of 64.8ms while consuming 4.12uW. Assuming one recovery per hour (pessimistic), FAR consumes on average 74.1pW. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|