Flexible, Self-Adaptive Sense-and-Compress SoC for Sub-microWatt Always-On Sensory Recording

Jaro De Roose; Haoming Xin; Ahmed Hallawa; Gerd Ascheid; Pieter J.A. Harpe; Marian Verhelst

doi:10.1109/LSSC.2020.3018382

Flexible, Self-Adaptive Sense-and-Compress SoC for Sub-microWatt Always-On Sensory Recording

Jaro De Roose (Corresponding author)
, Haoming Xin
, Ahmed Hallawa
, Gerd Ascheid
, Pieter J.A. Harpe
, Marian Verhelst

Research output: Contribution to journal › Article › Academic › peer-review

4 Citations (Scopus)

119 Downloads (Pure)

Abstract

We present a 5-sensor, fully integrated sensing system with interchangeable sensors and programmable configuration to create a sub-microWatt multisensor node that can tackle a wide range of sensing applications. Furthermore, the sensor node is capable of autonomously adapting its configuration to the application requirements hence minimizing system power. Such self-reconfiguration is enabled at low overhead by developing an automated offline optimization strategy, in combination with an autonomous embedded configuration controller, using the concept of behavioral trees (BTs). The resulting fully integrated platform consumes a maximum of 321 nW when sampling at 500 Hz and 3025 nW at 8 kHz. Furthermore, we demonstrate the end-to-end autonomous optimization flow for two different applications exploiting different sensors: 1) human activity recognition using accelerometers and 2) machine listening using a microphone. Both use cases demonstrate that the introduced system and methodology reduces the power by more than a factor 2 without losing significant application detection accuracy.

Original language	English
Article number	9171886
Pages (from-to)	362-365
Number of pages	4
Journal	IEEE Solid-State Circuits Letters
Volume	3
DOIs	https://doi.org/10.1109/LSSC.2020.3018382
Publication status	Published - 2020

Funding

Manuscript received May 18, 2020; revised July 14, 2020; accepted August 9, 2020. Date of publication August 20, 2020; date of current version September 29, 2020. This article was approved by Associate Editor Stefan Rusu. This work was supported in part by EU H2020 under Grant 66534; and in part by EU-ERC under Grant ERC-2016-STG-71503. (Corresponding author: Jaro De Roose.) Jaro De Roose and Marian Verhelst are with the Department of Electrical Engineering, KU Leuven, 3000 Leuven, Belgium (e-mail: [email protected]).

Keywords

Automated optimization
behavioral tree (BT)
flexible sensor node
low overhead flexible hardware
ultralow power

Access to Document

10.1109/LSSC.2020.3018382

publisher versionFinal published version, 2.04 MBLicence: TAVERNE

Cite this

@article{06c87b2da30641f694385dc1572d9a30,

title = "Flexible, Self-Adaptive Sense-and-Compress SoC for Sub-microWatt Always-On Sensory Recording",

abstract = "We present a 5-sensor, fully integrated sensing system with interchangeable sensors and programmable configuration to create a sub-microWatt multisensor node that can tackle a wide range of sensing applications. Furthermore, the sensor node is capable of autonomously adapting its configuration to the application requirements hence minimizing system power. Such self-reconfiguration is enabled at low overhead by developing an automated offline optimization strategy, in combination with an autonomous embedded configuration controller, using the concept of behavioral trees (BTs). The resulting fully integrated platform consumes a maximum of 321 nW when sampling at 500 Hz and 3025 nW at 8 kHz. Furthermore, we demonstrate the end-to-end autonomous optimization flow for two different applications exploiting different sensors: 1) human activity recognition using accelerometers and 2) machine listening using a microphone. Both use cases demonstrate that the introduced system and methodology reduces the power by more than a factor 2 without losing significant application detection accuracy.",

keywords = "Automated optimization, behavioral tree (BT), flexible sensor node, low overhead flexible hardware, ultralow power",

author = "\{De Roose\}, Jaro and Haoming Xin and Ahmed Hallawa and Gerd Ascheid and Harpe, \{Pieter J.A.\} and Marian Verhelst",

year = "2020",

doi = "10.1109/LSSC.2020.3018382",

language = "English",

volume = "3",

pages = "362--365",

journal = "IEEE Solid-State Circuits Letters",

issn = "2573-9603",

publisher = "Institute of Electrical and Electronics Engineers",

}

TY - JOUR

T1 - Flexible, Self-Adaptive Sense-and-Compress SoC for Sub-microWatt Always-On Sensory Recording

AU - De Roose, Jaro

AU - Xin, Haoming

AU - Hallawa, Ahmed

AU - Ascheid, Gerd

AU - Harpe, Pieter J.A.

AU - Verhelst, Marian

PY - 2020

Y1 - 2020

N2 - We present a 5-sensor, fully integrated sensing system with interchangeable sensors and programmable configuration to create a sub-microWatt multisensor node that can tackle a wide range of sensing applications. Furthermore, the sensor node is capable of autonomously adapting its configuration to the application requirements hence minimizing system power. Such self-reconfiguration is enabled at low overhead by developing an automated offline optimization strategy, in combination with an autonomous embedded configuration controller, using the concept of behavioral trees (BTs). The resulting fully integrated platform consumes a maximum of 321 nW when sampling at 500 Hz and 3025 nW at 8 kHz. Furthermore, we demonstrate the end-to-end autonomous optimization flow for two different applications exploiting different sensors: 1) human activity recognition using accelerometers and 2) machine listening using a microphone. Both use cases demonstrate that the introduced system and methodology reduces the power by more than a factor 2 without losing significant application detection accuracy.

AB - We present a 5-sensor, fully integrated sensing system with interchangeable sensors and programmable configuration to create a sub-microWatt multisensor node that can tackle a wide range of sensing applications. Furthermore, the sensor node is capable of autonomously adapting its configuration to the application requirements hence minimizing system power. Such self-reconfiguration is enabled at low overhead by developing an automated offline optimization strategy, in combination with an autonomous embedded configuration controller, using the concept of behavioral trees (BTs). The resulting fully integrated platform consumes a maximum of 321 nW when sampling at 500 Hz and 3025 nW at 8 kHz. Furthermore, we demonstrate the end-to-end autonomous optimization flow for two different applications exploiting different sensors: 1) human activity recognition using accelerometers and 2) machine listening using a microphone. Both use cases demonstrate that the introduced system and methodology reduces the power by more than a factor 2 without losing significant application detection accuracy.

KW - Automated optimization

KW - behavioral tree (BT)

KW - flexible sensor node

KW - low overhead flexible hardware

KW - ultralow power

UR - https://www.scopus.com/pages/publications/85092555405

U2 - 10.1109/LSSC.2020.3018382

DO - 10.1109/LSSC.2020.3018382

M3 - Article

AN - SCOPUS:85092555405

SN - 2573-9603

VL - 3

SP - 362

EP - 365

JO - IEEE Solid-State Circuits Letters

JF - IEEE Solid-State Circuits Letters

M1 - 9171886

ER -

Flexible, Self-Adaptive Sense-and-Compress SoC for Sub-microWatt Always-On Sensory Recording

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this