• Login
    View Item 
    •   UBIR Home
    • Theses and Dissertations (2018-present)
    • 2018-09-01 UB Theses and Dissertations (public)
    • View Item
    •   UBIR Home
    • Theses and Dissertations (2018-present)
    • 2018-09-01 UB Theses and Dissertations (public)
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Cooperated Sensing and Computing for Energy-Efficient Wearables

    Thumbnail
    View/Open
    Wang_buffalo_0656A_15910.pdf (12.50Mb)
    Date
    2018
    Author
    Wang, Aosen
    0000-0002-9825-9330
    Metadata
    Show full item record
    Abstract
    Nowadays, miniature and intelligence make wearable system advance to the leading-edge applications, such as telemedicine, patient rehabilitation and chronic disease analysis. However, energy-efficiency of wearable-end sensor nodes becomes a big issue for user experience in these applications. Huge volume of biodata sensing and intelligent bioevent analysis rapidly drain the battery storage in sensor node. Therefore, energy-efficiency of wearable sensor node needs to be tackled in urgent. With more energy savings, the lifetime of wearable system can be prolonged to reduce the recharging trouble. Better energy-efficiency can even improve the form factor of wearable nodes.To address the energy-efficiency challenge, we proposed a holistic compressed sensing (CS) based framework to jointly optimize the sensing architecture and computing architecture. We design our sensing architecture from three aspects, including parameter agility, information screen and signal dynamics, to exploit the opportunity of energy savings. We integrate circuit-level design parameter into CS architecture to explore better trade-off between energy and performance with the help of time-efficient design space exploration algorithm. We develop an information screening module to complete the low-effort data analysis task in the sensor node to avoid the energy-hungry wireless transmission. We also propose dynamic knob to make the sensing architecture adapting to the data dynamics, which is a common case in biosiganl processing. Finally, we investigate the optimization of computing architecture by cross-end implementation of the finer-grained computing primitives.Our contribution is three-fold: 1) We divide the intractable energy-efficiency problem of wearable system into two easier sub-parts, sensing and computing. We optimize our design from these two aspects jointly; 2) We model and develop algorithms to guide the improvements of the architecture and design; 3) We propose a holistic solution to tackle all the challenges and make large-step improvement on energy efficiency of wearable system.
    URI
    http://hdl.handle.net/10477/78510
    Collections
    • 2018-09-01 UB Theses and Dissertations (public)

    To add content to the repository or for technical support: Contact Us
     

     

    Browse

    All of UBIRCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsTypesThis CollectionBy Issue DateAuthorsTitlesSubjectsTypes

    My Account

    LoginRegister

    To add content to the repository or for technical support: Contact Us