ERA

Download the full-sized PDF of Signal Compression Methods for Low-Power ImplantsDownload the full-sized PDF

Analytics

Share

Permanent link (DOI): https://doi.org/10.7939/R35X0Q

Download

Export to: EndNote  |  Zotero  |  Mendeley

Communities

This file is in the following communities:

Graduate Studies and Research, Faculty of

Collections

This file is in the following collections:

Theses and Dissertations

Signal Compression Methods for Low-Power Implants Open Access

Descriptions

Other title
Subject/Keyword
neural
circuits
compression
Type of item
Thesis
Degree grantor
University of Alberta
Author or creator
Dodd, Russell D
Supervisor and department
Cockburn, Bruce (Electrical and Computer Engineering)
Examining committee member and department
Mushahwar, Vivian (Biomedical Engineering)
Wilton, Steve (Electrical and Computer Engineering, University of British Columbia)
Vorobyov, Sergiy (Electrical and Computer Engineering)
Department
Department of Electrical and Computer Engineering
Specialization
Integrated Circuits and Systems
Date accepted
2014-07-04T15:44:24Z
Graduation date
2014-11
Degree
Doctor of Philosophy
Degree level
Doctoral
Abstract
The pairing of implantable micro-electrode arrays with micro-electronic devices allows the observation of neural activity within nervous systems in a more comprehensive and potentially more effective way compared to single electrode recording. State-of-the-art tethered recording systems use implanted micro-electrode arrays and can record raw data rates of at least 24 Mbps over 100 channels. To make the systems fully implantable, the communication link needs to be wireless with the system power dissipation low enough for battery and/or near-field wireless operation. This thesis details a study of signal compression methods intended for implanted wireless low-power neural signal recording implementations. ASIC simulations of compression methods are presented for proof of concept and comparisons, while ASIC implementations provide power consumption measurements for select methods. It is shown by using a windowed noise-based dual-threshold neural spike detector that an energy savings of 90% can be expected compared to systems with no compression. Furthermore, an additional 80% compression per spike can be achieved using multilevel wavelet decomposition with no effect on the classification accuracy and relatively small increase in power consumption. Finally, a feature extraction method is presented and is simulated to demonstrate a neural spike compression ratio of 87% and classification accuracy of 96%.
Language
English
DOI
doi:10.7939/R35X0Q
Rights
Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
Citation for previous publication

File Details

Date Uploaded
Date Modified
2015-01-08T08:01:59.375+00:00
Audit Status
Audits have not yet been run on this file.
Characterization
File format: pdf (PDF/A)
Mime type: application/pdf
File size: 37874889
Last modified: 2015:10:12 12:58:17-06:00
Filename: Dodd_Russell_D_201407_PhD.pdf
Original checksum: 1064fc299afba190f6e9e2fc12530df1
Activity of users you follow
User Activity Date