A 41 μW real-time adaptive neural spike classifier

Conference paper (2016)

Authors

Amir Zjajo Signal Processing Systems -

T.G.R.M. van Leuken Signal Processing Systems -

Research Group

Signal Processing Systems () (TU Delft)

DOI: https://doi.org/10.1109/bhi.2016.7455941

Feature extraction Training Sorting Kernel Registers Support vector machine classification

To reference this document use:

http://resolver.tudelft.nl/uuid:3a6b78eb-1acd-42e7-94bf-ef076b20c910

More Info

expand_more

Published Date

21-04-2016

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Microelectronics

Research Group

Signal Processing Systems

Abstract

Robust, power- and area-efficient spike classifier, capable of accurate identification of the neural spikes even for low SNR, is a prerequisite for the real-time, implantable, closed-loop brain-machine interface. In this paper, we propose an easily-scalable, 128-channel, programmable, neural spike classifier based on nonlinear energy operator spike detection, and a boosted cascade, multiclass kernel support vector machine classification. The power-efficient classification is obtained with a combination of the algorithm and circuit techniques. The classifier implemented in a 65 nm CMOS technology consumes less than 41 μW of power, and occupy an area of 2.64 mm2.

Files

Zjajo201641.pdf

(.pdf | 1.05 Mb)