Skip to main content

Research Repository

Advanced Search

Consideration of multiplexing in neural network hardware

Craven, M.P.; Curtis, K.M.; Hayes-Gill, B.R.

Authors

K.M. Curtis

BARRIE HAYES-GILL BARRIE.HAYES-GILL@NOTTINGHAM.AC.UK
Professor of Electronic Systems and Medical Devices



Abstract

This paper presents the results of research into a scheme for overcoming the communications 'bottleneck' within hardware neural networks, utilising frequency division multiplexing of amplitude modulated neural signals. The introduction explains the problem of connectivity in neural networks and the need for multiplexing, followed by an introduction to the FDM solution. A comparison of FDM with TDM is then presented which shows that it is necessary to allow a moderate amount of band overlap in the FDM channel, to utilise the available bandwidth effectively. This approach is possible by virtue of the inherent error tolerance of the neural system..

Citation

Craven, M., Curtis, K., & Hayes-Gill, B. (1994). Consideration of multiplexing in neural network hardware. IEE Proceedings Circuits Devices and Systems, 141(3), 237-240. doi:10.1049/ip-cds:19941103

Journal Article Type Article
Publication Date 1994
Deposit Date Nov 27, 2018
Journal IEE Proceedings - Circuits, Devices and Systems
Print ISSN 1350-2409
Publisher Institution of Engineering and Technology
Peer Reviewed Peer Reviewed
Volume 141
Issue 3
Pages 237-240
DOI https://doi.org/10.1049/ip-cds%3A19941103
Public URL https://nottingham-repository.worktribe.com/output/1282385
Publisher URL https://ieeexplore.ieee.org/document/296523?arnumber=296523