Show simple item record

dc.contributor.advisorMorgan, Fearghal
dc.contributor.authorCawley, Seamus
dc.date.accessioned2015-07-24T11:45:54Z
dc.date.available2015-07-24T11:45:54Z
dc.date.issued2013-02-22
dc.identifier.urihttp://hdl.handle.net/10379/5102
dc.description.abstractThe automatic design of intelligent systems has been inspired by biology, specifically the operation of the human brain. Researchers hope to exploit and replicate the brain's ability to adapt and self repair in order to develop robust and error tolerant embedded hardware computational devices. Spiking Neural Networks (SNNs) emulate neural behaviour observed in biology. This thesis describes the successful development of a Network-on-Chip based hardware SNN(EMBRACE-FPGA) and the supporting GA-based SNN training and application implementation tools (SNNDevSys). Hardware SNNs can be configured for multiple applications through programming of neuron spike firing threshold potentials, synaptic weights and the hardware SNN interconnection topology. This thesis describes the hardware SNN architecture and prototype and its application to a range of benchmark control and classification problems. This work has contributed to ongoing EMBRACE-FPGA architecture development within the Bio-Inspired and Reconfigurable Computing research group to improve practical hardware SNN scalability. Phase II of this thesis describes the development of the Remote FPGA Laboratory (RFL). In recent years there has been a growing interest in the development of web-based e-learning systems. The RFL is a web-based distance learning application which enables the teaching of digital systems design using real FPGA devices through a standard web browser. The RFL allows users to configure and interact with FPGA devices via the Internet as part of a combined training and evaluation framework. The system has been designed as an interactive learning tool, which aims to increase student interaction and understanding through a learn-by-doing approach. The system enables better understanding of the operation of digital systems through animation of internal signals, real-time timing diagrams and single-stepping of hardware circuits.en_US
dc.subjectE-learningen_US
dc.subjectVHDLen_US
dc.subjectBio-inspired electronicsen_US
dc.subjectSpiking neural networken_US
dc.subjectHardware evolutionen_US
dc.subjectGenetic algorithmen_US
dc.subjectNetwork on chipen_US
dc.subjectWeb applicationsen_US
dc.subjectDigital systemsen_US
dc.subjectElectrical and Electronic Engineeringen_US
dc.subjectSchool of Engineering and Informaticsen_US
dc.titleHardware spiking neural network and remote FPGA laben_US
dc.typeThesisen_US
dc.contributor.funderSFIen_US
dc.local.noteThe thesis describes the development and application of a bio-inspired evolvable hardware spiking neural network computation device. System operation has been verified for robotics control and radar based breast cancer classification applications. The thesis also describes a web-based e-learning tool developed to enable user control and visualisation of digital systems.en_US
dc.local.finalYesen_US
nui.item.downloads1141


Files in this item

Attribution-NonCommercial-NoDerivs 3.0 Ireland
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. Please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.

The following license files are associated with this item:

Thumbnail

This item appears in the following Collection(s)

Show simple item record