Digital Commons Network^™

A Parallel Processing Architecture For Dqdb Protocol Implementation, Nilesh Vinubhai Gandhi

Theses

The high bandwidth transmission links, which have been provided by the advances of Fiber Optics Technology, reduce drastically the packet transmission times and place new demands on the nodal protocol processing. Segmentation and reassembly of packets, computation of checksums, introduction of source and destination addresses, etc., must be performed extremely fast in order to prevent node processing from becoming the bottleneck of the transmission. Parallel processing enables the execution of the previous tasks on multiple packets simultaneously and therefore has the potential of addressing the issue of fast node processing successfully. In this thesis we focus on the Medium Access …

Classification Of Patterns In Eeg Recordings : A Comparison Of Back-Propagation Networks Vs. Predictive Autoencoder Networks, Brian Armieri

Theses

Recent research exploring the use of neural networks for electro-encephalogram (EEG) pattern classification has found that a three-layer back-propagation network could be successfully trained to identify high voltage spike-and-wave spindle (HVS) patterns caused by epileptic seizures (Jando et. al., in press). However, there is no reason to predict that back-propagation is the best possible network architecture for EEG classification. A back-propagation neural network and a predictive autoencoder neural network were compared to determine which network was better at correct classifying both HVS and non-HVS patterns.

Both networks were able to classify 88%-89% of all patterns using a limited set of …