Retinomorphic Vision Systems II: Communication Channel Design

Creators: Boahen, Kwabena

Abstract

I discuss the tradeoffs faced when asynchronous pulse trains are transmitted among large, two-dimensional, arrays of neurons on different chips, using time-division multiplexing, and present an implementation of an arbitered, random-access, channel. The long cycle time that plagues arbitered channels is addressed in the implementation described here by pipelining. Cycle times ranging from 420ns to 730ns were achieved, for 64 x 64 arrays, in a 2µm CMOS process, yielding a peak throughput of 2.38M spikes/second.

Additional Information

© 1996 IEEE. I thank my advisor, Carver Mead, for sharing his insights into the operation of the nervous system. I thank Misha Mahowald for making available layouts for the arbiter, the address encoders, and the address decoder; John Lazzaro, Alain Martin, and Jose Tierno for helpful discussions on address-events and asynchronous VLSI; Tobi Delbruck for help with the Mac address-event interface; and Jeff Dickson for help with PCB design. This work is supported in part by the Office of Naval Research, by ARPA, by the Beckman Foundation, by the Center for Neuromorphic Systems Engineering as a part of the National Science Foundation Engineering Research Center Program, and by the California Trade and Commerce Agency, Office of Strategic Technology.

Attached Files

Published - 00598465.pdf

Files

00598465.pdf

Files (477.2 kB)

Name	Size	Download all
00598465.pdf md5:f9b924c87f218ea65d9b04954098df83	477.2 kB	Preview Download

Additional details

	All versions	This version
Views	0	0
Downloads	0	0
Data volume	0 Bytes	0 Bytes