Share Email Print
cover

Proceedings Paper

Neural network Z-plane implementation with very high interconnection rates
Author(s): John C. Carson
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Neural networks offer the potential for a quantum leap in the capabilities of imaging sensor systems. The critical neural network implementation factors are: weighted interconnect between all detector outputs; parallel, linear processing of each detector output; fan-out to multiple (thousands) processing nodes per detector output and the ability to independently change interconnect weights and processor node connections within the detector integration times. For a 128 x 128 pixel detector array, the number of desirable interconnects could be as high as iO per second, compared to the approximately iO rates achieved presently with off-focal plane digital processors. Irvine Sensors Corporation (ISC) has conceived a new way of interconnecting 3-D focal plane readout modules and of laying out their component integrated circuits that appears to fulfill the very high interconnect rate requirements. This concept is described and mterconnectivity and other performance attributes are discussed.

Paper Details

Date Published: 1 November 1990
PDF: 4 pages
Proc. SPIE 1339, Materials, Devices, Techniques, and Applications for Z-Plane Focal Plane Array Technology II, (1 November 1990); doi: 10.1117/12.23008
Show Author Affiliations
John C. Carson, Irvine Sensors Corp. (United States)


Published in SPIE Proceedings Vol. 1339:
Materials, Devices, Techniques, and Applications for Z-Plane Focal Plane Array Technology II
John C. Carson, Editor(s)

© SPIE. Terms of Use
Back to Top