A high-speed analog neural processor

Author

Masa, P. ; Hoen, K. ; Wallinga, H.

Author_Institution

MESA Res. Inst., Twente Univ., Enschede, Netherlands

Volume

14

Issue

3

fYear

1994

fDate

6/1/1994 12:00:00 AM

Firstpage

40

Lastpage

50

Abstract

Targeted at high-energy physics research applications, our special-purpose analog neural processor can classify up to 70 dimensional vectors within 50 nanoseconds. The decision-making process of the implemented feedforward neural network enables this type of computation to tolerate weight discretization, synapse nonlinearity, noise, and other non-ideal effects. Although our prototype does not take advantage of advanced CMOS technology, and was fabricated using a 2.5-μm CMOS process, it performs 6 billion multiplications per second, with only 2 W dissipation, and has as high as 1.5 Gbyte/s equivalent bandwidth.

Keywords

CMOS integrated circuits; feedforward neural nets; linear integrated circuits; microprocessor chips; neural chips; 2 W; 2.5 micron; CMOS process; feedforward neural network; high-energy physics research applications; high-speed analog neural processor; synapse nonlinearity; weight discretization; CMOS process; CMOS technology; Computer networks; Detectors; Feedforward neural networks; Hardware; Nanobioscience; Neural networks; Neurons; Physics;

fLanguage

English

Journal_Title

Micro, IEEE

Publisher

ieee

ISSN

0272-1732

Type

jour

DOI

10.1109/40.285223

Filename

285223