farjamit@eecs.cs.pdx.edu (Tom Farjami) (04/07/91)
Greatings,
I only got two emails in this regard so I am posting
this again. I am working on design and fabrication of a CMOS
motion detector IC that I have classified as being fuzzy.
The following is a description of the circuit. It has 10
digital inputs and a final output. A strong voltage on the
output (near VDD) means that majority of the inputs were
fired in the preferred direction, otherwise output is
silent. There is no feedbacks and therefor no neural-net
like back props in this circuit. It detects motion in real
time. Circuit is biologically motivated and operates in
subthreshold.
In my previous posting I asked the following
question,"Basically what are some of the applications of
this circuit in real life ?". I have thought of waveform
detection in scopes, pattern recognition and even speech
recognition myself. How would that be?. Consider the fol-
lowing. If inputs of the motion detector came from a set of
comparators such that all of the comparators were comparing
the same analog input signal( say a speech signal) against a
constant (programmable) voltages. That is each comparator
has a different reference voltage but takes in the same ana-
log signal as other comparators do. Now if the analog
waveform matched the wright pattern in the preferred direc-
tion ( which is basically what you programmed the compara-
tors with), motion detector will turn on and we have pattern
detection. Ofcourse for this I need more that 10 inputs,
perhaps hundreds to give me high resolution and a state
machine plus a A/D converter to program my reference vol-
tages; but this can be.
I especially like to hear comments from those doing
speech and pattern recognition.
Regards,
Tom Farjami
Portland State University
April 6, 1991