Title :
A Low-Power Haar-Wavelet Preprocessing Approach for a SNN Olfactory System
Author :
Allen, Jacob N. ; Hasan, Safa B. ; Abdel-Aty-Zohdy, Hoda S. ; Ewing, Robert L.
Author_Institution :
Oakland Univ. Rochester, Oakland
Abstract :
A low frequency and low power spiking neural network chip is designed to classify polymer film electronic nose patterns. A simulation model for polymer film chemical sensors is developed and compared to other approaches. The final chip design uses wavelet pre-processing for fault tolerance and operates at just 20 kHz.
Keywords :
Haar transforms; chemioception; electronic noses; neural chips; pattern classification; polymer films; wavelet transforms; Haar-wavelet preprocessing; frequency 20 kHz; olfactory system; pattern classification; polymer film chemical sensor; polymer film electronic nose; spiking neural network chip; Analytical models; Chemical sensors; Electronic noses; Force sensors; Neural networks; Olfactory; Polymer films; Semiconductor device modeling; Sensor arrays; Substrates;
Conference_Titel :
Design and Test Workshop, 2007. IDT 2007. 2nd International
Conference_Location :
Cairo
Print_ISBN :
978-1-4244-1824-4
Electronic_ISBN :
978-1-4244-1825-1
DOI :
10.1109/IDT.2007.4437464
