Title :
Spiking neural networks´ model with spike frequency adaptation for e-nose
Author :
Badiei, Shirin ; Abdel-Aty-Zohdy, Hoda
Author_Institution :
Dept. of Electr. & Comput. Eng., Oakland Univ., Rochester Hills, MI, USA
Abstract :
We create a spiking neural network of Integrate and Fire neurons with spike frequency adaption based on parameters adjusted for our e-nose device, and investigate the use of this model for odor classification. Addition of spike frequency adaptation term brings the model closer to the response of the olfactory system. Data from Cyranose 320, a polymer based 32-sensor array, is used to test the system and create unique dynamic spiking patterns. The results for four analytes are presented.
Keywords :
chemioception; electronic engineering computing; electronic noses; neural nets; pattern classification; polymers; Cyranose 320; dynamic spiking pattern; e-nose device; fire neurons; integrate neurons; odor classification; olfactory system; polymer based 32-sensor array; spike frequency adaptation; spiking neural network model; Adaptation models; Biological neural networks; Biological system modeling; Computational modeling; Mathematical model; Neurons; Training; e-nose; spiking neural networks;
Conference_Titel :
Aerospace and Electronics Conference (NAECON), Proceedings of the 2011 IEEE National
Conference_Location :
Dayton, OH
Print_ISBN :
978-1-4577-1040-7
DOI :
10.1109/NAECON.2011.6183078
