Paper Info
Title
An Information-Theoretic Framework For Joint Architectural And Circuit Level Optimization For Olfactory Recognition Processing
Abstract
Signal processing tasks such as classification or recognition may benefit from implementation strategies inspired by biological sensory pathways. In this paper, we employ an information-theoretic framework to explore the possible energy savings that can result from such an approach when applied to a specific problem, the artificial olfactory system. A preliminary evaluation of the efficiency versus SNR trade-offs for different signal representations demonstrates the advantage of one continuous-time discrete-valued (CTDV) signal representation in the low-precision regime with respect to the digital approach. These results are consequently applied to a joint circuit-architecture optimization of an artificial olfaction signal processing system, leading to promising indications of potential energy savings.
Year
DOI
Venue
2011
10.1109/SiPS.2011.6088943
2011 IEEE WORKSHOP ON SIGNAL PROCESSING SYSTEMS (SIPS)
Keywords
Field
DocType
Signal representation, recognition, olfactory system, information theory, low power
Information theory,Signal processing,Olfactory system,Olfaction,Computer science,Signal-to-noise ratio,Signal classification,Artificial intelligence,Sensory system,Machine learning
Conference
ISSN
Citations 
PageRank 
1520-6130
0
0.34
References 
Authors
4
3
Name
Order
Citations
PageRank
Ping-Chen Huang1142.76
David Macii232338.74
Jan M. Rabaey347961049.96