Pulse coupled neural networks (PCNN), wavelets and radial basis functions: olfactory sensor applications

摘要

This paper focuses on novel approaches to olfactory sensor signal analysis: 1) PCNN focusing, 2) wavelet characterization and 3) radial basis function detection of anomalies. Novel decision aide approaches are developed to facilitate design of routines to automatically trigger calibration and/or analysis when transient signals indicate an anomaly. These three signal analysis steps are integrated into a decision system to separate sensor faults from targets, and then recalibrate the target detection and identification procedures.