An application of wavelet transforms and neural networks for decomposition of millimeter-wave spectroscopic signals

Conference · Sat Jul 01 00:00:00 EDT 1995

OSTI ID:192491

Gopalan, K ^[1]; Gopalsami, N; Bakhtiari, S; Raptis, A C ^[2]

---

1. Purdue Univ. Calumet, Hammond, IN (United States)
2. Argonne National Lab., IL (United States)

This paper reports on wavelet-based decomposition methods and neural networks for remote monitoring of airborne chemicals using millimeter wave spectroscopy. Because of instrumentation noise and the presence of untargeted chemicals, direct decomposition of the spectra requires a large number of training data and yields low accuracy. A neural network trained with features obtained from a discrete wavelet transform is demonstrated to have better decomposition with faster training time. Results based on simulated and experimental spectra are presented to show the efficacy of the wavelet-based methods.

• Research Organization: Argonne National Lab., IL (United States)
• Sponsoring Organization: USDOE, Washington, DC (United States)
• DOE Contract Number: W-31109-ENG-38
• OSTI ID: 192491
• Report Number(s): ANL/ET/CP--87252; CONF-951117--1; ON: DE96005225
• Country of Publication: United States
• Language: English

Related Subjects

44 INSTRUMENTATION
INCLUDING NUCLEAR AND PARTICLE DETECTORS
54 ENVIRONMENTAL SCIENCES
99 GENERAL AND MISCELLANEOUS
AIR POLLUTION MONITORING
CHEMICAL ANALYSIS
DECOMPOSITION
NEURAL NETWORKS
REMOTE SENSING
SPECTROSCOPY