Lab Home | Phone | Search
	Center for Nonlinear Studies
	Home  ▶ People   Current   Executive Committee   Postdocs   Visitors   Students   Research   Publications  ▶ Conferences   Workshops   Sponsorship  ▶ Talks   Seminars   Postdoc Seminars Archive   Quantum Lunch   Quantum Lunch Archive   P/T Colloquia   Archive   Ulam Scholar     Postdoc Nominations  ▶ Student Requests   Student Program  ▶ Visitor Requests   Description   Past Visitors  ▶ Services   General     History of CNLS     Maps, Directions   CNLS Office   T-Division   LANL				Wednesday, November 07, 201809:00 AM - 10:00 AMNEW: ACL Conference Room (TA-3, Building 200, Room 116) Seminar Bayesian Inference for Raman Spectroscopy using a Particle Filter Matt MooresUniversity of Wollongong The SuperCam instrument for the Mars 2020 mission will include a Raman spectrometer, which will be the first time that measurements of this type have been performed on the Martian surface. Raman scattering produces peaks that correspond to the vibrational modes of the molecules in a sample, so it provides complementary information to LIBS and FTIR. The main difficulty in analysing this data is due to overlapping peaks and shoulders, which makes baseline correction extremely challenging. My talk will introduce the R package ‘serrsBayes,’ which uses informative Bayesian priors to fit a model of the peak broadening functions, while accounting for the baseline and additive white noise. We achieve this using a Rao-Blackwellized Particle Filter, which is a form of sequential Monte Carlo (SMC) algorithm. I will also discuss suitable sources of prior information, such as online databases of Raman spectra, a computational forward model known as time-dependent density functional theory (TD-DFT), as well as pre-flight calibration data. Host: Emily Casleton