Estimation of Poisson intensity in the presence of dead time

• 成果类型: Article
• 署名作者: He, SY; Yang, GL; Fang, KT; Widmann, JF
• 署名单位: Peking University; Peking University; University System of Maryland; University of Maryland College Park; National Institute of Standards & Technology (NIST) - USA; Hong Kong Baptist University; National Institute of Standards & Technology (NIST) - USA
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1198/016214504000001547
• 发表日期: 2005
• 页码: 669-679
• 关键词: spray; analyzer

摘要

Phase Doppler interferometry (PDI) is a nonintrusive technique frequently used to obtain information about spray characteristics. Understanding spray characteristics is of critical importance in many areas of science, including liquid fuel spray combustion, spray coatings, fire suppression, and pesticides. PDI measures the size and velocity of individual droplets in a spray. Due to the design of the instrument, recordings of the PDI contain gaps, called dead times. The presence of recurring dead times greatly complicates estimation of the diffusion rate of the droplets. Modeling the spray process as a homogeneous Poisson process, we construct consistent and asymptotic normal estimators of the diffusion rate (Poisson intensity) under various conditions. Simulation produced a good agreement between our estimators (in the presence of dead time) and the maximum likelihood estimates obtained without dead time. We use experimental data to illustrate the estimation method.