Accession Number:

ADA238183

Title:

A Formulation of a Stochastic Sampling Error Model and a Signal Detection Algorithm for the Aerodynamic Particle Size Analyser

Descriptive Note:

Corporate Author:

DEFENCE RESEARCH ESTABLISHMENT SUFFIELD RALSTON (ALBERTA)

Personal Author(s):

Report Date:

1991-06-01

Pagination or Media Count:

38.0

Abstract:

The determination of the distribution of airborne toxic particles as a function of the aerodynamic diameter provides important information as well as criteria for the definition of hazard as applied to levels of airborne contamination. This is because the aerodynamic particle size distribution embodies the information related to particle density, diameter, shape factor and slip correction that is critical for the characterization of particle motion in settling and impaction and it is these motions that are responsible for particle deposition in the respiratory tract and particle collection in aerosol sampling devices. For a given definition of hazard based on some parameter related to the aerodynamic size distribution, this paper develops a statistical sampling error model for the parameter that is based on the Poisson process. Given that an appropriate sampling program has been designed for the measurement of the size distribution-related parameter with the aerodynamic particle size analyzer, this paper proceeds to the derivation of an optimum detection algorithm for the detection of a signal aerosol sequence in a set of J aerosol samples with a common background. The detection algorithm is based on the generalized likelihood ratio test in which the received count associated with the aerosol sample is modeled as a Poisson distributed random variable.

Subject Categories:

  • Statistics and Probability
  • Air Pollution and Control

Distribution Statement:

APPROVED FOR PUBLIC RELEASE