The model of random signals generated by optical particle counter and the instrument improvement

In order to study and improve atmospheric and air pollution monitoring sensor, a new mathematical model of random signals is established based on measuring process of light scattering signals analyzed by laser particle counter which combines the high speed data acquisition card PCI-9812 and optical particles counting sensor. The measured random signals can be divided into stability constant part and random variation part. The performance of the instrument is improved by both this model and analytical methods. Statistical distributions of the amplitude of the standard particles with different diameters are studied by the original experiment and improved one. The resolving power of particle size could attain more than 90%. The results reveal statistical distributions match well with lognormal distribution with a natural number as an independent variable. The lognormal distribution plays an important role in describing the random fluctuation characteristics of random process in both theories and experiments. Furthermore, both normal and lognormal distribution fitting are applied in analyzing the experimental results and testified by chi-square distribution fit test and correlation coefficient for comparison.