Weak Signal Processing And Ion Mobilty Nonliner Fuction Coefficients Error Analysis In FAIKS

High-field asymmetric waveform ion mobility spectrometry (FAIMS) is a novel analytical technique for trace chemical detection. Performed under ambient conditions, this technique has shown great potential in the detection of explosives, pollutions, CWAs.There are broad application prospects in related field. FAIMS utilize the characteristic, which is nonlinear variation of ion mobility about different ions in the strong electric field, to achieve the separation and identification of ions. FAIMS sensitivity is high, core device is little and detecting time is short. To be a field rapid testing technique, sensitivity and resolution is very significant. Thus, how to improve further detection sensitivity and resolution are two valuable problems to be studied.There are many factors affect FAIMS detection sensitivity, such as the parameter of the mobility tube, efficiency of ionization source and identification method about spectrum signal characteristic. In case that structure of FAIMS is definitely, it is necessary to study the FAIMS spectrum signal itself By signal processing, we could accurately obtain the peak position of spectrum signal, meanwhile, it is could to improve the detection sensitivity of FAIMS. Currently, it is lack of research on FAIMS spectrum signal processing, which affects that FAIMS further enhance the detection sensitivity in a given structure.As we know, both extraction accuracy of spectrum characteristic and accuracy of α2and a4influencegreatly the resolution of FAIMS. On the one hand, when the signal is relatively weak, spectrum feature is interfered seriously by noise. Thus, it is necessary to improve the accuracy of fetching spectrum characteristic by filtering noise. On the other hand, existed α2and α4is not only lack of priors, but also short in error analysis method, so it is unable to verify the accuracy of corresponding solution. Therefore, it is necessary to establish error assessment criteria of a2and α4solution, on the basis, it is conductive to improve solution accuracy of α2and α4.For these two problems, the major work of this article includes the following two aspects:Firstly, according to feature of FAIMS spectrum, there is a variety of methods about FAIMS spectrum signal processing in this paper. Methods, like numerical average, Fourier transform and wavelet transform, are used to process FAIMS spectra signal, among them, wavelet transform perform best in removing noise and maintaining signal integrity. They could improve signal to noise ratio, increase the accuracy of acquired spectrum characteristics and availability of spectrum, therefore, enhance sensitivity and resolution of FAIMS detection.Secondly, considering that α2and α4is lack of priori and assessment methods, this paper established evaluation criteria of solution accuracy, on the ground, completed optimization of FAIMS experimental parameters. This article explored distribution characteristic and correlation of a2and α4, researched influence of number and method to fetch dispersion voltage detected points for error of solution. After fitting frequency of α2and a4, it was found that they conform to normal distribution, thus standard deviations of their distribution are able to evaluate error of solution. In addition, there is a strong negative correlation between them. With increasing of computing selected points, the corresponding error of solving results decrease. By comparing the standard deviation of method to fetch dispersion voltage detected points, found that solving effect was optimized. It is conducive for rapid detection and spectral analysis.