Novel methodology has been developed that simultaneously improves sensitivity and specificity of a low-resolution ion mobility spectroscopy (IMS) sensor. Wavelet transforms have been applied to IMS spectra in order to de-noise and enhance spectral features. Next, trigger metrics of the spectra were derived using a statistical evaluator (SE) and optimized using a genetic algorithm (GA). The combination of wavelets, SE, and GA has been demonstrated to differentiate between background, analyte, interferent, and a binary mixture of analyte and interferent. This results in an overall increase in resolving power. The new system is less sensitive to false positives due to increased selectivity, shows the ability to yield quantitative data at ultra-low concentrations for low level toxicity, has the ability to detect binary mixtures of compounds, and shows great potential in significantly improving chemical warfare detection capabilities under field conditions.