Porous suspended particles are hazardous to human health due to their strong absorption capacity for toxic substances. A fast, accurate, in situ and high-throughput method to characterize the microporous structure of porous particles has extensive application value. The polarization changes during the light scattering of aerosol particles are highly sensitive to their microstructural properties, such as pore size and porosity. In this study, we propose an overlapping sphere model based on the discrete dipole approximation (DDA) to calculate the polarization scattering characteristics of porous particles. By combining scattering calculations with multi-dimensional polarization indexes measured by a multi-angle polarized scattering vector detection system, we achieve the identification and classification of pore-type components in suspended particles. The maximum deviation based on multiple indexes is less than 0.16% for the proportion analysis of mixed particles. Simultaneously, we develop a quantitative inversion algorithm on pore size and porosity. The inversion results of the three porous polymer particles support the validity and feasibility of our method, where the inversion error of partial particles is less than 4% for pore size and less than 6% for porosity. The study demonstrates the potential of polarization measurements and index systems applied in characterizing the micropore structure of suspended particles.