The cloud microphysical properties such as cloud effective radius and cloud water path are fundamental properties for understanding the cloud formation, radiative impacts and interactions with aerosol and precipitation. The downwelling infrared radiance spectra is studied here to retrieve microphysical properties of clouds. The sensitivity of the downwelling radiance spectra and cloud emissivity spectra to the liquid cloud and ice cloud effective radius and optical depth is analyzed. The look-up-tables are established for optically thin clouds (cloud optical depth less than 6) that rely on parameters of the slopes and differences of the emissivity spectra. These parameters include the difference in the emissivity between 862.1 and 934.9 cm-1, the difference in the emissivity between 1 900.1 and 2 170.1 cm-1, the slope of the cloud emissivity and the radiation between 900 and 1 000 cm-1, the slope of the cloud emissivity and the radiation between 1 100 and 1 200 cm-1. The look-up-tables are constrained by the incorporation of mean ozone band transmissivity between 1 050 and 1 060 cm-1. Cloud effective radius and optical depth can be obtained with by least squares fitting between observed and modeled above-mentioned multiple spectral parameters. The cloud water path can then be derived from the experiential relationship. The inversion results are compared with the ARM baseline cloud microphysics product (MICROBASE). It is shown that, the cloud effective radius is roughly in the same order of magnitude while the water paths derived from both method are of large differences especially for the liquid cloud path. The algorithm proposed in this paper is efficient for retrieving microphysical properties of thin clouds with cloud optical depth less than 6.