Based on our prior research, this paper carries out the simulation of three typical short-wavelength radar detections (X-band-9.5GHz-3cm, Ka-band-35GHz-8mm, and W-band-94GHz-3mm) for non-precipitation cloud particles, particularly ice crystals. It further compares and analyzes the radar reflectivities stemming from variations in the particles" physical attributes. Additionally, the study explores the feasibility of employing these three radar wavelengths to discern the morphology of ice crystals. The results show that, despite the W-band"s short wavelength and heightened sensitivity, the backscattering characteristics of ice crystals exhibit a considerably more intricate pattern than those observed in the X-band and Ka-band. Variations in incident angle, particle morphology, and particle size elicit dramatic and complex alterations in backscattering, posing a formidable challenge to traditional inversion methods that rely on statistical or empirical fitting approaches to extract cloud particle microphysical properties. It has also been shown that utilizing the characteristics of the differences between multi-wavelength radar observations can help to provide microphysical information about particles in clouds. For example, in the ice cloud, a pronounced disparity in reflectivity factor between the W-band and either the Ka-band or X-band, particularly when it dips below -3 dBZ, suggests the likelihood of prolate particle presence. This offers invaluable insights and practical experience for refining and enhancing the accuracy of algorithms designed to invert cloud microphysical features based on triple-frequency radar data.