In this dissertation, we present a study based on the AzTEC/Large Millimeter Telescope (LMT) survey of dust continuum at 1.1mm on the central 200 parsecs (The Central Molecular Zone (CMZ)) of our Galaxy. Owing to its unusually high gas density and turbulence, strong magnetic field, and high cosmic ray flux, the CMZ represents an initial condition for star-formation typical of starburst galaxies in the distant universe. In order to understand dust properties in such an extreme environment. We perform a joint SED analysis of existing dust continuum surveys on the CMZ, from a wavelength of 160 μm to 1.1 mm. This analysis follows a Bayesian model incorporating the knowledge of Point Spread Functions (PSFs) in different maps, which enables full utilization of our high resolution (10.5”) map at 1.1 mm and achievement of unprecedented detailed information on the spatial distribution of dusty gas across the CMZ. There is a remarkable trend of increasing dust spectral index, from 2.0 − 2.5, toward dense peaks in the CMZ, indicating a deficiency of large grains or a fundamental change in dust optical properties. The latter scenario leads to an underestimate of dust temperature when using the conventional model. Depending on how the optical properties of dust deviate from the conventional model, dust temperature could be underestimated by 10−50%, and potentially even higher. We further develop new methods to explore the temperature and density structures of the CMZ molecular clouds, based on Hierarchical Bayesian Analysis. We propose a phenomenological model for line-of-sight temperature decomposition and show that the temperature profile of dust evolves with orbital phases, in agreement with previous studies on gas temperature. Finally, we show that at the 0.5-parsec spatial resolution achieved by our study, the Probability Density Function of the Column Densities (N-PDF) provides a robust indicator of the density structure.