Retrievals of sea ice thickness from passive microwave measurements have been limited to thin ice because microwaves penetrate at most the upper 50 cm of sea ice. To overcome such a limitation, a method of retrieving Arctic basin-scale ice thickness is developed. The physical background of this method is that the scattering optical thickness at microwave frequencies within the freeboard layer is linearly proportional to the physical thickness of the ice freeboard. In this study, we relate the optical thickness estimated from the Advanced Microwave Scanning Radiometer 2 (AMSR2) with ice freeboard estimated from the CryoSat-2 (CS2) by employing a piecewise linear fit. The results show a strong linear relationship between the AMSR2-estimated and CS2-measured ice freeboards with a correlation coefficient of 0.85 and bias and RMSE of 0.0001 and 0.04 m, respectively; this evidence suggests that the method can provide Arctic basin-scale ice freeboard with a comparable accuracy level of CS2. The method is also applied to estimate ice freeboard for the periods of the Scanning Multichannel Microwave Radiometer (SMMR) (1978-1987) and AMSR-E (2002-2011). It is shown that the area-averaged ice freeboard has decreased significantly with the linear trends of 1.5 cm/decade. In addition, there seems to be a change of ice freeboard distributions over the Arctic. Furthermore, the algorithm is extended to the ice thickness retrieval by using the hydrostatic balance equation, showing that operational basin-scale ice thickness retrieval will be possible from satellite passive microwave measurements if a realistic snow depth on sea ice is employed.