The frequent and direct remote sensing of the tropospheric ozone profile is a critical environmental measurement to be performed on a global scale by future satellite instruments. An approach has been developed for the retrieval of tropospheric ozone profiles from 9.6 ¿m limb-viewing clear-sky radiances generated for the spectral resolution and signal-to-noise of the Tropospheric Emission Spectrometer (TES). TES is a high-resolution Fourier transform spectrometer under development for NASA's Earth Observing System Aura platform (http://eos-chem.gsfc.nasa.gov/). The simulated radiance spectra are calculated from northern hemisphere midlatitude lidar profile measurements. The ozone profile retrieval and the associated errors are obtained as a function of atmospheric pressure level using the method of nonlinear least squares with regularization. In order to accelerate convergence a two-stage strategy has been applied in which the full profile retrieval has been preceded by a shape retrieval involving a smaller set of near independent parameters. Our analysis indicates that the O3 profile can be retrieved from the TES limb measurements with a relative uncertainty of 5% (1σ) in the middle and upper troposphere.