In many channel measurement applications, one needs to estimate some characteristics of the channels based on a limited set of measurements. This is mainly due to the highly time varying characteristics of the channel. In this contribution, it will be shown how free probability can be used for channel capacity estimation in MIMO systems. Free probability has already been shown to be an invaluable tool for describing the asymptotic behaviour of many systems when the dimensions of the system get large (i.e. the number of antennas). In particular, introducing the notion of free deconvolution, we provide hereafter an asymptotically (in the number of antennas) unbiased capacity estimator (w.r.t. the number of observations) for MIMO channels impaired with noise. Another unbiased estimator (for any number of observations) is also constructed by slightly modifying the free probability based estimator. The framework is then extended to MIMO channels with phase off-set and frequency drift for which no estimator has been provided so far.