Shlomo Shamai, Department of Electrical Engineering, Technion - Israel Institute of Technology
Abstract: The fundamental relationship between information theory and estimation theory in the Gaussian channel, is used for an alternative derivation of the secrecy capacity of the multiple-input multiple-output (MIMO) Gaussian wiretap channel, under a power-covariance constraint.
This approach yields closed form expressions in contrast to non-convex optimization based formulas provided by the standard different approaches (Khisti-Wornell, Oggier-Hassibi, Liu-Shamai). Furthermore, the input covariance matrix required in order to attain the capacity is specified, and the current technique provides also some insight into the concept of enhancement, and alternative expressions applied to this problem by Liu-Shamai. An alternative expression for the secrecy capacity of the MIMO Gaussian wiretap channel by Liu-Liu-Poor-Shamai is also examined via the information-estimation perspective. A short outlook will conclude the talk.
The talk is based on a joint work with R. Bustin, Technion, R. Liu and V. Poor, Princeton University.
Biography: