The Effect of Eavesdropper's Statistics in Experimental Wireless Secret-Key Generation
read the original abstract
This paper investigates the role of the eavesdropper's statistics in the implementation of a practical secret-key generation system. We carefully conduct the information-theoretic analysis of a secret-key generation system from wireless channel gains measured with software-defined radios. In particular, we show that it is inaccurate to assume that the eavesdropper gets no information because of decorrelation with distance. We also provide a bound for the achievable secret-key rate in the finite key-length regime that takes into account the presence of correlated eavesdropper's observations. We evaluate this bound with our experimental gain measurements to show that operating with a finite number of samples incurs a loss in secret-key rate on the order of 20%.
This paper has not been read by Pith yet.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.