Channel sparse representation based authentication for reconciliation of key generation
Ning Wang, Ting Chen Jiang, Zheng Zhou · 2016
We propose a novel authentication method based on sparse representation for the reconciliation in physical layer key Generation. In order to protect the legitimation of the channel, sparse representation is considered to enhance wireless security by using the unique channel location information. In contrast to traditional approaches, in this paper, we consider that this reconciliation channel is unauthentic, and we formulate this problem as one of improved sparse representation. In this scheme, the position of the primary decomposition coefficients is adopted to ascertain whether the signal comes from legitimate devices or not. The effectiveness of this new process model is validated through experiments and simulations. These results shown that the proposed scheme can achieve desired identification performance even at lower signal-to-noise ratio (SNR).