Jan. 28, 2013 to Jan. 30, 2013
S. Shimosaka , Tokyo Univ. of Agric. & Technol., Koganei, Japan
K. Umebayashi , Tokyo Univ. of Agric. & Technol., Koganei, Japan
Y. Suzuki , Tokyo Univ. of Agric. & Technol., Koganei, Japan
In this paper, a secure multiple-input multiple-output (MIMO) communication technique based on the singular value decomposition (SVD) is investigated. In a previous work, it has been shown that a secure MIMO communication based on the SVD, where interference signal and confidential signal are assigned to two different streams, can achieve positive secrecy rate. In this investigation, it was assumed that the stream assignment information is available at both the legitimate receiver and the eavesdropper. In fact, random stream assignment has a potential to increase the secrecy rate. Specifically, pilot signal is inserted into the pilot time frame and this is used to inform the stream assignment. If either the legitimate receiver or the eavesdropper decodes the pilot signal incorrectly, the receiver attempts to decode the interference signal in the following data time frame even if there is no information about the confidential signal. Therefore, we propose a signaling technique to expand the difference in mutual information between the legitimate receiver and the eavesdropper which leads to enhanced secrecy. In the proposed signaling technique, correlation between pilot signals in successive pilot time frames is provided. Due to this correlation, correct pilot signal decoding in the current pilot time frame requires the correct pilot signal decoding in the previous pilot time frame. Numerical result shows that the propose method can significantly decrease the mutual information between the transmitter and the eavesdropper which leads to increased the secrecy rate.
Mutual information, Receivers, Security, Interference, Vectors, MIMO, Matrix decomposition,
S. Shimosaka, K. Umebayashi, Y. Suzuki, "A study on stream assignment method for secure MIMO communication", ICOIN, 2013, The International Conference on Information Networking 2014 (ICOIN2014), The International Conference on Information Networking 2014 (ICOIN2014) 2013, pp. 12-17, doi:10.1109/ICOIN.2013.6496344