On the Impact of Quantized Channel Feedback in Guaranteeing Secrecy with Artificial Noise: The Noise Leakage Problem
Authors:
- 20 January 2011
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 10 (3) , 901-915
Abstract
The impact of quantized channel direction information (CDI) on the achievable secrecy rate is studied for multiple antenna wiretap channels. By assuming that the eavesdropper's channel is unknown at the transmitter, we adopt the transmission scheme where artificial noise (AN) is imposed in the null space of the legitimate receiver's channel to disrupt the eavesdropper's reception. It has been shown that, in the ideal case where perfect CDI is available at the transmitter, the achievable secrecy rate can be made arbitrarily large by increasing the transmission power. However, when only quantized CDI is available, the AN that was originally intended to jam the eavesdropper may now leak into the legitimate receiver's channel, causing significant secrecy rate loss. For a given number of feedback bits B and transmission power P, we derive the optimal power allocation among the message-bearing signal and the AN to maximize the secrecy rate under AN leakage. We show that, when B is sufficiently large, one should allocate power evenly among the message-bearing signal and the AN; whereas when B is small, one should be more conservative in allocating power to the AN. Moreover, by showing that the achievable secrecy rate under quantized CDI is bounded by a constant, we derive a scaling law between B and P that is necessary to maintain a constant secrecy rate loss compared to the perfect CDI case. The scaling of B is shown to be logarithmic of P. These results are first derived for the multiple-input single-output single-antenna-eavesdropper scenario and are later extended to the multiple-input multiple-output multiple-antenna-eavesdropper case. Numerical simulations are provided to verify our theoretical claims.Keywords
This publication has 17 references indexed in Scilit:
- Secure Transmission With Multiple Antennas I: The MISOME Wiretap ChannelIEEE Transactions on Information Theory, 2010
- On the Impact of Quantized Channel Direction Feedback in Multiple-Antenna Wiretap Channels2010
- On the impact of quantized channel feedback in guaranteeing secrecy with artificial noise2009
- Secrecy Capacity Region of a Multiple-Antenna Gaussian Broadcast Channel With Confidential MessagesIEEE Transactions on Information Theory, 2009
- Secure Communication Over Fading ChannelsIEEE Transactions on Information Theory, 2008
- Quantization Bounds on Grassmann Manifolds and Applications to MIMO CommunicationsIEEE Transactions on Information Theory, 2008
- Isotropic Fading Vector Broadcast Channels: The Scalar Upper Bound and Loss in Degrees of FreedomIEEE Transactions on Information Theory, 2005
- Random Matrix Theory and Wireless CommunicationsFoundations and Trends® in Communications and Information Theory, 2004
- On beamforming with finite rate feedback in multiple-antenna systemsIEEE Transactions on Information Theory, 2003
- Elements of Information Theory2001