Transmit design for MIMO wiretap channel with a malicious jammer

Duo Zhang,Weidong Mei,Lingxiang Li,Zhi Chen
DOI: https://doi.org/10.1109/VTCSpring.2015.7146141
2016-11-29
Abstract:In this paper, we consider the transmit design for multi-input multi-output (MIMO) wiretap channel including a malicious jammer. We first transform the system model into the traditional three-node wiretap channel by whitening the interference at the legitimate user. Additionally, the eavesdropper channel state information (ECSI) may be fully or statistically known, even unknown to the transmitter. Hence, some strategies are proposed in terms of different levels of ECSI available to the transmitter in our paper. For the case of unknown ECSI, a target rate for the legitimate user is first specified. And then an inverse water-filling algorithm is put forward to find the optimal power allocation for each information symbol, with a stepwise search being used to adjust the spatial dimension allocated to artificial noise (AN) such that the target rate is achievable. As for the case of statistical ECSI, several simulated channels are randomly generated according to the distribution of ECSI. We show that the ergodic secrecy capacity can be approximated as the average secrecy capacity of these simulated channels. Through maximizing this average secrecy capacity, we can obtain a feasible power and spatial dimension allocation scheme by using one dimension search. Finally, numerical results reveal the effectiveness and computational efficiency of our algorithms.
Information Theory
What problem does this paper attempt to address?
This paper attempts to solve the transmission design problem in multi - input multi - output (MIMO) wiretap channels when there are malicious jammers. Specifically, the paper focuses on how to improve the communication security and reliability of legitimate users by optimizing the transmission strategy in the case where the eavesdropper channel state information (ECSI) at different levels is available. The paper considers ECSI in three cases: 1. **Unknown ECSI**: When the eavesdropper's channel state information is completely unknown, the paper proposes a new optimization algorithm. First, a target data rate of a legitimate user is set, and then the optimal power allocation scheme is determined by the water - filling algorithm in reverse, and the spatial dimension allocated to artificial noise (AN) is adjusted by the step - by - step search method to achieve the target data rate. 2. **Statistical ECSI**: When only the statistical characteristics of the eavesdropper's channel state information are known, the paper uses the stochastic simulation method to generate multiple simulated channels distributed according to ECSI, approximately calculates the average secrecy capacity of these simulated channels, and determines the optimal power and spatial dimension allocation scheme by maximizing this average secrecy capacity. 3. **Complete ECSI**: When the eavesdropper's channel state information is completely known, the paper uses the alternating optimization (AO) algorithm or the generalized singular value decomposition (GSVD) algorithm to solve the transmission design problem of maximizing the secrecy rate. Through these different strategies, the paper aims to improve the security performance of MIMO wiretap channels in the presence of malicious jammers, and especially proposes effective solutions in the case of incomplete ECSI information.