Reasearch On Precoding Techniques For Physical Layer Security

The rapid development and popularity of mobile communication technology has brought a huge change to the people’s daily life. At the same time, the security issues of wireless communcaition are being more and more concerned. The openness of wireless channel, mobility of terminal, diversity of network topology and instability of wireless transmission make people face more security threats. The security of date transmission has been traditionally trusted to key-based cryptographic at the network layer, assuming that physical layer can provide error-free link. However, with appearance of some network topologies with low cost and low complexity, such as Adhoc network, distributed network, the traditional cryptography is unavailable. Therefore, to exploit new technology to gurantee the secure communication between transmitter and the legitimate receiver is inevitable. The physical layer security technique without relaying on the cryptography, based on the security model of Shannon, which utilizes the characteristic of wireless channels to gurantee secure message transmission with proper encoding, modulation and transmission scheme, has attacted significant attention. The rapid developlemt and innovation of the transmission techniques in physical layer enrich the available transmission scheme; also extend the research domain for physical layer security. The precoding technology in multiple input multiple output (MIMO) systems and cooperation relay systems have been investigated widely. The subject of dissertation is to investigate the precoding technique for physical layer security, analyse the secrecy performance, and design the corresponding precoding strategies to enhance the secrecy effectively.The existing procoding techniques for physical layer security are firstly reviewed and discussed in this dissertation. The precoding techniques for physical layer security can be summaried as two catagories: the precoding techniques in the centralized networks and the precoding techniques in the distributed networks. For the distributed networks, according to the protocol of collaborate, the precoding techniques in the distributed networks can be summaried as two sub-categories: cooperative relaying (CR) and cooperative jamming (CJ). The precoding techniques in these two categories are all investigated in this dissertation.The second part of this dissertation investigates the precoding techniques in the centralized networks. Firstly, for the multiple input single output wiretap channel, based on the delayed channel direction information (CDI) between the transmitter and the legitimate receiver, the joint precoding vector and AN vector design are proposed. Under this design, the expressions of security capacity, security capacity loss and power allocation with parameter delay are analysed. Secondly, for the single input multiple output wirtap channel, the joint receive antenna selection and AN generation strategy is proposed based on the full duplex assumption at the legitimate receiver. Based on this strategy, the secrecy performance, include that the nonzero secrecy capacity probability, the secrecy outage probability, the average secrecy capacity and the symbol error ratio (SER) at the legitimate receiver and eavesdropper are analysed. Furthermore,aiming to improve the secrecy performance, three different power allocation strategy between desired signal and AN are proposed, include that the criterion of secrecy outage probability minimization and that of secrecy capacity maximization.The third part of this dissertation investigated the precoding technique in the distributed networks. For the one-way relay wiretap channel, the secrecy performance of relay network with different forward protocol, cooperative protocol and power allocation strategy is investigated. Firstly, when the relay node employs amplified and forward (AF) protocol, different relay selection strategies based on the global CSI and the partial CSI knowledge are proposed. The secrecy performance of relay selection strategies under the partial CSI knowledge is analysed. Secondly, when the relay node employs decode and forward (DF) protocol, a joint relay node selection and cooperative jamming strategy is proposed. The mathematical model of the minimum secrecy rate in two slots maximization with power constraint, to optimize relay selection, power allocation and CJ vector, is built. Thirdly, the similar model for relay employs cooperative relaying (CR) is proposed, and an altenating iteration algorithm is proposed to jointly optimize the CR vector and power allocation. At last, an adapative strategy is proposed to make full use of advantage of both CJ and CR strategies.Different from the third part, the fourth part concentrates on the two-way relay wiretap channel. The cooperative relaying (CR) strategy is proposed to enhance the secrecy of two-way relay channel. Firstly, the mathematical model to describe the achievable rate region, which is based on the weighted sum rate maximization with the rate constraint at eavesdropper, is built. Two kinds of beamforming strategies, the optimal beamforming strategy and null space beamforming strategy, are proposed. An alternating iteration method is proposed to calculate the relay beamforming vector as well as sources’ transmit power. Secondly, for null space beamforming strategy, three schemes are proposed, include that the power consumption minimization with the secrecy rate constraint at each source, the sum secrecy rate maximization with power constraint, and the minimum secrecy rate maximization with power constraint. Similarly, the altermating iteration method is porposed to solve them effectively. Under the power minimization problem, a fast algorithm based on the sequential quadratic programming (SQP) to slove the beamforming vector problem is proposed, and its computional complexity is analysed.