Research Of Distributed Beamforming In The Cognitive Redio Networks With Relays

With the developing of communication technology and the increasing demanding of wireless data, spectrum resource becomes scarce. Cognitive radio is a technology that can realize the dynamic use of spectrum. It allows the secondary users to access the spectrum which is allocated to the primary users by multiple sharing methods. That is to say cognitive radio is good to solving the unbalancing using of the spectrum and improves the efficiency of spectrum. Relay is another technology which can be used to improve the distance of communication and reduce channel fading. On the other hand, distributed beamforming is a technology that can combine with relay. In distributed beamforming, multiple nodes are used to form a virtual MIMO system and the direction of the beam is controlled. By using this virtual MIMO system, diversity gain can be obtained. In addition, the estimation of channel can not be exactly accurate in practice. Thus, the research on distributed beamforming in the cognitive radio networks with relays is of great importance and it is of great significance to study beamforming while considering the errors of channel estimation at the same time. The main works of this paper includes:Firstly, beamforming in cognitive radio networks with two-way relay is considered. The sum-rate maximization problem subject to the interference power to the primary user constraint and the total power of the relays constraint is studied. The research aims to design the beamforming vector while considering the corresponding criterion and is based mainly on the convex optimization theory. Two different beamforming approaches are introduced. The first approach is based on the rate region. The maximum sum rate is obtained by charactering the rate region. The second approach is based on the first order Taylor polynomial. By analyzing the structure of the optimization problem, we found that the first order Taylor polynomial can be used to change the original non-convex problem to a solvable one. And an iterative algorithm is then introduced to solve the problem accurately. Simulation results show that the sum-rate obtained by these two methods are coincident. Secondly, beamforming in the cognitive radio with multiple transmitter-receiver peers is considered. Beamforming vector is designed to minimize the total transmit power of the relays while considering imperfect channel state information. Convex optimization theory and theory of probability are resorted to in this work. The original optimization problem with constraints in form of probability is firstly transformed to a normal form without probabilistic constraints. And a method based on Bernstein-type inequality and semidefine program is finally adopted to solve the problem. Simulation results indicate that minimum transmit power of the relays obtained by the beamforming method which considers the channel estimation error draws near to that obtained by the beamfroming method in perfect channel state information.