双向中继网络的波束成形设计大都基于对称信道的假设。该文考虑非对称信道的情况,通过合理设计波束成形因子,使得系统在中继总功率约束下,其和速率最大化。为解决此非凸的最优化问题,首先提出了一种基于分支定界(branch-and-bound)思想的优化算法,并证明了该算法可获得全局最优解。然后提出了一种次优算法,该算法假设次优解为两个已知向量的线性组合,因而求解次优解只需优化两个实数系数,具有较低的复杂度。仿真表明,分支定界算法性能优于现有其他算法;次优算法相对于分支定界算法有一定的性能损失,但是在节点数较小或中继节点平均功率较低时,次优算法优于其他现有算法。
Most of the study on the beamforming design for Two-Way Relaying Network(TWRN) is based on the reciprocity assumption.Assuming channel is nonreciprocal,this paper designs the beamforming coefficients to maximize the sum-rate of the network subject to the Total Relay Power Constraint(TRPC).To solve this nonconvex problem,a branch-and-bound based algorithm is firstly proposed and the global optimum is proved to be obtained.Secondly,a suboptimal algorithm is proposed.This algorithm assumes the suboptimal solution is the linear combination of two known vector,thus only two real variables need to be optimized,which requires low complexity.Simulation results show that the branch-and-bound algorithm performs better than other existing algorithm.The suboptimal algorithm suffers performance loss compared to the optimal solution,but outperforms other algorithm when the number of relay nodes is relatively small or the average power of relay is relatively small.