使用对偶分解的MU-CoMP-JT联合资源分配

Joint Resource Allocation Using Dual Decomposition for MU-CoMP-JT

在MU-Co MP-JT(Multi-User Coordinated Multiple-Points Joint Transmission)联合资源分配问题中,传统的迫零预编码矩阵会使得每根天线发送功率互不相同,当Co MP节点发射功率仅满足总功率约束时性能损失不明显,而当Co MP节点分布在不同的地理位置时将受到单节点功率约束,这势必会降低系统功率利用率。为了进一步提升系统吞吐量,基于对偶分解理论提出了一种联合预编码优化的资源分配算法。该算法以最大化用户权重速率为目标,将原优化问题分解成若干个优化的子问题,不同子问题对应不同接收天线数的联合优化问题。当子信道的发送天线数大于接收天线数时,通过多次迭代计算得到预编码矩阵,并且预编码矩阵会随着拉格朗日因子的变化而变化。仿真结果表明所提联合预编码优化的联合资源分配算法能够明显提升系统吞吐量,且提高天线功率利用效率。

In the joint resource allocation problem of multi-user coordinated multiple-points joint transmis-sion（MU-CoMP-JT）,the traditional zero-forcing precoding matrix makes the transmitting power of eachantenna different,the loss of performance is not obvious when the transmitting power of the CoMP node onlymeets the total power constraint,but it will be influenced by single node power constraint when the CoMPnode distributes in different geographical position,which is bound to reduce the power utilization of the sys-tem. In order to further improve the system throughput,a joint precoding optimized resource allocation algo-rithm is proposed based on dual decomposition theory. The algorithm is aimed at maximum user weight rateand decomposes the original optimization problem into multiple optimization sub-problems,different subproblemscorrespond to the joint optimization problem of the different number of receiving antennas. Whenthe number of transmit antennas is greater than that of receiving antennas,the precoding matrix can be ob-tained by several times of iterative computation and it will change with Lagrange factor. The simulation re-sults show that the proposed combined precoding optimization algorithm can improve the system throughputsignificantly and utilization efficiency of antenna power.