非理想信道下的分布式认知多小区波束形成

摘要

针对非理想信道下的波束形成问题，将传统稳健波束形成设计推广至认知多小区，在认知干扰和认知用户速率约束下，构造基于最小化认知系统总功率准则的优化问题。通过半定松弛及 S-Procedure 算法将其转化为凸优化问题，并采用 Primal 分解将该凸优化问题分解为一组独立的子问题，从而实现了分布式求解，在降低复杂度的同时减少所需的反馈信息。仿真结果表明：算法不仅对信道误差稳健，而且收敛速度很快。%To solve the beam forming problem for imperfect channel,the robust beam forming design method for conven-tional system is extended to cognitive multi-cell system.Under the constraints of cognitive interference and cognitive us-ers′rate,the optimization problem is formulated based on the criterion of minimization of cognitive system′s total power. The problem is converted to a convex optimization problem by means of semi-definite relaxation and S-Procedure algo-rithm,and is decomposed into a group of independent sub-problems by using Primal decomposition.So then a distributed solution is achieved and both computational complexity and required feedback information are reduced.Simulation results show that the algorithm is not only robust to channel errors but also convergences quickly.