能量有效的认知多小区协同波束赋形算法

摘要

针对认知多小区多用户下行传输链路，提出了一种基于能量效率最大化准则的协同波束赋形优化方法。该方法采用迫零消除小区内用户间干扰，在保证用户最小速率需求及认知干扰约束的同时，实现了能量效率和频谱效率的同步改善。为了分布式求解优化问题，通过约束泄露干扰并利用半定松弛，将其转换为凸问题，在此基础上，采用部分对偶分解方法将多小区联合优化问题分解为一组单小区优化问题，从而实现了分布式求解。仿真结果表明，该方法不仅实现了能量效率和频谱效率的有效折中，而且达到了集中式算法的性能。%A cooperative beamforming algorithm based on the maximize of energy efficiency( EE) was proposed for the cognitive multi-cell multiuser downlinks.By using zero-forcing algorithm to eliminate intra-cell interference, the pro-posed scheme makes a simultaneous improvement in energy efficiency and spectrum efficiency ( SE ) while guaranteeing minimum rate for the secondary users and the cognitive interference constraints.To implement a decentralized algorithm,the original problem was firstly transformed to a convex one via semi-definite relaxation and the leakage interference con-straints,and then was decomposed into a group of sub-problems on the basis of each cell by using dual decomposition,which admits a distributed computation of the problem.Simulation results show that the proposed scheme not only makes a good tradeoff between the EE and SE,but also attains the same performance as the centralized algorithm.