摘要
作为智能天线的关键技术之一,波束赋形算法引起了众多学者的广泛关注。为克服目前RVSSLMS算法比较简便,易于实现,但收敛速度较慢;RLS算法收敛速度较快,但其运算量大的问题,根据移动通信系统中波束赋形算法必须具有较快的响应速度和收敛速度的要求,对RVSSLMS算法进行了改进;结合RLS算法和RVSSLMS算法的优点,在开始迭代前的25次用RLS算法求加权系数W(k),再使RVSSLMS算法用RLS算法求出的加权系数W(k)作为初始值进行迭代求解,使其在保持原有运算量小的特点的同时,具有更快的收敛速度。用Matlab仿真对改进方法的有效性进行了验证,仿真结果表明:RLS-RVSSLMS算法既具有RLS算法收敛速度快的特点,同时保持了RVSSLMS算法计算量小的特点。
As the key technique of the smart antenna, the beam - forming algorithm has enjoyed widespread populaxity. In order to overcome the problem that the convergence speed of the RVSSLMS algorithm is slower though the realization of it is simpler and easier, meanwhile, the RLS algorithm's convergence speed is faster while its computation load is bigger, according to the condition that the beam - forming algorithm must have fast response speed and convergence speed in mobile communication' system, the RVSSLMS algorithm is improved by combining the advantages of RVSSLMS algorithm and RLS algorithm together: the RVSSLMS algorithm's initial iteration weight is from the RLS algorithm's weight whose length of data block is 25. The improved one not only keeps less computation load, but also attains faster convergence speed. The efficiency of the improved method is verified by Matlab simulation, the result indicates that the convergence speed of this RLS - RVSSLMS algorithm is as fast as that of RLS algorithm and the operation load of it is as small as that of the RVSSLMS algorithm.
出处
《空军工程大学学报(自然科学版)》
CSCD
北大核心
2008年第3期42-45,共4页
Journal of Air Force Engineering University(Natural Science Edition)
基金
陕西省自然科学基金资助项目(2005F23)
空军工程大学导弹学院研究生学位论文创新基金(DY06202)