摘要
传统的Schmidt方法对椭圆球面波函数(PSWF)脉冲组进行正交化时,脉冲组的最大互相关值受脉冲个数的影响,同时还存在计算复杂度高的问题。针对上述问题,提出了基于Householder变换的正交化方法,通过将PSWF脉冲离散化,再依次对各离散的PSWF脉冲进行Householder变换,实现了对PSWF脉冲组进行QR分解,并得到新的正交PSWF脉冲组。仿真结果表明,与传统Schmidt方法相比,基于Householder变换的正交化方法使得脉冲组的最大互相关值从10-4降低到10-11,改善了脉冲组的正交性能,突破了脉冲组在脉冲个数的限制,并降低了计算复杂度,同时能够保持PSWF脉冲带内能量集中度高的优势。
In using traditional Schmidt method in the orthogonalization of the Prolate Spheroidal Wave Function （PSWF） pulses set, the maximum cross-correlation value of the pulses set is under the influence of pulse number, and this method has a high computational complexity. To solve the problems above, an orthogonalization method based on Householder transformation is presented. The QR decomposition is real-ized through discretizing the pulses set and Householder transformation for the discrete pulse, then the new PSWF orthogonalization pulses set is obtained. Simulation results show that comparing with Schmidt meth-od, the orthogonalization method based on Householder transformation makes the max correlation value re-duced from 10^-4 to 10^-11 and is not restricted to the pulse number, reduces the computational complexity, while maintaining the advantage of high energy concentration.
出处
《电讯技术》
北大核心
2013年第11期1452-1458,共7页
Telecommunication Engineering
基金
国家自然科学基金资助项目(60772056)
山东省"泰山学者"建设工程项目
海军航空工程学院研究生创新基金项目~~
关键词
非正弦波通信
正交脉冲
椭圆球面波函数
Householder变换
nonsinusoidal wave communication
orthogonal pulse
prolate spheroidal wave function（PSWF）
Householder transformation
