摘要
针对传统单参数优化法设计的原型低通滤波器存在设计复杂,性能不理想的问题,提出了一种利用窗函数截取平方根升余弦函数获取滤波器系数的设计方法。该方法将具有功率互补特性的平方根升余弦函数作为原型函数,通过迭代算法来确定滤波器长度,进而利用窗函数设计原型低通滤波器。仿真结果表明,该方法设计的原型滤波器的阻带衰减达到了-116dB,滤波器组的重构误差在5×10^-4dB以内,性能优于传统单参数法设计的滤波器组,具有良好的重构特性。
In view of the critical shortcomings of the filter bank with traditional single parameter optimization,such as design complexity and unsatisfactory performance,a new approach that uses window function to intercept square root raised cosine function to obtain filter coefficients is proposed to design a bank of prototype filters.The square-root raised cosine function with power complementary property is used as a prototype function and the length of filter is determined by iteration.The simulation results show that the prototype filter designed by windowing has a stopband attenuation of-116 dB,and the reconstruction error of the filter bank is within 5×10^-4 dB.The performance is better than the traditional single-parameter method.
作者
张超
马宏
焦义文
ZHANG Chao;MA Hong;JIAO Yiwen(Space Engineering University of PLA ,Beijing 101416,China)
出处
《雷达科学与技术》
北大核心
2019年第3期335-338,344,共5页
Radar Science and Technology
关键词
单参数优化法
窗函数
平方根升余弦函数
功率互补特性
重构误差
single-parameter optimization
window function
square root raised cosine function
power complementary characteristic
reconstruction error
