李安宇,
石博雅
1.天津工业大学电子与信息学院 ??天津 ??300387
2.天津市光电检测技术与系统重点实验室 ??天津 ??300387
基金项目:国家自然科学基金(61501324)
详细信息
作者简介:徐微：女，1983年生，副教授，研究方向为滤波器设计与应用、超宽带和无线高速通信
李安宇：男，1993年生，硕士生，研究方向为低功耗的数字滤波器设计
石博雅：女，1981年生，讲师，研究方向为现代通信网络、无线高速通信
通讯作者:徐微 tjpulay@163.com
中图分类号:TN713计量
文章访问数:1041
HTML全文浏览量:347
PDF下载量:52
被引次数:0
出版历程
收稿日期:2018-06-04
修回日期:2018-12-25
网络出版日期:2019-01-02
刊出日期:2019-04-01
A Novel Design Algorithm for Low Complexity Sparse FIR Notch Filters
Wei XU,,Anyu LI,
Boya SHI
1. School of Electronics and Information Engineering, Tianjin Polytechnic University, Tianjin 300387, China
2. Tianjin Key Laboratory of Optoelectronic Detection Technology and System, Tianjin 300387, China
Funds:The National Natural Science Foundation of China (61501324)
摘要
摘要:FIR陷波滤波器具有线性相位、精度高、稳定性好等诸多优势，然而当陷波性能要求较高时，通常需要较高的阶数，导致FIR陷波滤波器硬件实现复杂度大大提高。该文基于稀疏FIR滤波器设计算法和共同子式消除的思想，提出一种低复杂度的FIR陷波滤波器设计方法。该方法首先采用稀疏滤波器设计算法得到满足频域性能设计要求的FIR陷波原始滤波器系数，然后对其进行CSD编码，并分析CSD编码量化系数集中所有的2项子式和孤子的灵敏度，最后根据灵敏度的大小依次选择合理的2项子式或孤子直接合成滤波器系数集。仿真结果表明，新算法设计实现的FIR陷波滤波器比已有的低复杂度设计方法最多可减少51%的加法器，有效地降低了硬件实现复杂度，大大节省了硬件资源。
关键词:FIR陷波滤波器/
共同子式消除/
稀疏滤波器设计
Abstract:FIR notch filter has many advantages such as linear phase, high precision and good stability. However, when the notch performance is required to be high, a higher order is usually required, resulting in increased greatly hardware complexity of the FIR notch filter. Based on sparse FIR filter design algorithm and common subexpression elimination, a novel algorithm is proposed for the design of low complexity sparse FIR notch filter. First, a sparse FIR notch benchmark filter that fulfills frequency response specifications is obtained from the sparse filter design algorithm. Then, each quantized filter coefficient is represented in Canonical Signed Digit (CSD). The sensitivities of all weight-two subexpressions and isolated nonzero digits of the quantized coefficient set are analyzed. Finally, the filter coefficient set with lower implementation cost is constructed by iteratively admitting subexpressions and isolated nonzero digits according to their sensitivities. Simulation results show that the proposed algorithm can save about 51% of adder compared with other low complexity filter design algorithms, which reduces effectively the implementation complexity and saves greatly the hardware cost.
Key words:FIR notch filter/
Common subexpression elimination/
Sparse filter design
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=755057f5-8e35-498f-bea5-d696020581d0
