摘要
提出了一种迭代算法 ,在 FIR数字滤波器的自适应设计过程中自动改变各个频率指标上代价因子的值 ,最终实现代价因子的最优取值。利用这种改进方法 ,可以实现 FIR数字滤波器和设计指标之间的最佳拟合。给出了两个设计实例 ,其中幅度响应每倍频程衰减 6 d B的滤波器 ,可以用于模拟水声阵列接收的宽带海洋噪声 ,而高精度时延的滤波器 ,则可以实现波束形成中所需要的非整数倍采样间隔的时延。
To date, to our best knowledge, there does not exist any analytical method for obtaining optimum cost functions in adaptively designing FIR digital filters. We propose an iterative algorithm for obtaining adaptively such optimum functions. Section 1 sums up the essentials of usual adaptive design of FIR digital filters. Section 2 deals with our iterative algorithm for enhancing the effectiveness of FIR digital filter design through adaptively obtaining optimum cost functions. Essentially, our iterative algorithm is based on the consideration that when responses at some frequencies do not satisfy the design requirements, we use an adaptive procedure to change the corresponding cost function values so as to reach a best fit between the FIR digital filter response and the design requirements. Section 3 uses only two examples to illustrate the applicability of our iterative algorithm to many engineering problems. In section 3.1, a filter with response having -6dB/oct attenuation above 100Hz was designed for broadband sea noise simulation (Figs. 2 and 3). In section 3.2, the algorithm was used to design a high precision time delay filter which implemented the time delay of fractional sample period with small errors over a wide frequency range (Figs. 4 and 5 and Tables 1 and 2).
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2002年第4期554-558,共5页
Journal of Northwestern Polytechnical University
基金
国家自然科学基金 (6 980 2 0 10 )
