用Kaiser窗方法设计一个台阶状滤波器。

代码:

%% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
%% Output Info about this m-file
fprintf('\n***********************************************************\n');
fprintf(' <DSP using MATLAB> Problem 7.15 \n\n'); banner();
%% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ % staircase bandpass 3-Band
w1 = 0; w2 = 0.3*pi; delta1 = 0.01;
w3 = 0.4*pi; w4 = 0.7*pi; delta2 = 0.005;
w5 = 0.8*pi; w6 = pi; delta3 = 0.001; tr_width = min(w3-w2, w5-w3); f = [0 w2 w3 w4 w5 w6]/pi;
m = [1 1 0.5 0.5 0 0]; [Rp1, As1] = delta2db(delta1, delta3);
[Rp2, As2] = delta2db(delta2, delta3);
As = min(As1, As2) M = ceil((As-7.95)/(2.285*tr_width)) + 1; % Kaiser Window Length if As > 21 || As < 50
beta = 0.5842*(As-21)^0.4 + 0.07886*(As-21);
else
beta = 0.1102*(As-8.7);
end fprintf('\nKaiser Window method, Filter Length: M = %d. beta = %.4f\n', M, beta); n = [0:1:M-1]; wc1 = (w2+w3)/2; wc2 = (w4+w5)/2; %wc = (ws + wp)/2, % ideal LPF cutoff frequency hd = ideal_lp(wc1, M) + 0.5*(ideal_lp(wc2, M) - ideal_lp(wc1, M));
w_kai = (kaiser(M, beta))'; h = hd .* w_kai;
[db, mag, pha, grd, w] = freqz_m(h, [1]); delta_w = 2*pi/1000;
[Hr,ww,P,L] = ampl_res(h); Rp1 = -(min(db(1 :1: w2/delta_w+1))); % Actual Passband Ripple
fprintf('\nActual Passband Ripple1 is %.4f dB.\n', Rp1); Rp2 = -(min(db(w3/delta_w+1 :1: w4/delta_w+1))); % Actual Passband Ripple
fprintf('\nActual Passband Ripple2 is %.4f dB.\n', Rp2); As = -round(max(db(floor(w5/delta_w)+1 : 1 : floor(w6/delta_w)+1 ))); % Min Stopband attenuation
fprintf('\nMin Stopband attenuation is %.4f dB.\n', As); [delta1, delta3] = db2delta(Rp1, As)
[delta2, delta3] = db2delta(Rp2, As) % Plot figure('NumberTitle', 'off', 'Name', 'Problem 7.15 ideal_lp Method')
set(gcf,'Color','white'); subplot(2,2,1); stem(n, hd); axis([0 M-1 -0.3 0.5]); grid on;
xlabel('n'); ylabel('hd(n)'); title('Ideal Impulse Response'); subplot(2,2,2); stem(n, w_kai); axis([0 M-1 0 1.1]); grid on;
xlabel('n'); ylabel('w(n)'); title('Kaiser Window'); subplot(2,2,3); stem(n, h); axis([0 M-1 -0.3 0.5]); grid on;
xlabel('n'); ylabel('h(n)'); title('Actual Impulse Response'); subplot(2,2,4); plot(w/pi, db); axis([0 1 -120 10]); grid on;
set(gca,'YTickMode','manual','YTick',[-90,-65,-6,0]);
set(gca,'YTickLabelMode','manual','YTickLabel',['90';'65';' 6';' 0']);
set(gca,'XTickMode','manual','XTick',[f]);
xlabel('frequency in \pi units'); ylabel('Decibels'); title('Magnitude Response in dB'); figure('NumberTitle', 'off', 'Name', 'Problem 7.15 h(n) ideal_lp Method')
set(gcf,'Color','white'); subplot(2,2,1); plot(w/pi, db); grid on; axis([0 2 -120 10]);
xlabel('frequency in \pi units'); ylabel('Decibels'); title('Magnitude Response in dB');
set(gca,'YTickMode','manual','YTick',[-90,-65,-6,0])
set(gca,'YTickLabelMode','manual','YTickLabel',['90';'65';' 6';' 0']);
set(gca,'XTickMode','manual','XTick',[f,1+f(2:6)]); subplot(2,2,3); plot(w/pi, mag); grid on; %axis([0 2 -100 10]);
xlabel('frequency in \pi units'); ylabel('Absolute'); title('Magnitude Response in absolute');
set(gca,'XTickMode','manual','XTick',[f,1+f(2:6)]);
set(gca,'YTickMode','manual','YTick',[0,0.5, 1]) subplot(2,2,2); plot(w/pi, pha); grid on; %axis([0 1 -100 10]);
xlabel('frequency in \pi units'); ylabel('Rad'); title('Phase Response in Radians');
subplot(2,2,4); plot(w/pi, grd*pi/180); grid on; %axis([0 1 -100 10]);
xlabel('frequency in \pi units'); ylabel('Rad'); title('Group Delay'); figure('NumberTitle', 'off', 'Name', 'Problem 7.15 Amp Res of h(n)')
set(gcf,'Color','white'); plot(ww/pi, Hr); grid on; %axis([0 1 -100 10]);
xlabel('frequency in \pi units'); ylabel('Hr'); title('Amplitude Response');
set(gca,'YTickMode','manual','YTick',[-delta3,0,delta3,0.5-0.005, 0.5+0.005,1-delta1,1,1+delta1])
%set(gca,'YTickLabelMode','manual','YTickLabel',['90';'45';' 0']);
set(gca,'XTickMode','manual','XTick',[f,2]); %% +++++++++++++++++++++++++++++++++++++++++++++++++
%% fir2 function method
%% +++++++++++++++++++++++++++++++++++++++++++++++++
f = [w1, w2, w3, w4, w5, w6]/pi;
m = [1 1 0.5 0.5 0 0];
ripple = [0.01 0.005 0.001]; fprintf('\n--------- use fir2 function ---------\n'); h_check = fir2(M-1, f, m, kaiser(M, beta)); [db, mag, pha, grd, w] = freqz_m(h_check, [1]);
%[Hr,ww,P,L] = ampl_res(h_check);
[Hr,ww,P,L] = Hr_Type2(h_check); %% -------------------------------------------
%% plot
%% -------------------------------------------
figure('NumberTitle', 'off', 'Name', 'Problem 7.15 fir2 Method')
set(gcf,'Color','white'); subplot(2,2,1); stem(n, hd); axis([0 M-1 -0.3 0.5]); grid on;
xlabel('n'); ylabel('hd(n)'); title('Ideal Impulse Response'); subplot(2,2,2); stem(n, w_kai); axis([0 M-1 0 1.1]); grid on;
xlabel('n'); ylabel('w(n)'); title('Kaiser Window'); subplot(2,2,3); stem([0:M-1], h_check); axis([0 M -0.3 0.5]); grid on;
set(gca,'XTickMode','manual','XTick',[0 M/2 M]);
xlabel('n'); ylabel('h\_check(n)'); title('Actual Impulse Response'); subplot(2,2,4); plot(w/pi, db); axis([0 1 -120 10]); grid on;
set(gca,'YTickMode','manual','YTick',[-90,-65,-6,0]);
set(gca,'YTickLabelMode','manual','YTickLabel',['90';'65';' 6';' 0']);
set(gca,'XTickMode','manual','XTick',[f]);
xlabel('frequency in \pi units'); ylabel('Decibels'); title('Magnitude Response in dB'); figure('NumberTitle', 'off', 'Name', 'Problem 7.15 h_check(n) fir2 Method')
set(gcf,'Color','white'); subplot(2,2,1); plot(w/pi, db); grid on; axis([0 2 -120 10]);
xlabel('frequency in \pi units'); ylabel('Decibels'); title('Magnitude Response in dB');
set(gca,'YTickMode','manual','YTick',[-90,-65,-6,0])
set(gca,'YTickLabelMode','manual','YTickLabel',['90';'65';' 6';' 0']);
set(gca,'XTickMode','manual','XTick',[f,1+f(2:6)]); subplot(2,2,3); plot(w/pi, mag); grid on; %axis([0 2 -100 10]);
xlabel('frequency in \pi units'); ylabel('Absolute'); title('Magnitude Response in absolute');
set(gca,'XTickMode','manual','XTick',[f,1+f(2:6)]);
set(gca,'YTickMode','manual','YTick',[0,0.5, 1]) subplot(2,2,2); plot(w/pi, pha); grid on; %axis([0 1 -100 10]);
xlabel('frequency in \pi units'); ylabel('Rad'); title('Phase Response in Radians');
subplot(2,2,4); plot(w/pi, grd*pi/180); grid on; %axis([0 1 -100 10]);
xlabel('frequency in \pi units'); ylabel('Rad'); title('Group Delay');

  运行结果:

Kaiser窗长M=74,两个通带衰减分别为0.0079dB和6.0345dB,阻带最小衰减65dB>60dB,满足设计要求。

用理想低通滤波方法设计的结果,实际脉冲响应、幅度谱(dB单位)

幅度谱(dB和绝对单位)、相位谱和群延迟响应

振幅响应(台阶状)

第一个台阶(通带)

第二个台阶(通带)

阻带

题目中暗示可以用fir1函数,但查了帮助和网上资料还是不会,只好用fir2函数的方法来设计,结果如下:

群延迟不是严格的常数了,非线性相位滤波器。

《DSP using MATLAB》Problem 7.15的更多相关文章

  1. 《DSP using MATLAB》Problem 6.15

    代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ %% Output In ...

  2. 《DSP using MATLAB》Problem 5.15

    代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ %% Output In ...

  3. 《DSP using MATLAB》Problem 4.15

    只会做前两个, 代码: %% ---------------------------------------------------------------------------- %% Outpu ...

  4. 《DSP using MATLAB》Problem 2.15

    代码: %% ------------------------------------------------------------------------ %% Output Info about ...

  5. 《DSP using MATLAB》Problem 8.15

    代码: %% ------------------------------------------------------------------------ %% Output Info about ...

  6. 《DSP using MATLAB》Problem 5.38

    代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ %% Output In ...

  7. 《DSP using MATLAB》Problem 5.31

    第3小题: 代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ %% Out ...

  8. 《DSP using MATLAB》Problem 5.22

    代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ %% O ...

  9. 《DSP using MATLAB》Problem 5.21

    证明: 代码: %% ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ...

随机推荐

  1. POJ - 2777——Count Color(懒标记线段树二进制)

    Count Color Time Limit: 1000MS   Memory Limit: 65536K Total Submissions: 53639   Accepted: 16153 Des ...

  2. Vue-router的传参方式和router使用技巧

    vue传参方法一 1,路由配置 { path: '/describe/:id', name: 'Describe', component: Describe } 2,使用方法 // 直接调用$rout ...

  3. 【webpack学习笔记】a06-生产环境和开发环境配置

    生产环境和开发环境的配置目标不一样,生产环境主要是让文件压缩得更小,更优化资源,改善加载时间. 而开发环境,主要是要开发更方便,更节省时间,比如调试比如自动刷新. 所以可以分开配置不同的开发环境,然后 ...

  4. 与Recommender System相关的会议及期刊

      会议 We refer specifically to ACM Recommender Systems (RecSys), established in 2007 and now the prem ...

  5. 自动调参库hyperopt+lightgbm 调参demo

    在此之前,调参要么网格调参,要么随机调参,要么肉眼调参.虽然调参到一定程度,进步有限,但仍然很耗精力. 自动调参库hyperopt可用tpe算法自动调参,实测强于随机调参. hyperopt 需要自己 ...

  6. redi应用-延时队列

    有mq和Kafka,,一般用不上,,两个线程 ..基于Redis 的 list(列表),使用rpush/lpush操作入队列,使用lpop 和 rpop来出队列.

  7. C++ Coroutine简明教程

    在C++里,一个函数如果其函数体实现中包含co_await.co_yield.co_return中任何一个关键字,那么这个函数就是一个coroutine.其中: co_await:挂起当前的corou ...

  8. lvs用户空间命令ipvsadm

    ipvs工作在内核空间,而ipvsadm工作在用户空间,是负责管理集群服务编写规则的命令行工具 ipvsadm需要手动安装. $ yum -y install ipvsadm ipvsadm管理命令 ...

  9. python 游戏 —— 汉诺塔(Hanoita)

    python 游戏 —— 汉诺塔(Hanoita) 一.汉诺塔问题 1. 问题来源 问题源于印度的一个古老传说,大梵天创造世界的时候做了三根金刚石柱子,在一根柱子上从下往上按照大小顺序摞着64片黄金圆 ...

  10. effective_java 第34条:用接口模拟可伸缩的枚举

    例如: /** * 加减乘除枚举 * Created by yulinfeng on 8/20/17. */ public enum Operation { PLUS { double apply(d ...