-
Notifications
You must be signed in to change notification settings - Fork 1
/
Untitled2.m
171 lines (156 loc) · 5.92 KB
/
Untitled2.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
% N = load('daopin_4_2.txt');
clc;clear
sample = csvread('sample.txt');
MM = zeros(250, 1);
[nrows, ncols] = size(sample);
% x = [-124,-100,-75,-50,-25,0,25,50,75,100,125];
% x = -124:5:125;
% % x = sort(randperm(250, 80)-125)';
% x = x';
x = load('dp62.txt');
% x = [-124 -116 -108 -100 -92 -84 -76 -68 -60 -52 -44 -36 -28 -20 -12 -4 ...
% 4 12 20 28 36 44 52 60 68 76 84 92 100 108 116 124]';
% csvwrite('dp62.txt',x')
xx = x+125;
noise = normrnd(-0,0.5, length(x),1)+normrnd(0,0.5, length(x),1)*1i;
noise_real = sample(xx, 3);
noise_imag =sample(xx, 4);
% noise_real = sample(xx, 1) + real(noise);
% noise_imag = sample(xx, 2) + imag(noise);
% %
% noise_real = xiaobo(noise_real);
% noise_imag = xiaobo(noise_imag);
% static_real = zeros(length(noise_real),1);
% static_imag = zeros(length(noise_imag),1);
% loop = 100000;
% u = 0;
% sigma = 1;
% for i=1:loop
% noise_r = normrnd(u, sigma, length(noise_real), 1);
% static_real = static_real + noise_real - noise_r;
% noise_i = normrnd(u, sigma, length(noise_real), 1);
% static_imag = static_imag + noise_imag - noise_i;
% end
% static_real = static_real / loop;
% static_imag = static_imag / loop;
noise = normrnd(0,0.04, 250,1)+normrnd(0,1, 250,1)*1i;
% predict_real = OMP_func(x, 62, 250, 1);
% predict_imag = OMP_func(x, 62, 250, 1);
%method 1
fitmodel_sample1 = createFit1(x, noise_real);
fitmodel_sample2 = createFit1(x, noise_imag);
predict_real = fitmodel_sample1(-124:125);
predict_imag = fitmodel_sample2(-124:125);
% for i=1:250
% for j=1:1
% MM(i,j) = predict_real(i,1)+predict_imag(i,1)*1i;
% end
% end
% MM = ifft(MM);
% lev = 1; wn = 'sym8';
% MM = wden(MM,'heursure','s','sln',lev,wn);
% MM = fft(MM);
% predict_real = real(MM);
% predict_imag = imag(MM);
% predict_imag = ifft(predict_imag);
% lev = 1; wn = 'sym8';
% predict_imag = wden(predict_imag,'heursure','s','sln',lev,wn);
% predict_imag = fft(predict_imag);
predict_real = smooth(predict_real,2, 'moving');
predict_imag = smooth(predict_imag,2, 'moving');
% %小波变换
predict_real = xiaobo(predict_real);
predict_imag = xiaobo(predict_imag);
% %小波去噪
% lev=10;wn='db45';ln = 'one';tptr='rigrsure';
% lev=6;wn='coif5';ln = 'sln';tptr='sqtwolog';
% 利用soft SURE阈值规则去噪
% predict_real= wden(predict_real, tptr, 's', ln, lev, wn);
% predict_imag= wden(predict_imag, tptr, 's', ln, lev, wn);
%
% %小波去噪
% lev=20;wn='sym8';
% % 利用soft SURE阈值规则去噪
% predict_real= wden(predict_real, 'minimaxi', 's', 'sln', lev, wn);
% predict_imag= wden(predict_imag, 'minimaxi', 's', 'sln', lev, wn);
% predict_real = predict_real - real(noise);
% predict_imag = predict_imag - imag(noise);
channel_real = sample(:, 1);
channel_imag = sample(:, 2);
C = sum((predict_real-channel_real).^2 + (predict_imag-channel_imag).^2);
H = sum(channel_real.^2 + channel_imag.^2);
fprintf('method1, Ea: %d, Eb: %d, E: %d\n', C/H, length(x)/2500, C/H + length(x)/2500);
plot(-124:125, predict_real, 'r')
hold on
digits(2);
predict_real= vpa(predict_real);
predict_imag = vpa(predict_imag);
%method 2
fitmodel_sample1 = createFit2(x, noise_real);
fitmodel_sample2 = createFit2(x, noise_imag);
predict_real = fitmodel_sample1(-124:125);
predict_imag = fitmodel_sample2(-124:125);
%小波变换
predict_real = xiaobo(predict_real);
predict_imag = xiaobo(predict_imag);
%小波去噪
lev=5;wn='sym8';ln = 'mln';tptr='heursure';
% lev=7;wn='coif5';ln = 'sln';tptr='sqtwolog';
% 利用soft SURE阈值规则去噪
predict_real= wden(predict_real, tptr, 's', ln, lev, wn);
predict_imag= wden(predict_imag, tptr, 's', ln, lev, wn);
% %小波变换
% predict_real = xiaobo(predict_real);
% predict_imag = xiaobo(predict_imag);
% predict_real = predict_real - real(noise);
% predict_imag = predict_imag - imag(noise);
%
% noise_real = smooth(noise_real, 'moving');
% noise_imag = smooth(noise_imag, 'moving');
channel_real = sample(:, 1);
channel_imag = sample(:, 2);
C = sum((predict_real-channel_real).^2 + (predict_imag-channel_imag).^2);
H = sum(channel_real.^2 + channel_imag.^2);
fprintf('method2, Ea: %d, Eb: %d, E: %d\n', C/H, length(x)/2500, C/H + length(x)/2500);
fprintf('percent, Ea: %d , Eb: %d , E: %d \n', C/H/(C/H + length(x)/2500), length(x)/2500/( C/H + length(x)/2500), C/H + length(x)/2500);
% consider noise: Ea: 4.264550e-02, Eb: 2.000000e-02, E: 6.264550e-02
% minus noise: Ea: 4.011335e-02, Eb: 2.000000e-02, E: 6.011335e-02 样本一次
% no 2 noise: Ea: 4.206490e-02, Eb: 5.000000e-02, E: 9.206490e-02
% no 1 noise: Ea: 4.123032e-02, Eb: 1.000000e-01, E: 1.412303e-01
% xiaobobianhuan: Ea: 7.030419e-02, Eb: 2.000000e-02, E: 9.030419e-02
% 统计平均100 : Ea: 4.256002e-02, Eb: 2.000000e-02, E: 6.256002e-02
% 统计平均10000 : Ea: 4.115769e-02, Eb: 2.000000e-02, E: 6.115769e-02
% 统计平均100000: Ea: 4.114562e-02, Eb: 2.000000e-02, E: 6.114562e-02
% 样本平均10000 : Ea: 4.010292e-02, Eb: 2.000000e-02, E: 6.010292e-02 % 当前最优，平均没用
% 小波变换 预测后再变换50个导频 : Ea: 3.667115e-02, Eb: 2.000000e-02, E: 5.667115e-02 4-14最优
% 65个导频 50%:50% : Ea: 2.263320e-02, Eb: 2.600000e-02, E: 4.863320e-02
% 改小波变换，改导频 62个导频
%method3
% S1 = sample(:,1)+1i*sample(:,2);
% S1 = S1.';
% S2 = sample(xx,3)+1i*sample(xx,4);
% S2 = S2.';
% % method = 'linear'/'spline'/'cubic'|插值方法
% P2 = interpolate(S2,xx',250,'cubic');
% plot(-124:125, real(P2),'y','linewidth',2)
% hold on
% P2t = ifft(P2);
% lev = 1; wn = 'sym8';
% P2t_modified = wden(P2t,'heursure','s','sln',lev,wn);
% P2_modified = fft(P2t_modified);
% Ea = sum(power(abs(S1-P2_modified),2))/sum(power(abs(S1),2));
% Eb = length(xx)/2500;
% E = Ea+Eb;
% fprintf('method3, Ea: %d, Eb: %d, E: %d\n',Ea, Eb, E);
% plot(-124:125, real(P2_modified),'r','linewidth',2)
% hold on
plot(-124:125, predict_real, 'g')
hold on
plot(-124:125, sample(:,1), 'b')
hold on
scatter(x', noise_real,'d','filled')
hold on
% scatter(x', noise_real-real(noise), 'filled')
% hold on
plot(-124:125, sample(:,3), 'm')
legend('predict_1','predict_2','real', 'given', 'noise')