Digital Signal Processing Reference
In-Depth Information
TABLE 14-3. Model Fitting Observations
(1)
(2)
(3)
(4)
(5)
(6)
(7) (8) (9) (10) (11) (12)
Predicted Residual
Eye Eye Eye Eye Eye Eye
Height Width Height Width Height Width
R
Tx
Z
diff
R
TT
L
EQ
Run
(mV)
(ps)
(mV)
(ps)
(mV)
(ps)
0
0
0
0
0
1
79.43
77
80.61
73.87
−
1
.
18
3.13
1
−
1
−
1
1
1
1
30.93
52
29.93
52.65
1.00
−
0
.
65
−
−
−
−
2
1
1
1
1
1
118.70
91
119.35
91.15
0
.
65
0
.
15
3
1
−
1
1
−
1
1
148.40
89
149.25
89.15
−
0
.
86
−
0
.
15
4
−
1
1
1
−
1
1
133.38
90
133.72
89.66
−
0
.
34
0.34
5
1
−
1
−
1
−
1
−
1
76.89
88
77.63
87.37
−
0
.
74
0.63
6
0
0
−
1
0
0
83.83
90
85.44
89.86
−
1
.
61
0.14
7
−
1
−
1
−
11
−
1
56.80
84
55.69
83.87
1.11
0.13
8
1
1
−
11
−
1
58.40
79
58.23
79.33
0.17
−
0
.
33
9
−
1
1
1
1
−
1
72.24
83
71.75
82.84
0.48
0.16
10
0
0
0
−
1
0
127.60
92
123.36
95.83
4.24
−
3
.
83
11
0
0
1
0
0
105.96
89
104.71
89.08
1.25
−
0
.
08
12
0
0
0
1
0
57.13
78
61.73
74.12
−
4
.
60
3.88
13
1
−
1
−
1
1
1
17.43
40
16.77
41.18
0.66
−
1
.
18
14
1
1
1
1
1
21.64
45
21.57
46.05
0.07
−
1
.
05
15
−
1
0
0
0
0
90.41
89
93.30
90.90
−
2
.
90
−
1
.
90
16
−
1
1
−
1
−
1
−
1
68.65
88
68.88
86.94
−
0
.
24
1.06
17
0
−
1
0
0
0
93.19
89
94.20
88.92
−
1
.
01
0.08
18
−
1
−
1
−
1
−
1
1
107.62
91
107.40
90.68
0.22
0.32
19
−
1
1
−
1
1
1
32.63
62
31.56
62.61
1.07
−
0
.
61
20
1
−
1
1
1
−
1
79.85
84
79.82
84.34
0.03
−
0
.
34
21
−
1
−
1
1
−
1
−
1
93.97
87
94.38
85.84
−
0
.
41
1.16
22
0
0
0
0
0
96.24
89
95.52
89.11
0.72
−
0
.
11
23
0
0
0
0
0
96.24
89
95.52
89.11
0.72
−
0
.
11
24
0
0
0
0
−
1
82.02
85
81.20
88.08
0.82
−
3
.
08
−
−
−
25
1
1
1
1
1
106.52
86
107.75
85.38
1
.
23
0.62
26
1
0
0
0
0
100.64
89
98.10
87.05
2.54
1.95
27
0
1
0
0
0
95.09
90
94.44
90.03
0.65
−
0
.
03
are available, and the details of DOE theory are beyond our scope. However,
those readers who adopt the RSM methodology will undoubtedly wish to acquire
a more in-depth understanding of DOE fundamentals. We refer them to topics by
Myers and Montgomery [1995], Steppan et al. [1998], and Montgomery [2005]
for further study.
The response surface model is a linear function of the model coefficients. For
a given model-fitting experiment, the input variables have predetermined values.
As a result, we can replace each of the higher-order terms with single variables
that have the same values without affecting the fit. For example, if our response
Search WWH ::
Custom Search