ETS-Lindgren 5411 GTEM! Test Cell - Manual (Page 87)

Radiated Emissions Test Performance of GTEM!
| 87
The next table illustrates the results of the Student’s-t comparisons between the
several sets of data and the null hypothesis test. D is the standard mean
difference of the data sets, SD is the standard deviation of the mean differences,
t is the interval from the Student’s-t distribution for a 95% confidence, and T is
the deviation of the difference of the sample means from zero normalized to
Student’s-t distribution.
Table 12: Summary of Statistical Analysis on All Measurements
<
 =0.05, H
O
: 
D
=0>
Data
Set
Dipole 1
(Tables
1, 2, 3)
Dipole 2
(Table 4) Dipole 3
(Table 5)
PC 1
(Tables
6 & 7)
PC 2
(Table 8)
All Dipoles All PCs All Data
D -0.8 +1.2 -1.6 +2.7 -0.2 -0.7 +1.7 -0.02
S
p
2.7 2.3 7.2 4.0 4.4 4.5 13.6 8.1
n 36 7 13 10 5 38 15 53
d, f 34 6 12 9 4 37 14 52
t +/-2.030 +/-2.447 +/-2.179 +/-2.262 +/-2.776 +/-2.029 +/-2.145 +/-2.008
T -0.295 +1.388 -0.812 +2.107 -0.097 -0.980 +0.489 -0.022
H
O
Accept Accept Accept Accept Accept Accept Accept Accept
S
UMMARY OF
S
TATISTICAL
F
INDINGS
The results show that there is no significant difference between the samples data
and we can accept the null hypothesis. That is, the GTEM! measurement is the
same as the OATS measurement. The results also show that for the personal
computers, the field strength measured in the GTEM! without cable manipulation
are essentially the same as those on the OATS with cable manipulation. This
implies a much faster and more consistent measurement.
M
EASUREMENT
E
RROR
, P
RECISION
,
AND
R
EPEATABILITY
The overall absolute rss error in the test instrumentation was 2.5 dB, and the
probable instrumentation error was 1.7 dB. The precision of the measurement
was 0.1 dB. From the descriptive statistics of the dipole-to-dipole OATS
measurements, the probable variation of repeatability was 0.7 dB.