當我們在測量一儀器或設備的電磁波強度時，一開放空間的EMI觀測
值是最直接也是最常被接受的標準測量方式。如果在不同頻率所紀錄觀察到的電磁波測量值(NSA)沒超過理論值(ideal values)加減4dB的範圍的話，這個場地就被認定為一個合格的標準場地，否則
只要有一個值超出範圍就是一個不合格的場地。在此之前，已有一些研究利用一個轉折點模型來配適觀測值。在本研究中，對於每一組觀測值和其相對應之理論值，我們分別配適一個轉折點模型，並根據所得到的迴歸參數估計值作比較。我們的目標是希望能探討觀測值與理論值間參數估計的差異，去判斷一個場地合格與否是否可行。所使用到的評估方法就是考慮觀測值所得之迴歸參數估計是否落在適當的信賴域中來做決定。最後，根據在四個不同測試場地所收集到的數據，分別利用本文中所提之統計方法配適統計模型，並進一步來比較四個測試場地的測量品質。

Open site measurement on the electromagnetic interference is the most direct and universally accepted standard
approach for measuring radiated emissions from an equipment or the radiation susceptibility of a component or equipment. In general, if the NSA measurements we recorded at different frequencies do not exceed the ideal value +-4dB, we would regard this site as
a normalized site, otherwise it is not a normalized site as long
as there is one measurement exceeds the range. A one change point
model had been used to fit observed measurements. For each set of
observations as well as the corresponding ideal values, we have
the estimated regression parameter for a one change point model.
Our ideal is using the difference of regression parameters between
ideal values and observations to assess whether a site is
qualified for measuring EMI or not. The assessment tool for
whether the testing site is normalized or not is referred to the
confidence region for the regression model parameters. Finally,
according to the data collected in this experiment, the estimated
parameters obtained from the observations will be used to do
further statistical analyses and comparing the qualities of the
four different testing sites.

Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . ii
List of Tables . . . . . . . . . . . . . . . . . . . . .iii
List of Figures . .. . . . . . . . . . . . . . . . . ..iv
1 Introduction . .. . . . . . . . . . . . . . . . . . . . 1
2 Data description . . . .. . . . . . . . . . . . . . . . 2
3 Statistical models and Analysis . . . . . . . . . . . 4
3.1 Statistical models . . . . . . . . . . . . . . . . . 5
3.2 Simulation procedure . . . . . . . . . . . . . . . 9
3.3 Quality assessment method . . . . . . . . . . . . 11
3.4 Pattern of the estimated parameter . . . . . . . . . 12
4 Results and Comparisons . . . . . . . . . . . . . . . 13
4.1 Simulation results . . . . . . . . . . . . . . . . . 14
4.2 Site comparison . . . . . . . . . .. . . . . . . . 19
5 Conclusion . . . . . . . . . . . . . . . . . . . . . . 20
References . . . . . . . . . . . . . . . . . . . . 22
Appendix . . . . . . . . . . . . . . . . . . . . . . . 24

References
1. Akaike, H. (1974). A new look at the statistical model identi‾cation, IEEE Trans-action on Automatic Control, AC-19, 716-723.

2. Akira, S. (1990). Formulation of Normalized Site Attenuation in Terms of Antenna Impedances, IEEE Transactions on Electromagnetic Compatibility, 32 , 257-263.

3. Akira, S. (1992). Correction Factors for Normalized Site Attenuation, IEEE Trans-actions on Electromagnetic Compatibility, 34, 461-470.

4. Page, E. S. (1954). Continuous inspection schemes, Biometrika, 41, 100-114.

5. Page, E. S. (1955). A test for a change in a parameter occurring at an unknown point, Biometrika, 42, 523-527.

6. Page, E. S. (1957). On problems in which a change in parameter occurs at an unknown point, Biometrika, 44, 248-252.

7. Richard H. J. and Indranil D. (1995). Determining one or more change points, Chemistry and Physics of LIPIDS, 76, 1-6.

8. Wang, D.Y. , Lin, K.-H. , Huang, M.-N. L. , Cheng, C.P. (2004). Uncertainty Studies of Radiated Emissions Test Site Based on Measured Normalized Site Attenuation
Data. manuscript. Department of Applied Mathematics, National Sun Yat-sen University, Kaohsiung, Taiwan 804, ROC.

9. Wang, D.Y. , Lin, K.-H. , Huang, M.-N. L. (2004). Quantitative Parameters for Normalized Site Attenuations, Maximum Electric Field Strengths and Antenna Factors.
manuscript. Department of Applied Mathematics, National Sun Yat-sen University, Kaohsiung, Taiwan 804, ROC.

10. Huang, M.-N. L. , Chiu, S.T. (2004). Statistical model building and inference about the Normalized Site Attenuation measurements for electromagnetic interference (EMI). manuscript. Department of Applied Mathematics, National Sun Yat-sen University, Kaohsiung, Taiwan 804, ROC.