• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.36 no.11
• /
• pp.813-818
• /
• 1987

In this paper, high resolution spectral estimation by AR modelling and principal comonent analysis is proposed. The given data can be expanded by the eigenvectors of the estimated covariance matrix. The eigenspectrum is obtained for each eigenvector using the Autoressive(AR) spectral estimation technique. The final spectrum estimate is obtained by weighting each eigenspectrum with the corresponding eigenvalue and summing them. Although the proposed method increases in computational complexity, it shows good frequency resolution especially for short data records and narrow-band data whose signal-to-noise ratio is low.

• Journal of Environmental Science International
• /
• v.5 no.3
• /
• pp.347-359
• /
• 1996

A study was carried out on the distribution of chlorophyll a and water quality in the dry season in Yosuhae bay and adjoining sea, Southern Korea, in July of 1994. Concentration of salinity and phosphate were lower in the outer bay than in the inner bay. While nitrate and silicate were higher in the former than in the latter. We were identified with coastal waters of origin from China with the lower salinity in outer bay. The China coastal water was characteristic of high nutrients and phytoplankton biomass, such as chlorophyll a. The principal component analysis-(PCA) on the analytical data proved that high density of phytoplankton biomass , occurred under the condition of low salinity and high concentration of nissoived Inorganic nutrients. It is thought that the thermoharine structure and biological produtions of Yosuhae bay were controlled by the China coastal water in the outer bay.