By Song X.

Show description

Read Online or Download A marginal model approach for analysis of multi-reader multi-test receiver operating characteristic PDF

Best analysis books

Systems Analysis and Modeling in Defense: Development, Trends, and Issues

This e-book comprises the complaints of an interna­ tional symposium dedicated to Modeling and research of safeguard strategies within the context of land/air struggle. It was once subsidized by means of Panel VII (on security functions of Operational study) of NATO's protection examine team (DRG) and came about 27-29 July 1982 at NATO headquarters in Brussels.

Extra resources for A marginal model approach for analysis of multi-reader multi-test receiver operating characteristic

Example text

Therefore, fullaperture extrapolation can be, at best, only reasonably stable for a finite number of extrapolation steps. , 1997). 8 Strictly speaking, “full-aperture” is also limited-aperture due to the finite acquisition survey or limited spatial extent of the data after zero padding. In this paper, the “limited-aperture” refers to the case where the full aperture is not used. 50 Both the Hale extrapolator and the Nautiyal extrapolator are designed to stabilize recursive wavefield extrapolation.

25Hz. 3c respectively as thin blue lines. The amplitude oscillation in the wavenumber domain is caused by spatial truncation of the extrapolator and the Gibbs’ phenomena. 3d). 3c, it is evident that a simple Hanning edge taper applied to the extrapolator in the space domain (thick green lines) attenuates the ripples. 3d): the longer the extrapolator, the better the stability. 3f), the instability is much less severe than with the 19- and 39-point extrapolators. Here, full aperture means that the length of the extrapolator is equal to the spatial extent of the finite wavefield (200 points in this example).

21) indicates that D (kˆ ) is symmetric with respect to kˆ (both the real and imaginary parts are even), which implies that h (n ) should also be even. Thus, the length of h (n ) , measured by N, should be odd, with the coefficient index n bounded by − ( N − 1) / 2 ≤ n ≤ ( N − 1) / 2 . Moreover, due to the symmetry of h (n ) , it is completely described by (N+1)/2 coefficients. 23) and δ mo is the Kronecker delta function defined by ⎧ 1. if m = 0; ⎩0. otherwise. 22) should be less than the number of filter coefficients (N+1)/2.

Download PDF sample

Rated 4.37 of 5 – based on 46 votes