𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Influencing the approach to the localization threshold

✍ Scribed by N. Garcia; J.H. Li; W. Polkosnik; T.D. Cheung; P.H. Tsang; A.A. Lisyansky; A.Z. Genack


Publisher
Elsevier Science
Year
1991
Tongue
English
Weight
600 KB
Volume
175
Category
Article
ISSN
0921-4526

No coin nor oath required. For personal study only.

✦ Synopsis


We argue that the Thouless criterion for localization is universal. Because the Thouless number 6 is the ratio of the width and spacing of the modes of the sample, it can be adjusted by changing the macroscopic characteristics of the sample, such as the transverse dimensions, the reflectivity at the boundaries and the character of internal microstructure resonances. We find the dependence of the correlation parameter 8 -1 and hence of the proximity to the localization threshold upon these characteristics. These results are illustrated by microwave and optical experiments.


πŸ“œ SIMILAR VOLUMES


On the Time – Dependent Approach to Ande
✍ Dirk Hundertmark πŸ“‚ Article πŸ“… 2000 πŸ› John Wiley and Sons 🌐 English βš– 201 KB

A simple proof of Anderson localization is obtained. This is done by giving a bound on the averaged time evolution and then ruling out the existence of any continuous spectrum for large disorder with the help of the RAGE theorem. Furthermore the decoupling lemma of Aizenman and Molchanov is extended

An LMI approach to the evaluation of ala
✍ Y. E. Faitakis; S. Thapliyal; J. C. Kantor πŸ“‚ Article πŸ“… 1998 πŸ› John Wiley and Sons 🌐 English βš– 115 KB πŸ‘ 2 views

A new approach is presented for selecting alarm thresholds in a simple fault detection system. Bounds are computed on the magnitudes of the minimum detectable fault and the maximum non-detectable fault. The use of the 2-norm for this calculation results in a linear matrix inequality (LMI) problem. A

Assessing local influence in PLS regress
✍ Bo Cheng; Xizhi Wu πŸ“‚ Article πŸ“… 2001 πŸ› Elsevier Science 🌐 English βš– 130 KB

Partial least squares (PLS) regression has received increasing attention in recent years. However, like other regression methods, PLS ΓΏtting could be substantially altered by one or a few in uential points. This paper assesses the local in uence by examining the second order derivatives on certain p