Concept of a neural system for real-time evaluation of spectroscopic measurements

被引：1

作者：

Redei, L ^{[1
]}

Fried, M ^{[1
]}

Lohner, T ^{[1
]}

Polgar, O ^{[1
]}

Barsony, I ^{[1
]}

Forizs, S ^{[1
]}

Horvath, ZG ^{[1
]}

Humlicek, J ^{[1
]}

Wallinga, H ^{[1
]}

机构：

[1] Res Inst Tech Phys & Mat Sci, H-1525 Budapest, Hungary

来源：

OPTIKA '98 - 5TH CONGRESS ON MODERN OPTICS | 1998年 / 3573卷

关键词：

neural networks; spectroscopic measurements; photosensor arrays; digital multi-chip module;

D O I：

10.1117/12.320999

中图分类号：

O43 [光学];

学科分类号：

070207 ; 0803 ;

摘要：

A hardware implementation of a Backpropagation (BP) feedforward neural network has been designed. The tool was proposed for reflectometric measurements integrated together with photosensor arrays. The intelligent reflectometric sensor is being implemented in a multi-chip-module approach. A logarithmic input transformation is applied for easing the misalignment and parameter scatter correction. It also allows for easy ratio calculation by subtraction for normalisation with the reference value. The neural network was designed for complexities up to 100 inputs, 30 hidden neurons and 5 outputs. The digital building blocks (neurons) utilise a logic approximation of the sigmoid nonlinearity and the possibility of weight scaling. These hardware solutions result in a simultaneous area reduction and speed gain, at the cost of slightly decreased performance. Simulations of the proposed neural system prove applicability for evaluation of optical measurements, were performed for reflectometric and ellipsometric data thin porous layers. Hardware simulations showed good correspondence to the optimum-case neural software simulations.

引用

页码：155 / 159

页数：5