# Estimating risk for earth-satellite attenuation prediction

Article Abstract:

Annual cumulative distributions of attenuation measurements and rain-rate measurements at several sites in Europe and the US are analyzed to estimate the year-to-year variability associated with prediction of the expected cumulative distribution. An ad hoc model and a probabilistic model were developed to estimate variability: the ad hoc model summarized several observations when compared to model predictions, and the probabilistic model applied the ideas of order statistics to the prediction problem when several independent attenuation or rain-rate events in a sample year could be estimated. The statistical uncertainty of a model prediction can be estimated using the models; the estimation procedure shows the number of years of observation required to provide an estimate of the empirical distribution with a specified statistical uncertainty.

Publication Name: Proceedings of the IEEE

Subject: Electronics

ISSN: 0018-9219

Year: 1993

User Contributions:

Comment about this article or add new information about this topic:

# A Numerical Weather Prediction Model - Computational Aspects on the Cray-1

Article Abstract:

A numerical atmospheric model called the community climate model (CCM) has been assembled at the National Center for Atmospheric Research (NCAR) for use on the CRAY-1 computer. The spectral model represents fields by a finite series of analytic functions, and derivatives are approximated by differentiating each function in the series. The equations of the model are presented and the solution of these equations is described. The pipeline or vector processing features of the CRAY-1 allow for high computational rates.

Publication Name: Proceedings of the IEEE

Subject: Electronics

ISSN: 0018-9219

Year: 1984

User Contributions:

Comment about this article or add new information about this topic:

# On the Relation Between Triangular Matrix Decomposition and Linear Prediction

Article Abstract:

Coefficients of linear prediction for a random process and prediction error variances are related to the covariance matrix through triangular decomposition. Coefficients of linear prediction and corresponding prediction error variances can be easily determined. A table gives the triangular decomposition of a covariance matrix.

Publication Name: Proceedings of the IEEE

Subject: Electronics

ISSN: 0018-9219

Year: 1983

User Contributions:

Comment about this article or add new information about this topic:

- Abstracts: Novel transmission lines for the submillimeter-wave region. Submillimeter-wavelength heterodyne spectroscopy and remote sensing of the upper atmosphere
- Abstracts: Automatic mesh generation using the symmetric axis transformation of polygonal domains. Properties of the Qn-Matrix in Bilinear Transformation
- Abstracts: OSI Management and Job Transfer Services. Determination of a Digital Transfer Function from its Real Part on the Unit Circle
- Abstracts: Data General Desktop Generation Model 10: Architecture and Implementation. Business Data Usage of OSI
- Abstracts: Applications of surface acoustic and shallow bulk acoustic wave devices. A New Method of Synthesizing Acoustic Holograms