THEORY OF OPTIMUM MULTIPLE MEASUREMENTS
Abstract
Multiple measurements with random inputs are studied. Methods of weighting and combining the measured signals are proposed. The relationships between the measured signals and the desired signal are assumed linear and time-invariant; the random inputs are assumed stationary with rational spectral density functions; the criterion of performance used is to minimize the mean squared value of the continuous error between the estimate and the desired signal; and the weighting operations are assumed linear. Two kinds of single-rate systems are studied: Single-rate multiple measurements with known spectral density functions of signal and noise, and Single-rate multiple measurements with known noise but unknown signal spectral density functions. A new method employing the frequency domain optimization theorems together with factorization theorems of rational matrices is proposed for obtaining the optimum system of the first kind. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1961
- Accession Number
- AD0269568
Entities
People
- James C. Hung
Organizations
- New York University