Method and apparatus for variance estimation in amplitude probability distribution model
AssigneeElectronics and Telecommunications Research Institute
InventorSeung Keun Park
Life of this patent
Estimating the term…
Abstract
A variance estimation apparatus for an amplitude probability distribution (APD) model includes: an APD curve calculation unit for calculating an APD curve of an APD model for noise of an electronic device; and a linearity determination unit for determining a linearity of a slope of the APD curve calculated by the APD curve calculation unit. The apparatus further includes a variance processing unit for selectively performing equivalent variance processing or different variance processing based on the linearity determination result obtained from the linearity determination unit.
Claims10
A variance estimation apparatus of an amplitude probability distribution (APD) model, the apparatus comprising:
- a processor to execute:
- calculating an APD curve of the APD model for noise of an electronic device;
- determining a linearity of a slope of the APD curve calculated;
- and
- selectively performing equivalent variance processing or different variance processing corresponding with a result of the linearity determination, wherein the different variance processing includes:
- obtaining a probability density function of a histogram for the calculated APD curve;
- searching an upper limit vertex and a lower limit vertex of the obtained probability density function;
- calculating measurement values of the upper limit vertex and the lower limit vertex of the probability density function;
- estimating a variance for each of the calculated measurement values of the upper limit vertex and that of the lower limit vertex;
- and
- obtaining a maximum probable bit error rate of a binary phase shift key (BPSK) system in a Gaussian channel environment using the variances estimated.
A variance estimation method of an amplitude probability distribution (APD) model, the method comprising:
- calculating, by a processor, an APD curve of the APD model for noise of an electronic device;
- determining whether a slope of the calculated APD curve is linear or nonlinear;
- and
- when the slope of the APD curve is nonlinear, selecting to perform different variance processing to estimate a variance, wherein the different variance processing includes:
- obtaining a probability density function of a histogram for the calculated APD curve;
- searching an upper limit vertex and a lower limit vertex of the obtained probability density function;
- calculating measurement values of the upper limit vertex and the lower limit vertex of the probability density function;
- and
- estimating a variance for each of the calculated measurement values of the upper limit vertex and that of the lower limit vertex;
- and
- obtaining a maximum probable bit error rate of a binary phase shift key (BPSK) system in a Gaussian channel environment using the variances estimated.
Dependencies are inferred from the claim text.