Multidimensional Likelihood Function in the Problem of Estimating Time-Frequency Parameters of Signals
Olga Safaryan, Irina A. Pilipenko, Nikolay Boldyrikhin, Vasiliy I. Yukhnov
Abstract
The paper deals with the construction and use of a multidimensional likelihood function for estimating time-frequency parameters. A signal representation model is proposed that takes into account both random and constant frequency deviations from the expected value during the observation interval. It is assumed that the random frequency deviation at each of the measurement intervals during the observation interval obeys the normal distribution law with a constant value of the standard deviation. Numerical simulation results are presented to illustrate the unbiased and asymptotic efficiency of the obtained estimates.
Topics & Concepts
Standard deviationMathematicsConstant (computer programming)Frequency deviationInterval (graph theory)Function (biology)Applied mathematicsStatisticsRepresentation (politics)Probability density functionExpected valueSIGNAL (programming language)Time–frequency analysisMathematical optimizationComputer scienceCombinatoricsProgramming languageAutomatic frequency controlRadarLawPoliticsBiologyTelecommunicationsEvolutionary biologyPolitical scienceFault Detection and Control SystemsSensor Technology and Measurement SystemsAdvanced Electrical Measurement Techniques