Estimation of the Initial Phase of Narrowband Radio Signal with Unknown Amplitude and Duration
| Autor: | Konstantin S. Kalashnikov, Andrey P. Trifonov, Yu. E. Korchagin, M. V. Trifonov |
|---|---|
| Přispěvatelé: | The study was carried out at the expense of the grant from the Russian Science Foundation (Project No. 17-71-10057). |
| Rok vydání: | 2018 |
| Předmět: |
Mathematical analysis
Block diagram Schematic 020206 networking & telecommunications 02 engineering and technology Expected value Exponential function Narrowband Amplitude maximum likelihood estimate quasi-likelihood estimate initial phase duration amplitude bias dispersion 0202 electrical engineering electronic engineering information engineering A priori and a posteriori 020201 artificial intelligence & image processing Electrical and Electronic Engineering Invariant (mathematics) Mathematics |
| Zdroj: | Radioelectronics and Communications Systems; Том 61, № 6 (2018); 231-241 |
| ISSN: | 1934-8061 0735-2727 |
| DOI: | 10.3103/s0735272718060018 |
| Popis: | The quasi-likelihood and maximum likelihood algorithms for estimating the initial phase of radio signal with arbitrary shaped envelope and unknown duration and amplitude have been synthesized. Schematic block diagrams of initial phase meters are proposed. Characteristics of the synthesized algorithms were determined, and errors of the initial phase estimates were compared. The structure of quasi-likelihood estimate of the initial phase is shown to be invariant with respect to the lack of knowledge of radio signal amplitude. However, a mismatch between the expected value of signal duration and its true value can lead to a marked increase of dispersion of the initial phase quasi-likelihood estimate. The loss in accuracy of the initial phase estimate due to a priori lack of knowledge of signal duration has been determined. Analytical expressions for statistical characteristics of maximum likelihood estimate of initial phase at large signal-to-noise ratios are shown to coincide asymptotically with the characteristics of maximum likelihood estimate of the initial phase of radio signal with a priori known amplitude and duration. Therefore, a priori unknown signal duration (asymptotically with an increase of signal-to-noise ratio) does not affect the accuracy of the maximum likelihood estimate of initial phase. The computer methods of statistical simulation were used to determine the applicability limits of asymptotic expressions for characteristics of the maximum likelihood estimate in respect of signals with linear and exponential envelopes. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink