Nonlinear Analysis of Angle Demodulation,

Abstract

The angle-demodulation problem is investigated from the viewpoint of nonlinear filtering theory. The mean square error criterion and state-variable representation of Kushner's classical nonlinear approach are adopted. The relationship between the optimum nonlinear filter and certain decoupled implementations is shown. The decoupled implementations are subsequently analyzed and nonlinear effects such as thresholding, cycle skipping, and phase locking are investigated by Fokker-Planck methods. The mean square estimation error is obtained exactly for the first-order signal-to-noise ratios (SNR) and demodulator gains. Linear methods of analysis are shown to lead to erroneous results even when the SNR is large. Since solutions to the Fokker-Planck equation for angle-modulation problems are usually difficult to obtain, an orthogonal projection technique is presented for obtaining approximate solutions. The approximate solutions are shown to compare favorably with simulations above and below threshold for the frequency demodulation of a message with a first-order spectrum. (Author)

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1972
Accession Number
AD0767714

Entities

People

  • Francis S. Nakamoto
  • Neil J. Bershad

Organizations

  • University of California, Irvine

Tags

DTIC Thesaurus Topics

  • Demodulation
  • Demodulators
  • Equations
  • Filters
  • Filtration
  • Fokker Planck Equations
  • Frequency
  • Mathematics
  • Modulation
  • Nonlinear Analysis
  • Simulations
  • Spectra

Fields of Study

  • Physics

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Radio communications and signal processing.