CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Long Range Channel Prediction Based on Non-Stationary Parametric Modeling

Ming Chen (Institutionen för signaler och system, Signalbehandling) ; Mats Viberg (Institutionen för signaler och system, Signalbehandling)
IEEE Transactions on Signal Processing (1053-587X). Vol. 57 (2009), 2, p. 622-634.
[Artikel, refereegranskad vetenskaplig]

Motivated by the analysis of measured radio channels and recently published physics-based scattering SISO and MIMO channel models, a new approach of long-range channel prediction based on nonstationary multicomponent polynomial phase signals (MC-PPS) is proposed. An iterative and recursive method for detecting the number of signals and the orders of the polynomial phases is proposed. The performance of these detectors and estimators is evaluated by Monte Carlo simulations. The performance of the new channel predictors is evaluated using both synthetic signals and examples of real world channels measured in urban and suburban areas. High-order polynomial phase parameters are detected in most of the measured data sets, and the new methods outperform the classical LP in given examples for long-range prediction for the cases where the estimated model parameters are stable. For the more difficult data sets, the performance of these methods are similar, which provides alternatives for system design when other issues are concerned.

Nyckelord: Radio propagation, Rayleigh channels, Nonlinear estimation, Adaptive Kalman filtering, Prediction methods

Denna post skapades 2007-10-15. Senast ändrad 2014-09-02.
CPL Pubid: 53666


Institutioner (Chalmers)

Institutionen för signaler och system, Signalbehandling


Elektroteknik och elektronik

Chalmers infrastruktur