Estimation and prediction for tracking trajectories in cellular networks using the recursive prediction error method

After considering the intrinsically erratic behavior of nodes in mobile networks, mobility prediction has been extensively used to improve the quality of services. Many methods have been proposed, inherited from technologies developed for signal processing and self-learning techniques and/or stochas...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2010 IEEE International Symposium on A World of Wireless, Mobile and Multimedia Networks s. 1 - 7
Hlavní autori: Milocco, R H, Boumerdassi, S
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 01.06.2010
Predmet:
Acceleration
Estimation
Mathematical model
Mobile communication
Mobile computing
Noise
Prediction algorithms
ISBN:9781424472642, 1424472644
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:After considering the intrinsically erratic behavior of nodes in mobile networks, mobility prediction has been extensively used to improve the quality of services. Many methods have been proposed, inherited from technologies developed for signal processing and self-learning techniques and/or stochastic methods. Among the latter the Extended Kalman Filter (EKF), using the received power as a measurement, is the most used. However, because the measure is not linear with distance, the EKF loses stability under certain circumstances and must be reset. Moreover, it requires the a priori knowledge of disturbances and measurement noise covariance matrices which are difficult to obtain. In this work, from the non-linear model, we derive a stable time-variant first order auto-regressive and moving average model (ARMA), and propose a prediction mechanism based on the well-known Recursive Prediction Error Method (RPEM) to predict the mobile location and then compare it with (EKF). Simulation results show that RPEM has a lower prediction error variance in most cases and similar in others to that obtained with EKF with the additional advantages that it has guaranteed stability and does not require the a priori knowledge of disturbances and measurement noise covariance matrices as in EKF.
ISBN:9781424472642
1424472644
DOI:10.1109/WOWMOM.2010.5534918