Kalman Predictions for Multipoint OFDM Downlink Channels

Rikke Apelfröjd , Uppsala University.

Technical Report, Signals and Systems, Uppsala University. Second Edition, March 2018

Also presented at the Swedish Communication Technologies Workshop (Swe-CTW), Västerås, Sweden, June 2014.

Report in Pdf (Second edition, March 2018)

Poster in Pdf presented at Swe-CTW 2014 in Västerås.

Abstract:

Coordinated Multipoint (CoMP) transmission provides high theoretic gains in spectral efficiency with coherent Joint Transmission (JT) to multiple users. However, this requires accurate Channel State Information at the Transmitter (CSIT). Unfortunately, coherent JT CoMP is often accompanied by long system delays, due to e.g. data sharing over backhaul links. These delays will cause the CSIT to be outdated.

This report provides a detailed description on how to increase the accuracy of the CSIT by utilizing Kalman filters to predict multipoint Orthogonal Frequency Division Multiplexing (OFDM) downlink channels.

The small scale fading of these channels are modeled by Auto Regressive (AR)models of finite order. The report includes descriptions on how to estimate these models based on past knowledge of the channel as well as analytical result on the predictability of such models. Different technical design aspects for deploying the Kalman filters in communication systems, such as pilot patterns, AR model estimations and the location of Kalman filters that predict downlink Frequency Division Duplex (FDD) channels, are also discussed.

The aim of the report is to in detail describe the prediction procedure used in previous work. Some of the results from this previous work are here presented and extended to provide a complete overview. All simulation results are based on measured channels.

The report also includes a description on how to simulate block-fading channels with a specified channel accuracy that would have been obtained with Kalman prediction.

Related publications:

PhD Thesis by Rikke Apelfröjd, 2018
(of which this is the first contribution paper).

Paper at IEEE ICC 2012 on robust Coordinated Multipoint transmission, that uses the described Kalman prediction.

EURASIP JWCN paper 2014 that evaluates robust linear precoding and simple user grouping on measured channels.

Paper at IEEE PIMRC 2012 that uses block-fading channel models with prediction errors generated as described in this report.

Paper at Eusipco 2007, comparing the use of overlapping (code-orthogonal) versus dedicated (resource-orthogonal) uplink pilots.

Channel Estimation and Prediction for MIMO OFDM Systems.
PhD Thesis by Danel Aronsson, Uppsala University 2011.

Prediction of Mobile Radio Channels.
PhD Thesis by Torbjörn Ekman, Uppsala University 2002.