uu.seUppsala University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Kalman Predictions for Multipoint OFDM Downlink Channels
Uppsala University, Disciplinary Domain of Science and Technology, Technology, Department of Engineering Sciences, Signals and Systems Group.
(English)Manuscript (preprint) (Other academic)
Abstract [en]

Coordinated Multipoint (CoMP) transmission provides high theoreticgains in spectral eciency with coherent Joint Transmission (JT) to mul-tiple users. However, this requires accurate Channel State Information atthe Transmitter (CSIT). Unfortunately, coherent JT CoMP often is accom-panied by long system delays, due to e.g. data sharing over backhaul links.Therefore, the CSIT will be outdated.This report provides a detailed description on how to increase the accu-racy of the CSIT by utilizing Kalman lters to predict Orthogonal FrequencyDivision Multiplexing (OFDM) downlink channels. The small scale fading ofthese channels are modeled by Auto Regressive (AR) models of nite order.The report includes descriptions on how to estimate these models basedon past knowledge of the channel as well as analytical result on the pre-dictability of such models. Dierent technical design aspects for deployingthe Kalman lters in communication, such as pilot patterns, AR model esti-mations and the location of Kalman lters that predict downlink FrequencyDivision Duplex (FDD) channels, are also discussed.The aim of the report is to in detail describe the prediction procedureused in previous work. Some of the results from this previous work arehere presented and extended to provide a complete overview. All simulationresults are based on measured channels.The report also includes a description on how to model block-fading chan-nels with a specied channel accuracy that would have been obtained withKalman predictions. This model can then be used for system simulations.V:

Keyword [en]
Linear predictions of OFDM channels, outdated channel state information (CSI), Coordinated Multipoint (CoMP), Predictability of radio channles, AR modelling of radio channels
National Category
Engineering and Technology
Research subject
Electrical Engineering with specialization in Signal Processing
Identifiers
URN: urn:nbn:se:uu:diva-224263OAI: oai:DiVA.org:uu-224263DiVA: diva2:716087
Available from: 2014-05-08 Created: 2014-05-08 Last updated: 2014-06-04
In thesis
1. Design Aspects of Coordinated Multipoint Transmission: A Study of Channel Predictions, Resource Allocation, User Grouping and Robust Linear Precoding for Coherent Joint Transmission
Open this publication in new window or tab >>Design Aspects of Coordinated Multipoint Transmission: A Study of Channel Predictions, Resource Allocation, User Grouping and Robust Linear Precoding for Coherent Joint Transmission
2014 (English)Licentiate thesis, comprehensive summary (Other academic)
Abstract [en]

Shadowed areas and interference at cell borders pose great challenges for future wireless broadband systems. Coordinated Multipoint (CoMP) coherent joint transmission has shown the potential to overcome these challenges by turning harmful interference into useful signal power. However, there are obstacles to overcome before coherent joint transmission CoMP can be deployed. Some of these are the investigated in this thesis.

First, coherent joint transmission requires very accurate Channel State Information (CSI), but unfortunately long system latencies cause outdating of the CSI. This can to some extend be counteracted by channel predictions. Two schemes are here investigated for predicting downlink Frequency Division Duplex (FDD) Orthogonal Frequency Division Multiplexing (OFDM) channels; Kalman filters and “predictor antennas”. The first is well suited for slow moving users, e.g. pedestrians or cyclists, as it does not require any special antenna setup. The second, which utilizes an extra antenna, located in front of the main receive antennas, is well suited for vehicular users, such as buses or trams, as these require long spatial prediction horizon.

Second, a user grouping and resource allocation scheme is investigated. This scheme forms CoMP groups by local resource allocations and provides multi-user diversity gains very close to the optimal gains, found through an extensive combinatorial search. It has very low complexity, requires less feedback capacity than other schemes and places no demands on backhaul capacity.

Finally, a linear precoder, which is robust to errors in the CSI, is investigated. This precoder takes the covariances of the channel errors into account while optimizing a Mean Squared Error (MSE) criterion. The MSE criterion includes design parameters that can be used as flexible tools for low dimensional searches with respect to an arbitrary optimization criterion, e.g. a weighted sum-rate criterion. The precoder design is also extended to handle backhaul constraints.

Results show that with the combination of these three schemes: channel predictions, the proposed user grouping and resource allocation scheme and the robust linear precoder, then coherent joint transmission will indeed provide large capacity gains.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 2014. xiv+39 p.
Keyword
Network MIMO, Centralized CoMP, Robust beamforming with imperfect CSI, Limited backhaul, Linear predictions of OFDM channels, Kalman filters, Predictor antennas, Long term channel prediction, Local resource allocation and user grouping
National Category
Signal Processing
Research subject
Electrical Engineering with specialization in Signal Processing
Identifiers
urn:nbn:se:uu:diva-224264 (URN)
Presentation
2014-08-22, 2002, Ångströmslaboratoriet, Lägerhyddsvägen 1, Uppsala, 09:15 (English)
Opponent
Supervisors
Available from: 2014-06-04 Created: 2014-05-08 Last updated: 2014-06-04Bibliographically approved

Open Access in DiVA

No full text

By organisation
Signals and Systems Group
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 575 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf