Full metadata record
DC FieldValueLanguage
dc.contributor.authorChung, Yao-Hsingen_US
dc.contributor.authorChang, Chung-Juen_US
dc.date.accessioned2014-12-08T15:24:36Z-
dc.date.available2014-12-08T15:24:36Z-
dc.date.issued2009en_US
dc.identifier.isbn978-1-4244-5123-4en_US
dc.identifier.urihttp://hdl.handle.net/11536/17065-
dc.identifier.urihttp://dx.doi.org/10.1109/PIMRC.2009.5449834en_US
dc.description.abstractIn this paper we propose an economized-CSI (channel state information) opportunistic scheduling (ECOS) scheme for OFDMA/TDD downlink systems. The ECOS scheme consists of a quality-of-service (QoS) guarantee scheduling (QGS) algorithm and a CSI overhead reduction (COR) algorithm. The QGS algorithm guarantees QoS requirements by adjusting priority value and anticipated rate of users dynamically. In addition, it achieves balance between QoS guarantee and throughput maximization by employing a priority threshold. The COR algorithm performs admissible user selection to economize the amount of CSI overhead. Since all users do not need to feed back CSI every frame, the transmission power used for CSI reporting is economized. Besides, the COR algorithm can reduce the computation complexity of the QGS algorithm because the number of users to be scheduled is reduced. Simulation results show that the proposed ECOS scheme greatly reduces the uplink bandwidth occupancy of CSI feedback. Also, it achieves high system throughput and maintains QoS guarantee at high traffic load.en_US
dc.language.isoen_USen_US
dc.titleOpportunistic Scheduling with Economized CSI Feedback for OFDMA/TDD Downlink Systemsen_US
dc.typeProceedings Paperen_US
dc.identifier.doi10.1109/PIMRC.2009.5449834en_US
dc.identifier.journal2009 IEEE 20TH INTERNATIONAL SYMPOSIUM ON PERSONAL, INDOOR AND MOBILE RADIO COMMUNICATIONSen_US
dc.citation.spage1391en_US
dc.citation.epage1395en_US
dc.contributor.department電信工程研究所zh_TW
dc.contributor.departmentInstitute of Communications Engineeringen_US
dc.identifier.wosnumberWOS:000305824601080-
Appears in Collections:Conferences Paper


Files in This Item:

  1. 000305824601080.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.