Decentralized user-centric scheduling with low rate feedback for mobile small cells

Wei Ni, Iain B. Collings, Ren Pingwei Liu

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)

Abstract

Small cells with wireless backhaul are promising, whereas challenges of severe overlapping coverage and strong interference are yet to be addressed. Coordinating small cells could resolve the challenges; however, existing multicell coordinated beamforming techniques involve high cost of communication overhead, synchronization and backhaul. Such problems may deteriorate in practical cellular applications, where there could be many users, each generating high channel feedback overhead to compete for an opportunity of being scheduled, and the downlink data signals of the coordinated cells need to be precisely synchronized at each of the users. We propose a new scheme, which cuts the overhead by 80% and enables the coordination to be practically implemented in a decentralized manner. Our scheme is a user-centric downlink scheduling approach, where mobile terminals trigger and participate in the scheduling decisions of small cells. We also develop a new recursive algorithm to optimize the quantization levels of mobile terminals' feedback, minimizing the feedback requirement. Analysis, confirmed by simulations, shows that our scheme is able to achieve 94.4% of the sum-rate upper-bound which can only be approached by idealized centralized coordination. In terms of throughput, given the 80% reduced overhead, our scheme is 139.5% better than the idealized centralized coordination approach.

Original languageEnglish
Article number6648620
Pages (from-to)6106-6120
Number of pages15
JournalIEEE Transactions on Wireless Communications
Volume12
Issue number12
DOIs
Publication statusPublished - Dec 2013
Externally publishedYes

Keywords

  • Channel quality indicator (CQI)
  • CQI quantization
  • Scheduling
  • Small cell
  • Wireless backhaul

Fingerprint Dive into the research topics of 'Decentralized user-centric scheduling with low rate feedback for mobile small cells'. Together they form a unique fingerprint.

Cite this