MIMO precoder selections in decode-forward relay networks with finite feedback

Duckdong Hwang, Junil Choi, Bruno Clerckx, Gil Kim

    Research output: Contribution to journalArticlepeer-review

    4 Citations (Scopus)

    Abstract

    We consider a system with precoding in the half duplex Decode-Forward (DF) relay networks, where multiple antennas are adopted at the source, the destination and the relay. Based on the channel state information (CSI), a few feedback bits are fed back from the destination to the source and the relay to select the source and the relay precoders. A suboptimal precoder selection criterion for the symbol vector error minimization and an optimal precoder selection criterion for the capacity maximization are proposed for the practical situation where relays are placed between the source and the destination. We analyze the diversity performance of the criteria and show that the full diversity is achieved when the rank one precoder sets span the vector channel space formed by the transmit antennas of the source and the those of the relay.

    Original languageEnglish
    Article number5771500
    Pages (from-to)1785-1790
    Number of pages6
    JournalIEEE Transactions on Communications
    Volume59
    Issue number7
    DOIs
    Publication statusPublished - 2011 Jul

    Bibliographical note

    Funding Information:
    Paper approved by D. L. Goeckel, the Editor for Space-Time and OFDM of the IEEE Communications Society. Manuscript received January 11, 2010; revised July 12, 2010, October 19, 2010, and February 16, 2010. the authors are with Samsung Electronics Co. (e-mail: [email protected]). This work is supported by Samsung Electronics Co. Digital Object Identifier 10.1109/TCOMM.2011.050911.100003

    Keywords

    • DF relay network
    • MIMO
    • codebook
    • diversity
    • precoder selection

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering

    Fingerprint

    Dive into the research topics of 'MIMO precoder selections in decode-forward relay networks with finite feedback'. Together they form a unique fingerprint.

    Cite this