Block Compressed Sensing For Feedback Reduction in Relay-Aided Multiuser Full Duplex Networks

Khalil Elkhalil1, Mohammed E. Eltayeb2, Abla Kammoun1, Tareq Al-Naffouri3, Hamid Reza Bahrami

  • 1King Abdullah University of Science and Technology
  • 2The University of Texas at Austin
  • 3King Abdullah University of Science and Technology (KAUST)

Details

11:30 - 12:45 | Wed 6 Jul | Salisbury A | S12.2

Session: Role of Sparsity in Communication

Abstract

Opportunistic user selection is a simple technique that exploits the spatial diversity in multiuser relay-aided networks. Nonetheless, channel state information (CSI) from all users (and cooperating relays) is generally required at a central node in order to make selection decisions. Practically, CSI acquisition generates a great deal of feedback overhead that could result in significant transmission delays. In addition to this, the presence of a full-duplex cooperating relay corrupts the fed back CSI by additive noise and the relay's loop (or self) interference. This could lead to transmission outages if user selection is based on inaccurate feedback information. In this paper, we propose an opportunistic full-duplex feedback algorithm that tackles the above challenges. We cast the problem of joint user signal-tonoise ratio (SNR) and the relay loop interference estimation at the base-station as a block sparse signal recovery problem in compressive sensing (CS). Using existing CS block recovery algorithms, the identity of the strong users is obtained and their corresponding SNRs are estimated. Numerical results show that the proposed technique drastically reduces the feedback overhead and achieves a rate close to that obtained by techniques that require dedicated error-free feedback from all users. Numerical results also show that there is a trade-off between the feedback interference and load, and for short coherence intervals, fullduplex feedback achieves higher throughput when compared to interference-free (half-duplex) feedback.