Dynamic decode-and-forward and amplify-and-forward cooperative strategy using distributed space-time code in uplink MIMO systems with multiple relays

  • Authors:
  • Yin Yang;Lilian Luo;Roger S. Cheng

  • Affiliations:
  • Department of Electrical and Computer Engineering, Hong Kong University of Science and Technology;Department of Electrical and Computer Engineering, Hong Kong University of Science and Technology;Department of Electrical and Computer Engineering, Hong Kong University of Science and Technology

  • Venue:
  • ICC'09 Proceedings of the 2009 IEEE international conference on Communications
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We extend the idea of hybrid decode-and-forward (DF)/amplify-and-forward (AF) relaying scheme for single relay to the multiple relays scenario and refer to it as dynamic DF/AF scheme. In this dynamic scheme, each relay in the system chooses, packet by packet, between DF and AF dynamically and independently by examining the cyclic redundancy check (CRC) result of the packet. The relays encode the mixed DF/AF signals into a distributed linear dispersion code (DLDC), and then transmit the coded signals to the destination. We show that our proposed dynamic DF/AF scheme outperforms the DF and AF schemes over a wide signal-to-noise ratio (SNR) range. We also propose a more general and flexible form of DLDC, which treats the source-relay link and relay-destination link separately and allows their packet durations to be different. In particular, we assume that the destination has multiple antennas and uses a general two-stage linear MMSE receiver to detect the possibly mixed DF/AF signals, without knowing the forwarding types adopted at the relays. A channel estimator is embedded in the receiver to estimate the source-relay link channel state information (CSI). With this receiver, only partial CSI (the CSI of the combined source-relay and relay-destination link) is required at the destination. The simulation results show that the degradation due to this two-stage receiver is minimal when compared to the optimal, but more complex, ML receiver with full CSI and knowledge of relay forwarding types.