Effect of feedback prediction on OFDMA system throughput

  • Authors:
  • Mohammad Abdul Awal;Lila Boukhatem

  • Affiliations:
  • Univ. Paris-Sud, CNRS, Orsay, France;Univ. Paris-Sud, CNRS, Orsay, France

  • Venue:
  • Proceedings of the 2009 International Conference on Wireless Communications and Mobile Computing: Connecting the World Wirelessly
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

In wireless communication systems, adaptive modulation and coding (AMC) is used to improve the downlink (DL) spectral efficiency by exploiting the underlying channel condition. However, AMC also reduces the uplink (UL) throughput because of feedback (FB) overhead which is used to carry channel information. Opportunistic mechanisms provide a good solution to reduce the FB overhead but in each scheduling slot, they are designed to serve a very few mobile stations (MSs) only out of a large number of simultaneously active MSs. In multichannel multiuser systems, the opportunistic FB scheme will have no effect in FB reduction if all the MSs need to be scheduled in each scheduling slot. To reduce the FB overhead, we propose a base station (BS) side FB prediction strategy that reduces FB overhead for MSs experiencing any channel condition. We use recursive least square (RLS) algorithm which adapts with SNR variations over time and provides predicted SNR with minimized error. With numerical analysis and simulations, we show that this FB reduction mechanism can be used to increase the UL throughput with non-noticeable DL degradation.