Capacity of multiple-antenna systems with both receiver and transmitter channel state information

Sudharman K. Jayaweera, H. Vincent Poor

Research output: Contribution to journalLetterpeer-review

109 Scopus citations


The capacity of multiple-antenna systems operating in Rayleigh flat fading is considered under the assumptions that channel state information (CSI) is available at both transmitter and receiver, and that the transmitter is subjected to an average power constraint. First, the capacity of such systems is derived for the special case of multiple transmit antennas and a single receive antenna. The optimal power-allocation scheme for such a system is shown to be a water-filling algorithm, and the corresponding capacity is seen to be the same as that of a system having multiple receive antennas (with a single transmitter antenna) whose outputs are combined via maximal ratio combining. A suboptimal adaptive transmission technique that transmits only over the antenna having the best channel is also proposed for this special case. It is shown that the capacity of such a system under the proposed suboptimal adaptive transmission scheme is the same as the capacity of a system having multiple receiver antennas (with a single transmitter antenna) combined via selection combining. Next, the capacity of a general system of multiple transmitter and receiver antennas is derived together with an equation that determines the cutoff value for such a system. The optimal power allocation scheme for such a multiple-antenna system is given by a matrix-water-filling algorithm. In order to eliminate the need for cumbersome numerical techniques in solving the cutoff equation, approximate expressions for the cutoff transmission value are also provided. It is shown that, compared to the case in which there is only receiver CSI, large capacity gains are available with optimal power and rate adaptation schemes. The increased capacity, is shown to come at the price of channel outage, and bounds are derived for this outage probability.

Original languageEnglish (US)
Pages (from-to)2697-2709
Number of pages13
JournalIEEE Transactions on Information Theory
Issue number10
StatePublished - Oct 2003
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences


  • Adaptive transmission
  • Channel capacity
  • Matrix water-filling
  • Multiple-antenna systems
  • Outage probability
  • Wishart distribution


Dive into the research topics of 'Capacity of multiple-antenna systems with both receiver and transmitter channel state information'. Together they form a unique fingerprint.

Cite this