Assuming that channel state information (CSI) can be available at the transmitter, we provide a simplified representation of multi-input multi-output (MIMO) systems and derive a new precoder which maximizes the smallest distance between the received symbols. When the number of transmit antennas, n/sub T/, is greater than the number of transmit data streams, b, R.W. Heath and A. Paulraj (see Proc. ICC 2001) have proposed a "hard" antenna selection (or switch precoder) which consists in choosing the best b (among n/sub T/) antennas of the transmitter according to a criterion based on the minimum distance. Using the same criterion, we propose a "soft" (or linear) precoder that performs power allocation among the transmit antennas. Comparisons in term of bit error rate (BER) between switch and linear precoders are performed considering b=2 independent data streams, a QPSK modulation and the maximum likelihood (ML) receiver.
[1]
Petre Stoica,et al.
Generalized linear precoder and decoder design for MIMO channels using the weighted MMSE criterion
,
2001,
IEEE Trans. Commun..
[2]
Robert W. Heath,et al.
Antenna selection for spatial multiplexing systems based on minimum error rate
,
2001,
ICC 2001. IEEE International Conference on Communications. Conference Record (Cat. No.01CH37240).
[3]
M. J. Gans,et al.
On Limits of Wireless Communications in a Fading Environment when Using Multiple Antennas
,
1998,
Wirel. Pers. Commun..
[4]
Emre Telatar,et al.
Capacity of Multi-antenna Gaussian Channels
,
1999,
Eur. Trans. Telecommun..
[5]
Olivier Berder,et al.
Minimum BER diagonal precoder for MIMO digital transmissions
,
2002,
Signal Process..