Digital Signal Processing Reference
In-Depth Information
Multi−User, 2 bits/sec/Hz
10
−2
Alamouti MUD without feedback in [5]
Precoding Scheme with perfect feedback in [8]
Our scheme with 8 bits feedback
10
−3
10
−4
10
−5
10
−6
12
14
16
18
20
22
24
Signal to Noise Ratio (dB)
Fig. 4.4
Comparison of our scheme, Alamouti MUD in [
1
] and Precoding scheme in [
13
]for2
users each with 4 transmit antennas and 1 receiver with 2 receive antennas
4.7 Simulation Results
In this section, we provide simulation results that confirm our analysis in the previous
sections. We assume a quasi-static Rayleigh fading channel. The performance of our
proposed scheme is shown in Figs.
4.3
,
4.4
and
4.5
. In each figure, the curves for Users
1 and 2 are identical. In Fig.
4.3
, we consider 2 users each equipped with 2 transmit
antennas and a receiver with 2 receive antennas. We compare our results using QPSK
with the results in [
1
] for the same configuration without channel information at the
transmitter and the results in [
12
] for the same configuration with perfect feedback.
Note that if the feedback is zero in our system (no channel information), we can
pick an identity matrix as our precoder and our transmitter will be the same as the
transmitter in [
1
]. In fact, this backward compatibility is the main reason for using
an Alamouti code. Otherwise, our scheme also works for other full rate space time
codes and all the above derivations are still valid.
In order to illustrate the effect of the number of bits, we provide the performance
with 1, 3, 6, 8 bits feedback, respectively. It can be seen that with 2 receive antennas,
the multi-user detection (MUD) method proposed in [
1
] can cancel the interfer-
ence but only provides a diversity of 2. The scheme proposed in [
12
] with perfect
feedback can achieve interference cancellation and provide a diversity of 4, full diver-
sity. In comparison, using the proposed scheme in this chapter, we can also achieve
Search WWH ::
Custom Search