Correlated Fading
In general, for correlated fading, the principal eigenvector of E[H
H] may use all transmit antennas.
It is easy to verify this by constructing an example. We leave this as an exercise to the reader. So
for correlated fading, one does gain from using multiple transmit antennas as well as multiple receive
antennas.
10.2
Space-time codes
The key result discussed in the previous subsection motivates the study of channel codes, called space-
time codes to pursue the very high throughput predicted by information theory. As we saw earlier, if
the transmitter knows the channel it is possible to transform it into several parallel non-interfering SISO
channels and the codec technology for SISO channels is well established. However if the transmitter does
not know the instantaneous channel, inherently multi-dimensional codes are required. Codewords are
now long matrices instead of vectors. The optimal decoding complexity of these codewords is exponential
in the number of antennas. Designing these codewords itself is a complex problem and represents a vast
area of research in itself. Some of the approaches explored include treating the transmission from each
antenna as an independent user using conventional scalar codes in conjunction with multiuser detection
techniques at the receiver (layered space time codes). However most of these suboptimal approaches
suffer significant performance penalties.
10.3
Smart Antennas
Smart antennas generally consist of an antenna array combined with signal processing in both space and
time. The spatial processing introduces a new degree of freedom in the system design with enormous
potential to improve performance, including range extension, capacity enhancement, higher data rates,
and better BER performance [53].
The main impediments to high-performance wireless communications are the interference from other
users (cochannel interference) and the intersymbol interference (ISI) and signal fading caused by multi-
path. The cochannel interference limits the system capacity, defined as the number of users which can
be serviced by the system. However, since interference typically arrives at the receiver from different
directions, smart antennas can exploit these differences to reduce cochannel interference, thereby increas-
ing system capacity. The reflected multipath components of the transmitted signal also arrive at the
receiver from different directions, and spatial processing can be used to attenuate the multipath, thereby
reducing ISI and flat-fading. Since data rate and BER are degraded by these multipath effects, reduction
in multipath through spatial processing can lead to higher data rates and better BER performance.
The complexity of spatial domain processing along with the required real estate of an antenna
array make the use of smart antennas in small, lightweight, low-power handheld devices unlikely in
next-generation systems. However the base stations for these systems can use antenna arrays with space-
time processing at the transmitter to reduce cochannel interference and multipath, providing similar
performance advantages as smart antennas in the receiver. An excellent overview of smart antennas can
be found in [53].
242