Multiple antennas at the transmitter and the receiver can be used either 
as a source of diversity or as a means to increase achievable data rates. 
The focus of this work is on the impact of correlation in the spatial 
domain on the diversity and multiplexing aspects of limited feedback 
systems. On the diversity aspect, we show that beamforming-like 
signaling, or in other words no coding across time, achieves a significant 
fraction of the diversity benefits of multi-antenna systems. We also 
establish conditions under which the low-complexity beamforming approach 
is information theoretically optimal. On the multiplexing aspect, we 
first describe the regime where any type of feedback, limited or otherwise, 
can result in substantial performance improvement. In this regime, we 
then show how to leverage statistical information to construct limited 
feedback codebooks of low-complexity that can bridge the gap in 
performance to the perfect channel information benchmark.