The Limited Feedback Revolution in Wireless Communication
Robert W Heath Jr.
Wireless communication systems have improved efficiency in several dimensions in the past decade. One prominent change is the sweeping incorporation of multiple antennas, at both the transmitter and receiver, to achieve higher capacity, better quality, and resilience to interference through new modes of operation. A corresponding advance, resulting from increasing flexibility in the transmitter configuration, is algorithms that adapt the transmit strategy to the dynamic propagation environment. The concept of limited feedback lies at the heart of these two fundamental changes in wireless design. Limited feedback is a methodology for obtaining and exploiting propagation channel state information at the transmitter. It uses a finite rate feedback control channel to convey quantized observations of the channel from the receiver to the transmitter. Limited feedback allows the transmitter to adjust how antennas are configured and change other parameters like the transmission rate to maximize performance. This presentation reviews several breakthroughs in limited feedback multiple antenna wireless communication. The connection between limited feedback and quantization on the Grassmann manifold is explained. Then, a new approach for adaptive high resolution limited feedback beamforming is revealed. It leverages temporal correlation in the channel through a new predictive coding framework that reduces feedback rates and/or increases effective resolution. The application of this new framework specifically for high resolution limited feedback interference alignment in interference channels is discussed.