Adaptive feedback for MIMO beamforming systems

Multiple-input multiple-output wireless systems can achieve significant diversity and array gain by using transmit beamforming and receive combining techniques. This however, requires channel information at the transmitter in the form of a beamforming vector. The literature describes methods to quantize the beamforming vector and feedback that information to the transmitter assuming a particular channel distribution, commonly uncorrelated Rayleigh. When the channel is correlated, these quantization strategies result in considerable degradation of signal to noise ratio. In such cases, the spatial and temporal correlations in the beamforming vector are exploited to develop quantization schemes that adapt to the distribution of the channel. Two scenarios are considered - (i) when the feedback rate is fixed and distortion is minimized (ii) when distortion is constrained and rate is minimized. Both the algorithms largely outperform quantization schemes that are based on uncorrelated Rayleigh fading assumption.