Performance analysis of MIMO ad-hoc networks with quantized beamforming and imperfect channel state information

This paper deals with the network spectral efficiency/capacity analysis of multiple-input multiple-output (MIMO) ad hoc network with k simultaneous communicating transceiver pairs in two scenarios; quantized beamforming for achieving finite rate feedback and imperfect channel state information at the transmitter (CSIT). In the quantized beamforming, each receiver chooses the best beamforming vector from the code book that has been designed using the concept of Grassmannian beamforming. Each receiver sends the label of the best beam-forming vector to the transmitter through finite rate limited feedback. Code book is shared between each transmitter-receiver pair to prevent the decoding of other users information. An expression for spectral efficiency is derived for imperfect CSIT, with the assumption that each transmitter is equipped with t antennas and each receiver with r antennas and each receiver implements single-user detection. In this work, the capacity loss due to quantized beamforming as a function of codebook size and the capacity/SINR loss due to imperfect CSIT as a function of channel estimation error are also derived and analyzed. The expressions are examined to provide interesting insights in the high and low SINR regime. Given a target spectral efficiency, numerical and simulation results are presented to confirm the closeness between them.