Subspace Estimation and Decomposition for Large Millimeter-Wave MIMO Systems

摘要: Channel estimation and precoding in hybrid analog-digital millimeter-wave (mmWave) MIMO systems is a fundamental problem that has yet to be addressed, before any of the promised gains can harnessed. For matter, we propose method (based on well-known Arnoldi iteration) exploiting channel reciprocity TDD sparsity channel’s eigenmodes, estimate right (resp. left) singular subspaces channel, at BS MS). We first describe algorithm context conventional systems, derive bounds error presence distortions both MS. later identify obstacles hinder application such an architecture, address them individually. In view fulfilling constraints imposed by further iterative for subspace decomposition, whereby above estimated subspaces, are approximated cascade analog digital precoder/combiner. Finally, evaluate performance our scheme against perfect CSI, fully case (i.e., equivalent system), conclude similar achieved, especially medium-to-high SNR (where gap less than 5%), however, with drastically lower number RF chains ( ${\sim}4$ 8 times less).