Quasi-Static and Time-Selective Channel Estimation for Block-Sparse Millimeter Wave Hybrid MIMO Systems: Sparse Bayesian Learning (SBL) Based Approaches

A Spatial–Temporal Subspace-Based Compressive Channel Estimation Technique in Unknown Interference MIMO Channels

“…The first seven items 8 in Table III are the complexity for Algorithm 2 f ℓ1 ICA (·|A [n] ) performed at Step 5 in Algorithm 3, whereas the last item describes the complexity for the AAD algorithm performed at Step 7 of Algorithm 3. As observed from Table III, the complexity for the ℓ1 iST is dominated by that needed to update the CCM 9…”

“…A sparse Bayesian learning (SBL)-based algorithm [9] improves estimation accuracy over the OMP by exploiting the block sparsity property [9] commonly observed for the angular domain channel gain vectors in certain L M measurements. Note that the estimation algorithms [8], [9] do not consider the MAI problem directly. The signal model in both studies is formulated as a collection of received single-input multi-output (SIMO) signals from each transmission (TX) beam.…”

A Sequential Subspace Method for Millimeter Wave MIMO Channel Estimation

“…Hence, the overall complexity of the proposed SASE algorithm is O(m 2 N r + LDN r ) = O(LDN r ). The computational complexities of benchmarks, i.e., the angle estimation methods OMP [8], SBL [9], and ACE [17] along with the subspace estimation methods Arnoldi [16], SD [10], and MF [11] are compared in Table II, where K denotes the number of channel uses. For a fair comparison, when comparing the computational complexity, we assume the number of channel uses, K, is equal among the benchmarks.…”

“…By exploiting the fact that mmWave propagation exhibits low-rank characteristic, recent researches formulated the channel estimation task as a sparse signal reconstruction problem [8], [9] and low-rank matrix reconstruction problem [10]- [15]. By using the knowledge of sparse signal reconstruction, orthogonal matching pursuit (OMP) [8] and sparse Bayesian learning (SBL) [9] were motivated to estimate the sparse mmWave channel in angular domain. Alternatively, if the channel is rank-sparse, it is possible to directly extract sufficient channel subspace information for the precoder design [10], [11], [16].…”

“…Though the sparse signal reconstruction [8], [9] and matrix completion [10], [11] techniques can reduce the channel use overhead compared to traditional beam alignment techniques, the training sounders of these techniques [8]- [11] are predesigned and high-dimensional, which leads to the fact that these works suffer from explosive computational complexity as the size of arrays grows. To reduce the computational complexity, the adaptive training techniques have been investigated in [4], [16], [17], where the training sounders can be adaptively designed based on the feedback or two-way training.…”