Block-Partition Sparse Channel Estimation for Spatially Correlated Massive MIMO Systems

Massive multiple-input and multiple-output (MIMO) technique is regarded as one of the most promising technique in the fifth-generation (5G) wireless communication systems. However, accurate channel estimation technique poses a challenge for spatial correlated 3D MIMO systems. Based on the conventional general sparse channel model, sparse channel estimation method using compressive sampling matching pursuit (CoSaMP) algorithm cannot efficiently exploit the block-structure information in the 3D MIMO channel. To fully take advantage of the prior information, in this paper, we propose a block-partition compressive sampling matching pursuit (BP-CoSaMP) algorithm to exploit the block-structure sparsity in angular domain, so that it can further improve channel estimation performance. Simulation results imply that the proposed algorithm not only can reduce pilot overhead, but also can reduce compute complexity.