Fast Codebook-Based Beamforming Training For Mmwave Mimo Systems With Subarray Structures

Liang Zhou,Yoji Ohashi
DOI: https://doi.org/10.1109/VTCFall.2015.7390946
2015-01-01
Abstract:Beamforming (BF) with phased arrays is a critical and natural solution for millimeter-wave (mmWave) wireless communications due to the increased signal propagation attenuation. In this paper, we propose fast and efficient codebook-based BF training to estimate antenna weight vectors (AWVs) for next generation mmWave WLANs/WPANs or the fifth generation (5G) wireless systems with spatial multiplexing (SM). In the proposed mmWave SM systems, the transmit and receive antenna arrays are grouped into a number of subarrays with a predetermined subarray separation based on the optimal placement in line of sight (LOS) environment. Unlike the stochastic gradient algorithm (SGA) based blind BF training which has high complexity, we propose codebook-based initial coarse BF training in inner-subarray beamforming, and BF refinement based on the selection criteria of inner-subarray beamforming or inter-subarray spatial multiplexing to reduce training complexity greatly. Complexity analysis and simulations show that the proposed techniques achieve a small performance loss compared with the benchmark, and outperform the SGA at different geometrical placements, even using much less BF training time and smaller feedback overhead.
What problem does this paper attempt to address?