Efficient channel estimation for millimeter wave MIMO with limited RF chains

In this paper, an efficient channel estimation scheme is proposed for the point-to-point lens-antenna-array enabled millimeter wave (mmWave) multiple-input multiple-output (MIMO) system with limited number of radio frequency (RF) chains at both source and destination. By exploiting the energy focusing property of lens antenna arrays at both the transmitter and receiver, as well as the multi-path sparsity of mmWave channels, the proposed scheme employs two-way training to achieve energy-based antenna selections at both the source and destination, thus significantly reducing the number of RF chains required. The proposed scheme has low complexity for implementation and yet does not require any prior knowledge of the channel. It is shown by simulations that compared to the traditional channel training method for MIMO systems without lens antenna, the effective data rate achieved by the lens array enabled mmWave MIMO system can be significantly improved, thanks to the energy focusing capability of lens antenna arrays and the matching efficient channel estimation scheme proposed.