A deep learning-based antenna selection approach in MIMO system

Multiple Input Multiple Output (MIMO) systems offer interesting advantages in terms of throughput and reliability thanks to the diversity and spatial multiplexing technique. However, the system complexity and cost, caused by the RF elements, are still an ongoing challenge. To remedy this, antenna selection technique arises as an alternative to reduce the requirement on RF chains by selecting a subset of the available transmit antennas. This will decrease the overall cost without a significant degradation of the performance. The exhaustive search method, considered as optimal, scans all possible antennas subsets to select the optimal one. It involves a high computational complexity even more with the increase in the number of the antennas. Hence, transferring this huge computation to offline system seems to be the only way to benefit from this technique. In this paper, we propose to analyse the performance of the deep neural network to construct a deep learning decision server to assist the MIMO system for making intelligent decision for antenna selection that maximize either the capacity or the energy efficiency. Extensive simulations show that the designed network can suitably resolve the antennas selection problem with high accuracy and robustness against imperfect channel estimation while leading to low online computational time compared to different optimization-driven decision making method.