Stochastic Learning Robust Beamforming for Millimeter-Wave Systems with Path Blockage

We introduce a new robust, outage minimum, () () beamforming scheme to combat the random path blockages typical of systems. Unlike state-of-the-art methods, which are of limited applicability in practice due to their combinatorial nature which leads to prohibitive complexity, the proposed method is based on a stochastic-learning-approach, which learns crucial blockage patterns without resorting to the well-known worst-case optimization framework. Simulation results demonstrate the superior performance of the proposed method both in terms of outage probability and effective achievable rate.