Massive MIMO Adaptive Modulation and Coding Using Online Deep Learning Algorithm

The letter describes an online deep learning algorithm (ODL) for adaptive modulation and coding in massive MIMO. is based on a fully connected neural network, which initially trained the output of traditional then incrementally retrained by service feedback its output. We show advantage our solution over state-of-the-art Q-learning approach. provide system-level simulation results to support this conclusion various scenarios with different channel characteristics user speeds. Compared OLLA, shows 10% 20% improvement throughput full-buffer case.