Abstract: Channel estimation of millimeter wave communication needs large system load. In order to reduce the load, a low-rank and sparse feature of the wireless channel is combined, and a channel estimation algorithm framework based on non-convex low-rank approximation is proposed. Aiming at the large computation of the channel model-based dictionary learning algorithm, a dictionary learning algorithm for deep neural network channel feature classification is designed. The simulation shows that the average square error of the proposed method is better than the channel model-based dictionary learning method, the channel estimation method under the Bayesian framework, and the compressed sensing channel estimation method under the specific city microcellular channel model. The signal-to-noise ratio required by the algorithm is the lowest when the mean square error is the same. The number of pilots required is lower than the above three methods.

Key words: massive MIMO, non-convex, deep neural network, channel state information, dictionary learning