Channel estimation method for time-frequency double-selection channel of orthogonal frequency division multiplexing system

The invention discloses a channel estimation method for a time-frequency double-selection channel of an orthogonal frequency division multiplexing system. The channel estimation method comprises the following steps: constructing a deep neural network (DNN) and a long-short term memory neural network (LSTM-NN), and forming a DL-CE channel estimation network which is separately used for OFDM symbolestimation with pilot frequency and OFDM symbol prediction without pilot frequency; training the DL-CE channel estimation network to obtain all parameters of the DL-CE channel estimation network; estimating a frequency domain channel, namely acquiring CSI at a pilot frequency by adopting a least square method in an OFDM symbol, and acquiring a channel response of each data symbol position throughthe DNN; and tracking the change of the channel by using the LSTM-NN, and predicting the CSI of the subsequent OFDM symbol without pilot frequency in the time domain. According to the method, the problem of channel estimation