15-08-2023 дата публикации
Номер: CN116599800A
Принадлежит:
The invention discloses an OFDM (Orthogonal Frequency Division Multiplexing) channel estimation method based on a deep convolutional network, which belongs to the technical field of OFDM channel estimation and comprises the following steps: step 1, a communication receiving end performs data processing on received data Y; step 2, establishing an OFDM channel estimation model based on a deep convolutional network, specifically, firstly establishing a channel estimation network based on DenseBlock, namely a DenseBlock-alpha network, then performing transverse expansion on the DenseBlock-alpha network to obtain an MNet-L6 network, and finally performing network training on the MNet-L6 network based on a Keras deep learning framework; and step 3, loading the OFDM channel estimation model based on the deep convolutional network to an OFDM system receiving end to complete a channel estimation task. According to the OFDM channel estimation method based on the deep convolutional network, the DenseBlock ...
Подробнее