Channel Estimation Algorithm Combining Deconvolution and Dilated Convolution

 As the key of wireless communication, channel estimation has become a research hotspot in related fields in recent years. In this paper, a deep learning method based on deconvolution network and dilated convolution network was proposed to solve the problem that the performance of traditional channel estimation algorithm in orthogonal frequency division multiplexing (OFDM) system was difficult to meet the communication requirements of complex scenes and was greatly affected by noise. In this method, a lightweight deconvolution network was constructed by using the correlation of the channel, and a few deconvolution operations were used to realize the channel interpolation and estimation step by step, which achieved the channel estimation well with low complexity. In order to improve the estimation performance, a dilated convolution network was constructed to suppress the channel estimation noise and improve the accuracy of channel. Simulation results show that the proposed deep learning method based on deconvolution and dilated convolution has lower estimation error with low complexity than traditional methods under different SNR conditions.