A DNN-Based Channel Estimation Scheme for PMCH in LTE-Based 5G Terrestrial Broadcast System

This paper proposes a deep neural network (DNN)based channel estimation scheme for the physical multicast channel (PMCH) in Long-Term Evolution (LTE)-based fifthgeneration (5G) terrestrial broadcast system. In Release 16, rooftop receiver scenarios in high power high tower (HPHT) and medium power medium tower (MPMT) are supported for LTEbased 5G terrestrial broadcast. Therefore, channel environments between transmitters and receivers may have a very long delay spread which induces severe frequency selectivity. To improve the performance of a channel estimation under the very long delay spread environments, the DNN-based channel estimation scheme is proposed. Simulation results show that the proposed DNNbased channel estimation scheme outperforms the conventional scheme, such as linear and discrete Fourier transform-based interpolation schemes under mobile channels with a very long delay spread. Furthermore, it shows that the mobile speed used for the generation of the training data has little effect in training the proposed DNN- based scheme.