TY  - JOUR
T1  - Channel Estimation for OFDM Systems Using an Improved SRGAN Architecture
AU  - Jo, Eunwoo 
AU  - Lee, Nayoung 
AU  - Park, Hosung 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2025
DA  - 2025/1/1
DO  - 10.7840/kics.2025.50.1.78
KW  - Channel estimation
KW  - Deep learning
KW  - Neural network
KW  - SRGAN
AB  - It is important to accurately estimate channel values in orthogonal frequency division multiplexing systems.
In the channel estimation (CE), the channel values in the positions of pilot signals can be regarded as low-resolution images, therefore deep learning (DL)-based super-resolution (SR) algorithms can be applied to estimate all the channel values. The existing DL-based CE algorithm with the SR generative adversarial network (SRGAN) achieves good performance and we further improved the neural network architecture, enhancing the CE accuracy with lower complexity.