IRO Journals
IRO Journal on Sustainable Wireless Systems

PG-GAT based Channel Estimation for Enhanced OFDM Performance

Open Access
Published: 08 July, 2026
Pages: 160-174
  • Karthik K. , Karthik K.
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
  • Rithika S G. , Rithika S G.
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
  • Praseetha M. , Praseetha M.
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
  • Sowndarya K. Sowndarya K.
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
    Department of Electronics and Communication Engineering, Velammal College of Engineering and Technology (Autonomous), Madurai, India
view PDF
view PDF

How to Cite

K., Karthik, Rithika S G., Praseetha M., and Sowndarya K. 2026. “PG-GAT Based Channel Estimation for Enhanced OFDM Performance”. IRO Journal on Sustainable Wireless Systems 8 (3): 160-74. https://doi.org/10.36548/jsws.2026.3.003.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

OFDM (Orthogonal Frequency Division Multiplexing)
Channel Estimation
GAT (Graph Attention Network)
Physics-Guided Learning
High-Mobility Communication
NMSE (Normalized Mean Squared Error).

Abstract

Accurate channel estimation is essential in enhancing the performance and spectral efficiency of an Orthogonal Frequency Division Multiplexing (OFDM) system especially in high mobility environment. This paper presents a Physics-Guided Graph Attention Network (PG-GAT mini v2) to be used in pilot-based channel estimation through combining the concepts of wireless channel physics and graph attention learning. Multi-head graph attention, residual learning, and Layer Normalization are used in order to optimize initial channel estimates by means of LS interpolation as well as to ensure stability of training procedure. Performance of the method is estimated in terms of NMSE, BER and standard deviation in the conditions of the high mobility Rayleigh fading channel with the carrier frequency of 2 GHz, SNRs from 0 to 30 dB and receiver velocities from 0 to 300 km/h. According to experimental results, PG-GAT mini v2 decreases the total NMSE from 0.7846 to 0.3079, which corresponds to 60.76% performance gain compared to LS interpolation.

References

  1. Nithya, B., D. Brijesh, S. Karan Kumar, and J. Pathmakarthik. "Pilot Based Channel Estimation of OFDM Systems Using Deep Learning Techniques." International Journal of Information Technology 2023, vol. 15, no. 2, 819-831.
  2. Ye, Hao, Geoffrey Ye Li, and Biing-Hwang Juang. "Power of Deep Learning for Channel Estimation and Signal Detection in OFDM Systems." IEEE Wireless Communications Letters 2017, vol. 7, no. 1, 114-117.
  3. Chun, Chang-Jae, Jae-Mo Kang, and Il-Min Kim. "Deep Learning-Based Channel Estimation for Massive MIMO Systems." IEEE Wireless Communications Letters 2019, vol. 8, no. 4, 1228-1231.
  4. Mei, Kai, Jun Liu, Xiaoying Zhang, Kuo Cao, Nandana Rajatheva, and Jibo Wei. "A Low Complexity Learning-Based Channel Estimation for OFDM Systems with Online Training." IEEE Transactions on Communications 2021, vol. 69, no. 10, 6722-6733.
  5. Tian, Guoda, Xuesong Cai, Tian Zhou, Weinan Wang, and Fredrik Tufvesson. "Deep-Learning Based Channel Estimation for OFDM Wireless Communications." 23rd International Workshop on Signal Processing Advances in Wireless Communication (SPAWC), IEEE, 2022, 1-5.
  6. Le Ha, An, Trinh Van Chien, Tien Hoa Nguyen, Wan Choi, and Van Duc Nguyen. "Deep Learning-Aided 5G Channel Estimation." 15th international conference on ubiquitous information management and communication (IMCOM), IEEE, 2021, 1-7.
  7. Ye, Ming, Xiao Liang, Cunhua Pan, Yinfei Xu, Ming Jiang, and Chunguo Li. "Graph Neural Networks Based Channel Estimation for Mmwave Massive MIMO Systems." IEEE Transactions on Vehicular Technology 2025, vol. 74, no. 12, 19420-19435.
  8. Liu, Zishen, Nan Wu, Dongxuan He, Weijie Yuan, Yonghui Li, and Tony QS Quek. "GNN-Assisted Big-AMP: Joint Channel Estimation and Data Detection for Massive MIMO Receiver." IEEE Transactions on Wireless Communications 2025, vol. 24, no. 6, 4631-4646.
  9. Ren, Shuhui, Zhenkun Qiu, and Wuyang Zhou. "Residual Channel Attention Network-Based Channel Interpolation Using Noise2Noise for Massive MIMO-OFDM Systems." 97th Vehicular Technology Conference (VTC2023-Spring), IEEE, 2023, 1-5.
  10. Liao, Yong, Yuanxiao Hua, Xuewu Dai, Haimei Yao, and Xinyi Yang. "Chanestnet: A Deep Learning Based Channel Estimation for High-Speed Scenarios." International Conference on Communications (ICC), IEEE, 2019, 1-6.
  11. Siriwanitpong, Aphitchaya, Kosuke Sanada, Hiroyuki Hatano, Kazuo Mori, and Pisit Boonsrimuang. "Deep Learning-Based Channel Estimation with 1D CNN for OFDM Systems Under High-Speed Railway Environments." IEEE Access 2025, vol. 13, 13128-13142.
  12. Pihlajasalo, Jaakko, Dani Korpi, Mikko Honkala, Janne MJ Huttunen, Taneli Riihonen, Jukka Talvitie, Mikko A. Uusitalo, and Mikko Valkama. "Deep Learning Based OFDM Physical-Layer Receiver for Extreme Mobility." 55th Asilomar Conference on Signals, Systems, and Computers, IEEE, 2021, 395-399.
  13. Hu, Fengye, Zhuang Ling, Tanda Liu, Hailong Li, and Bo Ai. "Wireless Perception of High-Speed Railway Communication: Challenges, Framework, and Future Directions." IEEE Wireless Communications 2024, vol. 31, no. 4, 284-292.
  14. Zhang, Qingmiao, Hanzhi Dong, and Junhui Zhao. "Channel Estimation for High-Speed Railway Wireless Communications: A Generative Adversarial Network Approach." Electronics 2023, vol. 12, no. 7, 1752.
  15. Yang, Junjie, Jie Xu, Yinghui Zhang, Hao Zheng, and Tiankui Zhang. "Graph Attention Network-Based Precoding for Reconfigurable Intelligent Surfaces Aided Wireless Communication Systems." IEEE Transactions on Vehicular Technology 2024, vol. 73, no. 6, 9098-9102.