A Physical-layer Network Coding Scheme Based on Orthogonal Model Division Multiple Access
Lihui Huang, Haotai Liang, Chen Dong, Xiaodong Xu, Xiaoyi Liu, Weijie Zheng · 2025
Physical-layer network coding (PNC) has been widely studied for its ability to enhance the capacity of wireless communication systems. While deep learning offers potential performance improvements for PNC, it has to consider the symbol bit offset problem again, leading to high bit error rates. This paper proposes a PNC paradigm based on orthogonal model division multiple access (OMDMA-PNC) for two-way relay communication (TWRC) systems. This paradigm combines OMDMA technology with PNC to efficiently process multi-user information. While simultaneously supporting the asynchronous transmission of user information, it significantly overcomes the impact of symbol bit offset. Additionally, this system requires users to upload their knowledge base information to the relay to ensure accurate decoding and reliable transmission, thereby reducing the model management requirements. The experimental results demonstrate that the proposed framework enables communication parties to accurately decode information without requiring other participants’ knowledge base information, while maintaining high robustness to asynchronous transmission.