A CNN-GRU based Approach for Strong In-Band Interference Mitigation in Wireless Communication Systems
Yanru Wang, Zhiyong Luo, Xiti Wang · 2023
With the advent of the sixth generation (6G), the scarcity of spectrum resources and interference issues are receiving increasing attention. In order to address the issue of decreased communication reliability caused by the presence of interference signals, various methods have been proposed in previous research. However, these traditional methods often involve manual parameter selection, lacking automation and intelligence. And in the face of strong in-band interference, they are unable to effectively achieve interference mitigation. Research has shown that deep learning can be used for signal processing. But most of them are used for handling frequency modulated continuous wave (FMCW) radars and rarely applied to address strong in-band interference issue in wireless communication systems. Therefore, we leverage the advantages of convolutional neural network (CNN) and recurrent neural network (RNN) with gated recurrent unit (GRU) in processing sequential data, and propose a fusion network to achieve strong in-band interference mitigation in wireless communication systems. Through experiments, it has been demonstrated that our proposed approach exhibits better suppression performance in strong interference scenarios, reducing the bit error rate by over 75 percent. Additionally, in the presence of various types of interference, our approach is still capable of effectively reducing the bit error rate and enhancing the reliability of the communication system.