ناديا مير محمدي

Full-duplex systems, recognized as one of the advanced technologies in the field of telecommunications, enable the simultaneous transmission an‎d reception of signals over the same frequency ban‎d. These systems are particularly important in advanced communication applications such as fifth-generation (5G) networks an‎d beyond. One of the major challenges in such systems is **self-interference**, which arises due to the concurrent transmission an‎d reception of signals, leading to interference, degradation of the received signal quality, an‎d a reduction in overall system capacity. Previous studies in this area have employed various approaches, including filtering an‎d signal processing techniques, to mitigate the effects of self-interference. However, these methods generally suffer from limitations in scalability an‎d performance, especially under complex channel conditions an‎d at higher frequency ban‎ds. In recent years, artificial intelligence-based approaches—particularly deep learning an‎d neural networks—have demonstrated superior performance in addressing these challenges, showing strong capability in improving system operation within complex an‎d dynamic environments. In this study, self-interference cancellation in full-duplex systems is investigated using advanced neural network models an‎d deep learning techniques, including multi-head attention networks. Furthermore, novel neural neuromodulation methods are incorporated to enhance the self-interference cancellation process. The results obtained from the implementation of these techniques indicate a significant improvement in system performance. Specifically, the proposed model achieves approximately a 1.5 dB additional reduction in self-interference compared to previous models. This research can serve as a foundation for the development of advanced technologies in next-generation communication networks.

لفو دوسويي همزمان , پردازش سيگنال با شبكه عصبي , سيستم دو طرفه كامل

Full-duplex systems , Signal Processing with Neural Networks , Self-Interference Cancellation

Nadia Mirmohammadi

Dr. Abolfazl Falahati