Enhancement of Physical Layer Security for Relay Aided Multihop Communication Using Deep Transfer Learning Technique

In order to meet the skyrocketing demand for wireless connectivity, fifth-generation (5G) technologies have lately gained a lot of interest. Device-to-device (D2D) communication, that lets nearby users connect without sending information over the network’s base station (BS), has generated a lot of attention since it can relieve 5G networks of their burdensome traffic. Multihop D2D communication provides extended coverage, improved reliability and enhanced connectivity in disaster relief scenarios where emergency responders can maintain critical communication even when the cellular base stations are non-operational. Unfortunately, there are various risks to D2D communications, including jamming, data manipulation, and privacy violations. By just using the features of wireless channels, physical layer security (PLS) is regarded as an innovative method for enhancing wireless security. PLS focuses on the inherent properties of the wireless signal, such as signal strength, interference patterns, and channel fading, to secure communication, and adds additional protection for existing security measures. Wireless networks should use deep transfer learning for physical layer security that applies a previously trained D2D communication system model to a new wireless environment. The selective transfer of relevant knowledge from pre-trained model reduces the risk of overfitting and in turn minimizes interference caused by task-specific noise. The proposed work analyzes PLS for underlay multihop D2D communication using deep transfer learning. The performance parameters such as cellular outage probability (COP), the secrecy outage probability (SOP), and the probability of nonzero secrecy (PNSC) are evaluated to enhance physical layer security for reliability, transmission confidentiality, and receiver channel capacity. Deep transfer learning is utilized to improve PLS by leveraging knowledge gained from related tasks to manage interference and enhance the secrecy rate.