Research Article
Enhancing Mobile Communication System Security via Neural Cryptography Applications
@INPROCEEDINGS{10.1007/978-3-031-78806-2_9, author={Lela Mirtskhulava and Nana Gulua and Khatuna Putkaradze}, title={Enhancing Mobile Communication System Security via Neural Cryptography Applications}, proceedings={Smart Grid and Innovative Frontiers in Telecommunications. 8th EAI International Conference, EAI SmartGIFT 2024a, Santa Clara, United States, March 23-24, 2024, Proceedings}, proceedings_a={SMARTGIFT}, year={2025}, month={1}, keywords={5G and beyond TPMs synchronization security neural cryptography}, doi={10.1007/978-3-031-78806-2_9} }- Lela Mirtskhulava
Nana Gulua
Khatuna Putkaradze
Year: 2025
Enhancing Mobile Communication System Security via Neural Cryptography Applications
SMARTGIFT
Springer
DOI: 10.1007/978-3-031-78806-2_9
Abstract
The given paper aims to understand how the emerging technologies of 5G and Beyond, and artificial intelligence could affect the security of mobile communication systems and to explore ways to mitigate any potential risks. We analyze the vulnerabilities of 5G and beyond systems to attacks and the potential risks associated with the use of AI in these systems. We optimise a security protocol and systems that can withstand the power of quantum computing and enhance the security of IoT and “5G and beyond” systems. We identify potential areas of regulatory intervention and ensure that AI is used responsibly and transparently in 5G and beyond systems. The proposed method is novel - implementing a type of synchronization mechanism between two Tree Parity Machines (TPMs) using feedback. A feedback mechanism helps to adjust the weights of one TPM based on the outputs of both TPMs on the same input. This is different from traditional methods of training neural networks, which usually involve minimizing a cost function or optimizing weights using some form of gradient descent. The use of TPMs instead of traditional neural networks is also somewhat novel as TPMs are a type of recurrent neural network that has been proposed for use in cognitive modeling. In the training phase of TPMs, the Hebbian learning rule is employed to adjust connection weights between perceptrons.
