Informatics Publishing Limited

T.R. Ramesh ¹, R. Raghavendra ², Sushiladevi B. Vantamuri ³, R. Pallavi ⁴, Balamurugan Easwaran ⁵

Affiliations
1 Department of Computer Applications, SRM Institute of Science and Technology, Tiruchirappalli, India., IN
2 School of Computer Science and Information Technology, Jain (Deemed-To-Be University), India., IN
3 Department of Computer Science and Engineering, S.G. Balekundri Institute of Technology, India., IN
4 Department of Computer Science and Engineering, Presidency University, India., IN
5 School of Information Technology, Texila American University, Zambia., ZM

Abstract

Vehicular Ad hoc Networks (VANETs) are one of the most advanced transportation networks that have attracted much attention in recent years. The VANETs are characterized by a large number of traffic flows, which make them a good choice for a wide range of applications. However, due to the unique characteristics of the VANET, routing algorithms present a significant obstacle that must be surmounted. In order to improve the communication quality, the research uses federated learning. The research demonstrates the capacity of the model to learn from its previous errors while also delivering more accurate projections using the federated learning. The findings of the simulation demonstrate that the model with a prediction accuracy of 4 packets/s has the highest accuracy when compared to its contemporaries as well as other predicted models. The results show that the proposed method achieves higher rate of accuracy in transmitting the packets with reduced overhead than the other existing methods.

Keywords

Communication Quality, VANET, Federated Learning, Overhead.

Full Text

     

S. Karthigai ¹, Sushiladevi B. Vantamuri ², C. Anupriya ³, Vinod Sesai ⁴, A. Nicholas Daniel ⁵

Affiliations
1 Department of Computer Applications, Navarasam Arts and Science College, IN
2 Department of Computer Science and Engineering, S.G. Balekundri Institute of Technology, IN
3 Department of Computer Science, PSG College of Arts and Science, IN
4 Department of Computer Science and Engineering, BLDEA’s V. P. Dr. P. G. Halakatti College of Engineering and Technology, IN
5 Department of Chemistry, School of Mathematics and Natural Sciences, Mukuba University, ZM

Abstract

Wireless Sensor Networks (WSNs) play a pivotal role in various domains, including environmental monitoring, surveillance, and industrial automation. However, the inherent vulnerabilities in WSNs make them susceptible to various security threats, such as data modification attacks, which can compromise the integrity and reliability of collected sensor data. To address this issue, we propose a novel approach called Residual Recurrent Transfer Learning (RRTL) to enhance the security of WSNs and eliminate data modification in sensor nodes. RRTL leverages the power of deep learning and transfer learning techniques to develop an intelligent and adaptable security framework. The proposed approach trains a deep residual recurrent neural network (RNN) model using a large dataset of normal sensor readings. This model learns the temporal patterns and dependencies in the data, enabling it to identify abnormal sensor readings that might indicate data modification attempts. To evaluate the effectiveness of RRTL, we conducted extensive experiments using a real-world WSN deployment. The results demonstrate that our approach significantly outperforms existing security mechanisms in terms of accuracy, detection rate, and false positive rate. Furthermore, RRTL exhibits robustness against adversarial attacks and dynamic environmental conditions, making it suitable for real-time applications in challenging WSN environments.

Keywords

Securing, Wireless Sensor Networks, Residual Recurrent Transfer Learning, Eliminating, Data Modification, Sensor Nodes.

Full Text