Research Area:  Wireless Sensor Networks
Intrusion detection system has been termed as events which have surpassed the security measures and makes normal data transfer as vulnerable. The context of wireless sensor nodes with reduced traffic volume has been also susceptible leading to scarce resource utilization by attackers. Thus, as the transfer of data takes place, it is inevitable to preserve the resources of sensors from network undoing. This work contributes to processing time of the flow for particular traffic in the presence of attacker does not impact heavily with unfairness of the traversing flow. The process of conserving network flow from source to sink using Graph Neural Networks is being incorporated. It is achieved with shared weight as a function of traffic flow and assigning the adjacency and degree matrix, which help in associating the traffic structures in a hop by hop basis. The proposed work effectively detects the presence of fairness as the traffic flows in multi-hop communication with resilience to varying links. Simulation works with network simulator-2 has been done for the proposed work to ensure fair resource consumption among sensor nodes.
Author(s) Name:  S. Narasimha Prasad,K. Senthamil Selvan,B. Lakshmi Dhevi
Journal name:  Wireless Personal Communications
Publisher name:  Springer
Paper Link:   https://link.springer.com/article/10.1007/s11277-021-09294-2