Generative Boltzmann Adversarial Network in Manet Attack Detection and QOS Enhancement with Latency

Abstract

Mobile Ad-Hoc Network (MANET) are considered as self-configured network those does not have any centralized base station for the network monitoring and control. MANET environment does not control architecture for routing for the frequent maintenance of topology. The drastic development of Internet leads to adverse effect of development in MANET for different multimedia application those are sensitive to latency. Upon the effective maintenance of the QoS routing route discovery is performed to calculate queue and contention delay. However, the MANET requirement comprises of the complex procedure to withstand the Quality of Service (QoS) with Artificial Intelligence (AI). In MANET it is necessary to compute the MANET attacks with improved QoS with the reduced latency as existing model leads to higher routing and increased latency. In this paper proposed a Generative Boltzmann Networking Weighted Graph (GBNWG) model for the QoS improvement in MANET to reduce latency. With GBNWG model the MANET model network performance are computed with the weighted graph model. The developed weighted graph computes the routes in the MANET network for the estimation of the available path in the routing metrices. The proposed GBNWG model is comparatively estimated with the conventional QOD technique. Simulation analysis stated that GBNWG scheme exhibits the improved performance in the QoS parameters. The GBNWG scheme improves the PDR value by 12%, 41% reduced control packets and 45% improved throughput value.