A Potent Model for Unwanted Traffic Detection in QoS Network Domain

Unwanted traffic injection which amplifies the traffic loading and exhausts network resources is counted as network security threat. Despite of the numerous protection systems, this threat is still lacking powerful remedy till day. This paper proposes a potent model for unwanted traffic detection in edge-to-edge QoS network domain. It is designed to detect service violations and bandwidth theft in the network. The ingress and egress edges at the ISP domain are used to feed Risk Management Unit (RMU) by information about delay, jitter and throughput. RMU computes users' ratios to verify the violations in the Service level Agreement (SLA) and to identify who is behind these violations. The simulation results indicate that our model is capable to detect the violations in SLA. Furthermore it is precise for dropping the unwanted traffic without exceeding that to the wanted traffic.