Multiscale admission control algorithm for statistical QoS guarantee

Statistical properties of the traffic were calculated to estimate the burst level through online measurement for selecting proper traffic model.In order to guarantee traffic QoS,the effective envelope of self-similarity and multiracial traffic was derived using the transformation theory between the effective bandwidth and effective envelope.Then,QoS parameters were obtained using network calculus to perform admission decision;furthermore appropriate admission time was chosen to reduce the proposed algorithm’s impact on QoS performance.The simulations prove that the traffic model can be selected accurately,statistical delay can be ensured and utilization efficiency of network is improved.