The Markov-modulated Poisson process (MMPP) cookbook
Performance Evaluation
IEEE Journal on Selected Areas in Communications
Neural approximation of open-loop feedback rate control in satellite networks
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
This paper compares several approaches for dynamic allocation in geo-stationary networks based on DVB-RCS system. Each Satellite Terminal (ST) regularly sends requests to the Network Control Center (NCC) which in turn allocates resource to the users. Unfortunately, this delayed request-assignment makes the dynamic bandwidth allocation very difficult. Simple mechanisms such as a fixed allocation or requests based on the current size of the terminals' queue are compared to predictive methods based on control theory techniques which have been previously proposed. A lower bound is also derived by considering that the actual size of the buffer can be instantaneously known. It is shown that if the traffic is not really bursty, a fixed allocation which implies lighter signalling mechanisms leads to good results. In bursty traffic conditions, simple mechanisms for which the requests correspond to the actual size of the buffer may lead to the best performance results.