Dynamic Traffic Allocation Scheme for Optimum Distribution in Heterogeneous Networks

We present an explicit adaptive traffic allocation scheme for optimum distribution in heterogeneous networks, where data traffic is broken down into multiple sub-flows, such that the aggregate of these multiple sub-flows at the terminal are merged into the original flow. For simplicity, parallel data transmission through inter-network cooperation is modeled as a sequent M/M/1 queue system. We aim to minimize the queue delay cost and maximize the probability of packets meeting a given delay constraint. The optimization problem has been achieved through the use of Lagrange multiplier. Depending on the utility the formulation done in this work is shown to optimally divide the rate of incoming data traffic into different paths. The simulation results show the convergence of the proposed scheme and significant improvement in performance comparing with other schemes.