摘要
In packet switched communication networks, traffic between any source destination (SD) pair is expected to be transmitted over multiple paths. In this paper, we focus on the problem how to optimally distribute the load on a set of paths so that the overall network performance is maximized. This problem is first formulated as a nonlinear programming problem and then we propose a neural network (NN) to solve it. The NN architecture is discussed in detail. At last, we verify the effectiveness of our approach by applying it to a specific network model. The experimental results demonstrate that our method can yield optimal solutions with good stability.
In packet switched communication networks, traffic between any source destination (SD) pair is expected to be transmitted over multiple paths. In this paper, we focus on the problem how to optimally distribute the load on a set of paths so that the overall network performance is maximized. This problem is first formulated as a nonlinear programming problem and then we propose a neural network (NN) to solve it. The NN architecture is discussed in detail. At last, we verify the effectiveness of our approach by applying it to a specific network model. The experimental results demonstrate that our method can yield optimal solutions with good stability.
