基于主动网络的最优化流量控制

主动网络的分布式计算能力为网络流量控制提供了新途径，将网络的流量控制归结为一个用户收益的全局最优化问题，提出了改进的分布式算法来求解并设计了基于这种算法的主动网络最优化流量控制协议，证明了通过选取适当的收益函数，只需在主动结点中提供FIFO的调度机制就可根据用户需求分配不同的带宽，从而达到了区分服务的目的，仿真结果证明了协议的有效性。

Optimization Flow Control Based on Active Networks

The distributed computation ability of active networks provides a new approach to network flow control. This paper formulates the network flow control as a global optimization problem of user profit. An improved distributed algorithm was proposed to solve the problem, then the active network optimization flow control protocol was designed according to this algorithm. It is proved that by choosing proper user profit functions, only the FIFO scheduling mechanism in active nodes is required to allocate different band according to user's requirement. So it can achieve the goal of differentiation service. The effectiveness of the protocol is proved by the simulation result.