变长分组Banyan网中排队的神经网络调度

Neural Network Based Scheduling for Banyan Network with Variable-Length Packets

提出了适用于变长分组情形下进行调度的系统结构 ;给出了分组调度的具体原则 ;基于所给出的调度原则 ,研究了利用 Hopfield神经网络实现优化调度的方法和在硬件电路上的可实现性 .

The input-output-queuing scheduling problem in Banyan network with variable-length packets was studied, and a scheduling approach based on neural network was proposed. For the proposed approach, a scheduling system structure fit for the variable-length packet case is proposed at first. Then some rules for scheduling are given. At last, based on the rules, an optimal scheduling approach using Hopfield neural network is given. Furthermore, the paper indicated that the approach can be realized by hardware circuit. The simulation results show the effectiveness of the proposed approach.