Reinforcement Optimization to Implement AntNet-Based Routing on Packet Forwarding Engines

AntNet is an adaptive agent-based routing algorithm that imitates the activities of the social insect. In previous works, the efficiency of the AntNet algorithm has been verified in dynamic traffic environments. In spite of this merit, its implementation has been limited till now. There are implementation constraints due to complex arithmetic calculations for determining a reinforcement value and issues of compatibility with network devices. We propose a new optimized AntNet algorithm to overcome these problems. This can be efficiently implemented onto conventional network processors in real networks. The results of the simulation show that the proposed AntNet is more adaptive and effective in the performance of the implementation than the original AntNet.