In order to speed up the state convergence in the multi-Agent formation control process, a formation control method based on multi-hop network technology was proposed. Firstly, the relative velocity deviation between Agents of Multi-Agent System (MAS) was introduced into the control protocol. Then, the absolute displacement deviation between Agents and the standard displacement was introduced. Finally, the multi-hop network technology was applied in the communication topology, so more information was passed around and each Agent enlarges its "available" neighborhood. A six-Agent formation control example was used to verify the proposed protocol. The simulation results show that the proposed control method can make the system build up the specified formation; and the time required for state convergence is reduced by nearly 10 seconds compared with the control method that does not take into account multi-hop network technology, which verifies that the proposed method is more efficient.

With the advent of multimedia service, the traditional fair traffic service mode can no longer satisfy the need of their quality of service (QoS). DiffServ scheme can differentiate different applications by setting priority for each flow and policing traffic at the edge of network. Classifier is one of the parts in implementation of DiffServ. This complement of DiffServ four-tuple classifier is based on a general-purpose network processor IXP2800 and has been tested under workbench - Intel simulation environment.