论文部分内容阅读
Performance evaluation shows that Hierarchical Mobile IPv6 (HMIPv6) cannot outperform standard Mobile IPv6 (MIPv6) in all scenarios. Thus, adaptive protocol selection under certain circumstances is desired. Moreover, it is necessary to balance the load among different Mobility Anchor Points (MAPs). This paper proposes an efficient three-level hierarchical ar- chitecture for mobility management in HMIPv6 networks, in which a mobile node (MN) may register with either a higher/lower MAP or its home agent according to its speed and the number of connecting correspondent nodes (CNs). An admission control algorithm and a replacement algorithm are introduced to achieve load balancing between two MAP levels and among different MAPs within the same MAP level. Admission control is based on the number of CNs but not MNs that an MAP serves. In case there is no capacity for an MN, the MAP chooses an existing MN to be replaced. The replaced MN uses the MAP selection al- gorithm again to choose another mobility agent. Simulation results showed that the proposed scheme achieves better performance not only in reducing the signaling overhead, but also in load balancing among different MAPs.
As such, it is necessary to balance the load on different Mobility Anchor Points (MAPs) . This paper proposes an efficient three-level hierarchical architecture for mobility management in HMIPv6 networks, where a mobile node (MN) may register with either a higher / lower MAP or its home agent according to its speed and the number of connecting correspondent nodes (CNs). An admission control algorithm and a replacement algorithm are introduced to achieve load balancing between two MAP levels and among different MAPs within the same MAP level. Admission control is based on the number of CNs but not MNs that an MAP serves . The replaced MN uses the MAP selection al- gorithm again to choose anoth er mobility agent. Simulation results showed that the proposed scheme achieves better performance not only in reducing the signaling overhead, but also in load balancing among different MAPs.