SmartHO: mobility pattern recognition assisted intelligent handoff in wireless overlay networks

Abstract

Handoff management providing continuous mobility services is important for end-users in wireless overlay networks. The existing studies focus on improving the handoff performance with little attention to determine the current mobility pattern to provide different advanced services for end-users, and the handoff process is not intelligent as well. Due to the randomness and fuzziness of human mobility, predicting the mobility is still an open issue. In this paper, a mobility pattern recognition assisted intelligent handoff (SmartHO) is proposed. By exploring the regularity and rationality in the seemingly random daily itinerary and analyzing the typical mobility patterns, a spatial---temporal mobility pattern model is firstly constructed, and the mobility pattern recognition algorithm based on the joint space---time correlation then is designed. In different handoff cases, the specific handoff optimization strategies corresponding to the mobility patterns are employed to make the handoff process more intelligent. With the configurations from the authors real life, the simulations conducted on the NS-2 platform show that SmartHO outperforms FHMIPv6 in the critical handoff performance metrics.