Research Article
A Study on Vertical Handoff for Integrated WLAN and WWAN with Micro-Mobility Prediction
@INPROCEEDINGS{10.1109/BROADNETS.2006.4374352, author={Pin-Han Ho and Ying Wang and Fen Hou and Sherman Shen}, title={A Study on Vertical Handoff for Integrated WLAN and WWAN with Micro-Mobility Prediction}, proceedings={3rd International ICST Conference on Broadband Communications, Networks, and Systems}, publisher={IEEE}, proceedings_a={BROADNETS}, year={2006}, month={10}, keywords={Markov chain motion prediction pre-handoff vertical handoff wireless LAN wireless WAN}, doi={10.1109/BROADNETS.2006.4374352} }- Pin-Han Ho
Ying Wang
Fen Hou
Sherman Shen
Year: 2006
A Study on Vertical Handoff for Integrated WLAN and WWAN with Micro-Mobility Prediction
BROADNETS
IEEE
DOI: 10.1109/BROADNETS.2006.4374352
Abstract
The integration of the third generation (3G) wireless wide-area networks (WWAN) and the IEEE 802.11 wireless local-area networks (WLAN) has drawn much attention from both industry and academia. To achieve an effective and efficient integration between the two networks with very different characteristics, nonetheless, is still an open issue. One of the challenges is to provide an integrated strategy for achieving a seamless vertical handoff of mobile users roaming between the two network domains where the delay, delay jitter, and packet loss probability can be well controlled. This paper is committed to study a two-step vertical handoff mechanism based on linear regression which is further modeled through an analytical approach. The proposed vertical handoff scheme is characterized by its adaptability to different quality of service (QoS) requirements by manipulating a threshold on the expected handoff instant. A new approach of mobility analysis is introduced to facilitate modeling of vertical handoff delay by taking advantage of Markov chain techniques. We have seen merits gained in our scheme in achieving a good trade-off between the average handoff delay and the multi-tunnel time by manipulating a threshold value, where both analytical and simulation results prove the effectiveness.