A scoring method improvement of analytic hierarchy process using linear programming technique for vertical handover decision

Abstract

Recently, heterogeneous wireless networks have been integrated to provide mobile users to access services without restriction of different access networks, e.g., cellular systems and WLANs. Thus, the mobile users can obtain the services with higher data rate and quality in anywhere and at any time seamlessly. To support such scenario, it requires a vertical handover (VHO) mechanism that can decide to select the best connection among different access networks from various factors. An analytic hierarchy process (AHP) is one of the methods used in VHO decision by finding the optimal solution based on multiple attribute approach. However, there is a validity acceptance limitation of AHP, due to the score and weight of attributes, or network factors, are assigned by decision makers. This article proposes the VHO decision mechanism that improves a scoring method of AHP by deploying a linear programming (LP) technique instead of human knowledge. The LP is used to find the optimized values among various network factors. Then, the optimized values are taken to replace the human score in the AHP for VHO decision. The received signal strength (RSS), data rate, and throughput are used as the attributes, or variables, related to network factors. The experiment is conducted by simulation to verify the result of proposed VHO decision mechanism between a cellular network and WLANs. The simulation result expresses that the LP can provide the optimized network factors used for scoring in the AHP. It could also be implied that the proposed VHO decision mechanism yields the best network connection based on the optimized values obtained from the LP used instead of human knowledge.