[Paper Review] Network Selection Decision Based on Handover History in Heterogeneous Wireless Networks

Mohamed Lahby, Cherkaoui Leghris+1 more|arXiv (Cornell University)|Jun 7, 2012

IPv6, Mobility, Handover, Networks, Security12 references21 citations

TL;DR

This paper proposes a novel network selection decision framework in heterogeneous wireless networks using a hybrid FAHP-TOPSIS approach that incorporates handover history to improve decision accuracy. By weighting criteria via Fuzzy Analytical Hierarchy Process and ranking networks via TOPSIS, the method reduces unnecessary handovers and enhances QoS, with simulations showing improved selection reliability and reduced handover frequency compared to conventional methods.

ABSTRACT

In recent years, the mobile devices are equipped with several wireless interfaces in heterogeneous environments which integrate a multitude of radio access technologies (RAT's). The evolution of these technologies will allow the users to benefit simultaneously from these RAT's. However, the most important issue is how to choose the most suitable access network for mobile's user which can be used as long as possible for communication. To achieve this issue, this paper proposes a new approach for network selection decision based on Saaty's Fuzzy Analytical Hierarchy Process (FAHP) and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS). The FAHP method is used to determine a weight for each criterion, and the TOPSIS method is applied to rank the alternatives. Simulation results are presented to illustrate the effectiveness of our new approach for network selection.

Motivation & Objective

• To address the challenge of selecting the most suitable network in heterogeneous wireless environments with multiple radio access technologies (RATs).
• To improve network selection decisions by incorporating historical handover data to enhance decision accuracy and reduce unnecessary handovers.
• To develop a decision-making framework that dynamically weights network selection criteria based on user mobility patterns and network performance.
• To evaluate the effectiveness of the proposed method in minimizing handover-related disruptions and maintaining stable connectivity.

Proposed method

• The Fuzzy Analytical Hierarchy Process (FAHP) is used to determine the relative weights of network selection criteria such as signal strength, bandwidth, and delay, based on expert judgment and fuzzy logic.
• The Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) ranks available networks by measuring their proximity to the ideal solution using the weighted criteria.
• Handover history is integrated into the decision model to reflect past network performance and user mobility behavior, improving prediction accuracy.
• The framework processes real-time network measurements and historical handover events to update weights and rankings dynamically.
• A simulation environment is implemented to evaluate the proposed method under varying mobility and network load conditions.
• The method is compared against conventional single-criterion or non-history-based multi-criteria approaches to assess performance gains.

Experimental results

Research questions

• RQ1 How can handover history be effectively leveraged to improve network selection decisions in heterogeneous wireless networks?
• RQ2 To what extent does integrating historical handover data reduce unnecessary or ping-pong handovers?
• RQ3 How does the hybrid FAHP-TOPSIS model compare to traditional network selection methods in terms of decision accuracy and stability?
• RQ4 What is the impact of dynamically weighted criteria on network selection performance and user QoS?

Key findings

• The proposed FAHP-TOPSIS model with handover history integration significantly reduces the frequency of unnecessary handovers compared to conventional methods.
• The method improves network selection accuracy by incorporating mobility patterns and past network performance into the decision-making process.
• Simulation results demonstrate enhanced QoS metrics, including reduced handover blocking and improved connection stability.
• The use of fuzzy logic in FAHP allows for more robust handling of uncertain and imprecise network quality indicators.
• The hybrid approach outperforms single-criterion and non-history-based multi-criteria methods in terms of decision consistency and reliability.