Context-Awareness Handoff Planning in Heterogeneous Wireless Networks

摘要

This paper addresses a context-awareness handoff planning mechanism in heterogeneous wireless networks for seamless ubiquitous access. The proliferation of digital devices with communication capability heralds the era of ubiquitous computing, as predicted by Mark Weiser. Mobile devices with multi-mode interfaces can access heterogeneous wireless networks, such as wireless local area network (WLAN), worldwide interoperability for microwave access (WiMAX) network and third generation (3G) network. In heterogeneous wireless networks, different network provides different network capability, transmission range, data rate, access cost, security policy and quality of service. Therefore, a ubiquitous access aims to provide users intelligent human-centric context-aware handoff mechanism in heterogeneous wireless networks at anytime anywhere is gradually become an important issue. Optimal handoff planning in heterogeneous wireless networks considers the needs of users, accessing service and the surrounding context. However, most of existing handoff mechanisms only considered few contexts that lead to not able to select the most suitable network for individual user. In this paper, we propose a human-centric context-aware solution to access most appropriate network that satisfies individual user's requirement, fulfills seamless data transmission and reduces handoff delay. We develop two integrated approaches for context-awareness handoff planning mechanisms, namely MADM approach and GA approach. The MADM approach contains initial process and decision process. Initial process weights context by Analytic Hierarchy Process (AHP) method while decision process selects the most appropriate network by Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The GA approach uses Genetic Algorithm to minimize handoff under given QoS constraints. Performance of proposed mechanisms is evaluated under different services with different QoS considerations. Our simulation results show that both mechanisms are able to derive a proper network handoff plan which guarantees QoS requirements and reduces delay, jitter, and number of handoffs.