Context-Aware Handoff Decision for Wireless Access Networks using Bayesian Networks

摘要

One of the challenging tasks of current generation wireless networks consist of supporting service continuity when users roam around different networks such as Wi-Fi, Bluetooth, and Cellular networks. This happens especially when dealing with applications such as multimedia streaming that distribute time continuous data with stringent Quality of Service (QoS) requirements like bandwidth, jitter, delay and packet loss. Session handover can solve this problem by ensuring service continuity through minimization of flow interruptions when users are mobile. Building upon a new Bayesian Belief framework, this paper proposes a context-aware handoff decision process that uses a multi-criteria decision making model to support QoS-based mobility in wireless networks. Using the widely deployed "Hugin Expert" Bayesian belief support tool, we illustrate its deployment in a network environment where mobility decisions are taken to roam around Wi-Fi or Bluetooth networks.