Network Selection Solution For Heterogeneous Wireless Environment

Abstract

The future generation of wireless network consists of heterogeneous wireless networks that provide uninterrupted services to user when roaming through different access networks. This future generation of wireless network will provide increase in data rate, decrease in end to end latency, energy efficient network environment etc. This is possible due to different enabling technologies i.e. Device to Device Communication, Spectrum Sharing, Enhancement of Quality of Experience in the selection process, Energy Harvest Network and Multi-Radio Access Technologies Association. The selection of network that supports these technologies for different traffic class is done using different decision making techniques. In this work, decision making techniques like Multi Attribute Decision Making algorithms and Game Theory Model (Bankruptcy and Evolutionary Game Model) are used for selection of network in a environment composed of Universal Mobile Telecommunication System, Wireless Local Area Network and Wireless Interoperability for Microwave Access. The different network attributes (available bandwidth, packet delay, packet jitter, packet loss and cost per byte) are taken into consideration for analysis. The analysis is done for conversational, streaming and background traffic class. Also, preference to network attribute`s is introduced using Analytical Hierarchy Process and Standard Deviation Method. The network selection results using Multi Attribute Decision Making algorithm and Game Theory Model are compared and it is found that Game Theory based selection involves learning process therefore consider more than one maximum weighted network attribute for selection whereas MADM algorithm depends on maximum weighted attribute. Further, different decision making techniques are compared on the basis of Relative Standard Deviation. The accuracy of algorithm can be calculated on the basis of difference between ranking values of alternatives.