AMONET : a method for detecting and mitigating the data rate degradation due to interference over wireless NETworks

Abstract

Current optimizations in the Wi-Fi networks require an accurate but low-cost interference detection and modeling method in wireless network. In this thesis, we present an extensive measurement study on the interference detection in dynamic environment. We find out that some previous methods can hardly be accurate to detect all the interference cases. We define a new kind of problem caused by the collision induced interference in the Wi-Fi networks, named the Data Rate Degradation by Interference (DRDI), which causes a wireless link to degrade its physical data rate in order to endure interference. Through mathematical modeling and experiments, we show that the DRDI can cause dramatic throughput decrease in the Wi-Fi networks. Moreover, to address the above problem, we present A Method for detecting and mitigating the data rate degradation due to the interference Over wireless NETworks (AMONET), which is helpful to detect and mitigate either the Hidden Terminal problem or the DRDI in the wireless networks. We integrate the AMONET with a centralized scheduling system and demonstrate that the interference relations detected by AMONET can benefit the centralized scheduling system more than that of the previous methods.