HMM Driven White Space for IEEE 802.11
Nayan Dubay, Vishank Patel · International Journal of Engineering Trends and Technology · 2014
Wi-Fi has been adopted more and more extensively in various applications such as wireless patient monitoring in hospitals, herds monitoring in pastures, smart control in home networking and game remote controllers. Most of these applications are performancesensitive so the throughput and packet delivery ratio should be guaranteed for Wi-Fi system to work properly. However, since Wi-Fi operates in unlicensed ISM spectrum. White spaces between frame clusters in time domain observed in Wi-Fi its advantage should be taken to improve the performance of the system. In this paper, we propose a novel approach that ensures high performance of Wi-Fi data trace to time keep the Wi-Fi performance almost unaffected. First we learn an HMM (Hidden Markov Model) based on the data traces of the network. With HMM model we can accurately characterize the dynamic distribution of the durations of white spaces in different times. Second, we estimate the data trace of the Wi-Fi in time domain. Thus based on the HMM model of the white space and the performance estimation we develop a novel frame control protocol called HMM-driven White Space for Wi-Fi.