Dynamic Estimation of Unsaturated Buffer in Context-Aware M2M WiFi Network

We propose a Particle Filter framework to perform online estimation for an unsaturated buffer of the stations in the Machine to Machine (M2M) WiFi network. The dynamical variation of the traffic affects the performance severely in the M2M WiFi network. The exist researches for analyzing the unsaturated condition in the network is based on the steady-state model, whereas this proposed method is devoted to dynamically estimate the probability distribution of the packet existence in the unsaturated buffer of the stations. The estimation accuracy and effectiveness are evaluated by Root Mean Square Error. The proposed dynamic estimation is more aware of the traffic change in the varying wireless M2M WiFi network compared to the other works by the static analyzing model.