HMM-Based Predictive Power Saving Mechanism in WiMAX

Several studies have showed that network features (e.g., packet interval and packet size) may be well modeled by a hidden Markov model (HMM) with appropriate hidden variables that capture the current state of the network. In this paper, we propose a prediction mechanism on the basis of the HMM model to assist the Power Saving (PS) in WiMAX. In comparison with prior models whose analyses are often with the assumption of Poisson arrival, the prediction-based PS relies no more on that, which means a rich variety of traffic patterns such as heavy-tailed or burst that are often encountered in usual networks can be applied. In addition, our mechanism balances in a more effective way the tradeoff between packet delay and energy consumption, and we find that in fact they can be both improved sometimes. Validation results confirm that the prediction-based PS is of great reduction in energy (about 5% to Instant Message and 5%-25% to HTTP) under the trace driven simulations.