A transmission and scheduling scheme based on W-learning algorithm in wireless networks

摘要

In this paper, we propose a transmission and scheduling scheme based on W-learning algorithm used in wireless network. With the introduction of W-learning algorithm, we can transmit packets intelligently in wireless networks, namely, reduce package lose under the premise of energy saving. We offer our own system model based on Markov decision process (MDP), in order to prove the efficiency of our scheme, we compare the simulation transmission results of our scheme with the results of several heuristic algorithms and the approximately optimal result of successive approximation method. A state aggregate method and action set reduction scheme are also used in this paper to reduce the number of system states, downsize the calculate amount of iterative process. We prove and simulate that the state aggregate method and action set reduction scheme will have little influence of the efficiency of algorithms while reduce the calculate amount.