Application-Level Data Rate Adaptation in Wi-Fi Networks Using Deep Reinforcement Learning

Wireless technologies are used in almost every application domain. They are easy to deploy and some of them offer high data rates. Modern applications require more bandwidth to cope with the growing quality of network content (i.e. multimedia). The increasing demand for network capacity in terms of bandwidth and the growing number of users are causing densification in the deployed networks. Most wireless technologies, such as Wi-Fi, suffer from deterioration of Quality of Experience (QoE) in dense deployments. To overcome this problem, one of the techniques that can be used is data rate adaptation depending on the state of the network. In this paper, we propose a Deep Reinforcement Learning (DRL) approach for decentralized application-level data rate adaptation in dense Wi-Fi networks. We present the training procedure of the DRL model using NS-3 simulator and TensorFlow. The model is then evaluated in dense scenarios and compared to an existing approach from the literature. Results show that using DRL can help to better cope with the current capacity of the wireless network.