Incentive reward for efficient WiFi offloading using Q-learning approach

As mobile data traffic continues to increase exponentially, finding an efficient solution to offload the cellular network traffic becomes very crucial for operators. The deployment of small cells yet help resolve part of this problem, but user traffic demands always exceeded the capacity of mobile networks. WiFi technology came as one of the most attractive alternative for traffic offloading to cellular operators. It becomes especially appealing when it is about very high capacity demanding networks such as 5G. However, WiFi offloading decision as well as WiFi-Access Points (W-AP) selection should be carefully studied in order not to affect the offloaded user's quality of service. In this paper, a new Reinforcement-Learning framework is presented in order to enhance WiFi traffic offloading in a network where several W-APs coexist within a macrocell. We propose a distributed Q-learning algorithm in which each cellular user learns about his local environment and selects the best base station (macro-BS or W-AP) after reaching convergence. We introduce a new reward parameter which takes into account the load of each detected W-AP, the duration of the vertical Handover, the offered gain as well as the achieved Signal-to-Interference-plus-Noise Ratio. With our Q-learning scheme, each user decides to join the WiFi offloading or not, depending on the received reward from his environment and from his previous learning. Simulation results showed the effectiveness of our proposed Q-learning based scheme as compared to common WiFi offloading scheme in terms of cellular residence time.