QoS-aware cell association in 5G heterogeneous networks with massive MIMO

5G cellular networks aim to improve the connection speed to ensure ultra-low latency (e.g., to support real-time video streaming service) through heterogeneous environments. HetNets composed of different types of cells (e.g., macrocells and small cells) are the major direction of 5G network design. In this article, we first introduce the key features and highlight resource management issues in 5G HetNets. We then focus on the cell association problem of multiple classes of users in HetNets. Users in different classes have different data rate requirements and will choose to associate with the cell(s) (e.g., a macrocell or small cells) that yield(s) the highest data rate. Therefore, the cells also have to allocate resources in terms of antennas to different classes of users to maximize their total revenue. We introduce cell association and antenna allocation algorithms based on the evolutionary game theory. The algorithms can achieve equilibrium solutions, ensuring that the users and the cells cannot gain higher data rate and total revenue, respectively, by changing their cell association and antenna allocation.