Joint optimal multi-connectivity enabled user association and power allocation in mmWave networks

The millimeter wave (mmWave) spectrum has been involved in the fifth generation wireless systems (5G) for its enormous spectrum resources. However, signals over mmWave band suffer from severe path loss and are vulnerable to be blocked by obstacles due to the extremely high frequency, which greatly degrades the quality of transmission and connection reliability. One of the promising ways to overcome this problem is the multi-connectivity technology, which allows a user in an mmWave network to associate with multiple mmWave base stations (mBSs). And for multi-connectivity, there are two essential challenges: one is the optimal user association, that is, which are the most appropriate mBSs a user should be connected to, and the other is the optimal power allocation of each connection. In view of this, this paper aims to jointly optimize the user association and optimal power allocation for multi-connectable mmWave networks. Different from most existing works, the optimization objectives in our work are three-fold, namely maximizing the overall energy-efficiency and meanwhile balancing the achievable rates among all users and the load among all mBSs, respectively, under the QoS constraint. Considering that this multi-objective problems is a mixed integer programming and NP-hard, this paper proposes a method based on the novel non-dominated sorting genetic algorithm II (NSGA-II) to obtain the near-optimal solutions. Simulation results show that the proposed scheme enables optimal multi-connectivity enabled user associations and power allocations simultaneously in ultra-dense mmWave networks and yields EE-fairness tradeoff.