Optimization for User Centric Massive MIMO Cell Free Networks via Large System Analysis

In traditional cellular networks, cell edge users suffer from severe interference. This paper considers a cell free network scenario that multiple base stations (BSs) serve a group of users without cell boundary. Each user can only be served by its own BS according to the traditional cellular network paradigm, while in the user centric cell free network, users can be associated with multiple nearby BSs. With the cell free model, interference could be converted into useful signal. Several BSs close to the users can form virtual cell and serve the cell edge users, thus eliminating the severe interference. We try to achieve the potential of the cell free network by optimization based on channel statistics. Due to the coupled nature of interference, it is difficult to get the global optimal solution. We aim to adopt a heuristic precoder that balances signal and interference to maximize the network sum rate. However, the optimization of the problem requires instantaneous channel state information. In order to reduce the information exchange in massive MIMO system, an effective coordinated power allocation algorithm only based on statistical channel state information is proposed by deriving the large system asymptotic optimization problem. In this way, the problem can be optimized in a long-term timescale and signaling overhead and computational cost can be reduced significantly. Numerical results validate the effectiveness of the proposed asymptotic optimization algorithm and show that the cell free network model can efficiently boost the throughput and maintain good fairness performance.