Downlink Power Optimization for Cell-Free Massive MIMO Over Spatially Correlated Rayleigh Fading Channels

Cell-free massive multiple-input multiple-output (mMIMO) is regarded as a promising technology in the future wireless communication. This paper aims to minimize the downlink power consumption of cell-free mMIMO system under the downlink rate constraints of users and the power constraints of per-antenna over spatially correlated Rayleigh fading channels. Firstly, this paper studies the performance of two different downlink transmission modes: non-coherent joint transmission and coherent joint transmission. Then, for the two transmission modes, the corresponding downlink power optimization model is established and an efficient power optimization algorithm is proposed based on the Lagrange multiplier (LM) method. Under the power optimization model, this paper analyzes the impact of different factors on the downlink power consumption of cell-free mMIMO system. Simulation results show that coherent joint transmission performs better than non-coherent joint transmission on spectral efficiency (SE) and energy efficiency (EE). In addition, under the same power control strategy, the downlink power consumption of cell-free mMIMO is much lower than co-located mMIMO. Moreover, the total transmission power decreases when more antennas are utilized at each access point (AP) or spatial correlation of channels becomes weak.