Environment-aware based access point deployment optimization for cell-free massive MIMO system
Cell-free massive multiple-input multiple-output(MIMO)systems deploy a large number of access point(AP)across the coverage area which can provide uniform high-rate services to users.However,the quality of cover-age would be affected by path loss,shadow fading scatters,and environmental occlusions around the randomly placed AP in conventional cell-free massive MIMO systems that do not consider their impact.Considering the impact of ac-tual wireless propagation environments,an AP deployment scheme was proposed to acquire uniform and consistent coverage.Firstly,a hybrid probabilistic path loss model was utilized to characterize various wireless propagation en-vironments.Then,the AP deployment optimization problem was solved with the objective of maximizing the average throughput.Finally,the problem was transformed into a Markov game process and solved by the multi-agent deep deterministic policy gradient(MADDPG)algorithm.The simulation results demonstrate that the proposed scheme can provide more uniform coverage in complex environments and serve users with reliable and consistent service compared to random AP deployment and existing AP deployment methods.
cell-free massive MIMOAP deploymenthybrid probabilistic path loss modelMADDPG algorithm
NETL
NSTL
万方数据
