Unsupervised Learning for C-RAN Power Control and Power Allocation

This letter applies a feedforward neural network trained in an unsupervised fashion to the problem of optimizing transmit powers centralized radio access networks operating on cell-free basis. Both uplink and downlink are considered. Various objectives entertained, some leading convex formulations that do not. In all cases, performance proposed procedure is very satisfactory and, terms computational cost, scalability manifestly superior solvers. Moreover, optimization relies directly measurable channel gains, with no need for user location information.