Unsupervised Learning for Cellular Power Control

This letter applies a feedforward neural network trained in an unsupervised fashion to the problem of optimizing the transmit powers in cellular wireless systems. Both uplink and downlink are considered, with either centralized or distributed power control. Various objectives are entertained, all of them such that the problem can be cast in convex form. The performance of the proposed procedure is very satisfactory and, in terms of computational cost, the scalability with the system dimensionality is markedly superior to that of convex solvers. Moreover, the optimization relies on directly measurable channel gains, with no need for user location information.