LeAP: Learning based Adaptive Uplink Power Control for 4G LTE Cellular Networks

LTE's uplink (UL) efficiency critically depends on how the interference across different cells is controlled. The unique characteristics of LTE's modulation and UL resource assignment poses considerable challenges in achieving this goal. In this work, we propose LeAP, a measurement databased machine learning paradigm for power control in LTE. The data driven approach has the inherent advantage that the solution adapts based on network traffic, propagation and network topology that is increasingly heterogeneous consisting of multiple overlays of cells. The LeAP system consists of the following designs: (i) design of UE (user equipment) measurement reports that are succinct yet expressive enough to capture the network dynamics, and (ii) design of two learning based algorithms that use the reported measurements to set the power control parameters and optimize the network performance. LeAP is standards compliant and can be implemented in a centralized SON (self organized networking) server. We perform extensive evaluations using real LTE radio network plans by a leading operator in New York City. Our results show that, compared to existing approaches, LeAP provides around 3- gain in 20th%􀀀 tile of user data rate for most cells, 2 gain in median data rate for most cells.