FURI | Spring 2023
Optimal Power Allocation in Cellular Networks Through Machine Learning
Increasing reliance on wireless networks demands a system design that is robust and efficient while minimizing interference between competing transmitters. In this research, we assume multiple transmitters each trying to send a signal to its intended receiver. Optimal power allocation using formulas derived from Information Theory is known to be an NP-hard problem. This work leverages machine learning to allocate the transmission power among all users in the system with the objective of maximizing the network sum rate. We present algorithms that are optimized for a wide range of scenarios such as interference-limited as well as noise-limited setups. The incorporation of low computational cost methods provides for dynamic allocation of transmitter power while preventing latency concerns due to computationally expensive traditional algorithms.
Hometown: Denver, Colorado, United States
Graduation date: Fall 2025