Considerations of Reinforcement Learning within Real-Time Wireless Communication Systems

This paper investigates the practical limitations and considerations of implementing a reinforcement learning framework within real-time wireless communication systems. Tradeoffs between latency, memory, goodput, and effects on algorithm accuracy are among the considerations that are analyzed within this work that are not typically considered within simulationbased prior art. To perform this investigation, a representative real-time OFDM transmit/receive chain is implemented within the GNU Radio framework. The system, operating over-theair through USRPs, leverages reinforcement learning, e.g. Q-Learning, in order to avoid interference with other spectrum users. Performance analysis of this representative system provides a systematic approach for helping to predict limiting factors within an implemented real-time system and thus, aim to provide guidance on how to design these systems with these practical limitations in mind.