Achieving Energy Efficiency via Drowsy Transmission in Cognitive Radio

Energy efficiency is increasingly important for wireless communication systems. An optimization policy for cognitive radio communication systems is derived by employing dynamic programming, thus improving the energy efficiency via drowsy transmission. An alternative Q-learning approach is introduced when the environment model is unknown. Numerical simulation results demonstrate that our proposed optimization is effective. Significant performance gain is achieved especially when the traffic load of secondary user is small, which reduces the energy consumption up to 50%.