Deep Reinforcement Learning Based Power control for Wireless Multicast Systems
This work addresses a specific bottleneck in wireless communication systems, offering an incremental improvement over existing methods for power control.
The paper tackles the intractable problem of optimal power control in wireless multicast systems by using deep reinforcement learning with a deep neural network to approximate the Q-function, achieving learnable optimal control for reasonably large systems and enabling tracking of time-varying statistics with a modified algorithm.
We consider a multicast scheme recently proposed for a wireless downlink in [1]. It was shown earlier that power control can significantly improve its performance. However for this system, obtaining optimal power control is intractable because of a very large state space. Therefore in this paper we use deep reinforcement learning where we use function approximation of the Q-function via a deep neural network. We show that optimal power control can be learnt for reasonably large systems via this approach. The average power constraint is ensured via a Lagrange multiplier, which is also learnt. Finally, we demonstrate that a slight modification of the learning algorithm allows the optimal control to track the time varying system statistics.