Yasar Sinan Nasir

Yasar Sinan Nasir, Dongning Guo

A wireless network operator typically divides the radio spectrum it possesses into a number of subbands. In a cellular network those subbands are then reused in many cells. To mitigate co-channel interference, a joint spectrum and power allocation problem is often formulated to maximize a sum-rate objective. The best known algorithms for solving such problems generally require instantaneous global channel state information and a centralized optimizer. In fact those algorithms have not been implemented in practice in large networks with time-varying subbands. Deep reinforcement learning algorithms are promising tools for solving complex resource management problems. A major challenge here is that spectrum allocation involves discrete subband selection, whereas power allocation involves continuous variables. In this paper, a learning framework is proposed to optimize both discrete and continuous decision variables. Specifically, two separate deep reinforcement learning algorithms are designed to be executed and trained simultaneously to maximize a joint objective. Simulation results show that the proposed scheme outperforms both the state-of-the-art fractional programming algorithm and a previous solution based on deep reinforcement learning.

Yasar Sinan Nasir, Dongning Guo

Deep reinforcement learning offers a model-free alternative to supervised deep learning and classical optimization for solving the transmit power control problem in wireless networks. The multi-agent deep reinforcement learning approach considers each transmitter as an individual learning agent that determines its transmit power level by observing the local wireless environment. Following a certain policy, these agents learn to collaboratively maximize a global objective, e.g., a sum-rate utility function. This multi-agent scheme is easily scalable and practically applicable to large-scale cellular networks. In this work, we present a distributively executed continuous power control algorithm with the help of deep actor-critic learning, and more specifically, by adapting deep deterministic policy gradient. Furthermore, we integrate the proposed power control algorithm to a time-slotted system where devices are mobile and channel conditions change rapidly. We demonstrate the functionality of the proposed algorithm using simulation results.