基于Q-Learning的深度神经网络自适应退避策略

An adaptive back-off strategy based on deep Q-Learning neural network

摘要:

针对无人机自组织网络，结合Q-Learning和深度神经网络，提出一种自适应退避策略，以提高基于竞争的MAC协议通信性能。以Matlab为仿真平台，仿真比较了自适应退避策略与二进制指数退避策略的性能。

Abstract:

An adaptive back-off strategy based on Q-Learning and deep neural network is proposed to improve the communication performance of MAC protocol based on competition for unmanned aerial vehicle self-organizing network. In the experiment, Matlab is used as the simulation platform to compare the performance of adaptive back-off strategy and binary exponential back-off strategy.