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

作　　者：毛中杰 俞晖[1] 麻智超 王政[1] MAO Zhongjie;YU Hui;MA Zhichao;WANG Zheng(Shanghai Jiao Tong University,Shanghai 200240,China;Beijing Research Institute of Telemetry,Beijing 100076,China)

机构地区：[1]上海交通大学,上海200240 [2]北京遥测技术研究所,北京100076

出　　处：《遥测遥控》2021年第1期19-25,共7页Journal of Telemetry,Tracking and Command

基　　金：国防基础科研计划“十三五”项目(NO.JCKY2017203B082)。

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。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.

关 键 词：无人机 Q-LEARNING 深度神经网络 退避策略 

分 类 号：TN830.6[电子电信—信息与通信工程]