基于GPSR和Q网络的流量感知无人机ad-hoc网络路由协议（英文）
发布时间：2024-06-04 00:22
在流量密集的无人机ad-hoc网络中,流量拥塞会增加网络时延和丢包,大大限制网络性能。因此,需要一个流量平衡策略控制流量。本文提出TQNGPSR,一个基于GPSR和Q网络的流量感知无人机ad-hoc网络路由协议。该协议利用邻居节点的拥塞信息实现流量平衡,并用一种强化学习算法—Q网络算法—评价当前节点每条无线链接的质量。基于对这些链接的评估,该协议可在多个选择中做出合理决定,降低网络时延和丢包率。在仿真环境中测试TQNGPSR、AODV、OLSR、GPSR和QNGPSR。结果表明,相比于GPSR和QNGPSR, TQNGPSR有更高包到达率和更低端到端时延。在高节点密度场景中,TQNGPSR在包到达率、端到端时延和吞吐量上优于AODV和OLSR。
【文章页数】：14 页
【文章目录】：
1 Introduction
2 Background and related works
2.1 Traffic balancing
2.2 Reinforcement learning
2.3 Reinforcement learning based routing protocols
3 Traffic-aware Q-network enhanced geographic routing
3.1 Traffic balancing
3.2 Q-network based route selection
4 Simulation results
4.1 Simulation results and comparison with other protocols
4.2 Simulation results under different penalty factors
4.3 Complexity analysis and comparison with other protocols
5 Conclusions
本文编号：3988600
【文章页数】：14 页
【文章目录】：
1 Introduction
2 Background and related works
2.1 Traffic balancing
2.2 Reinforcement learning
2.3 Reinforcement learning based routing protocols
3 Traffic-aware Q-network enhanced geographic routing
3.1 Traffic balancing
3.2 Q-network based route selection
4 Simulation results
4.1 Simulation results and comparison with other protocols
4.2 Simulation results under different penalty factors
4.3 Complexity analysis and comparison with other protocols
5 Conclusions
本文编号：3988600
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/3988600.html
