WDQAR: a weighted double Q-learning based adaptive routing protocol for flying ad hoc networks

Reliable routing in flying ad hoc networks (FANETs) faces challenges such as frequent topology changes and unstable links, which are caused by the rapid mobility of UAVs. Existing Q-learning-based protocols still suffer from biased Q-value estimation, slow convergence, and high control overhead. This paper proposes WDQAR, an adaptive routing protocol based on weighted double Q-learning. WDQAR consists of two stages: neighbor discovery and routing. In the neighbor discovery phase, each UAV dynamically adjusts the HELLO message broadcast interval based on link duration and speed, promptly perceiving topology changes and reducing control overhead. In the routing phase, WDQAR applies a sector-based candidate neighbor filtering mechanism to narrow down the next-hop candidate set, thereby accelerating the selection of the optimal path. The reward function comprehensively considers residual communication energy, link duration, geographical progress, and valid neighbor ratio. Then, a weighted double Q-learning framework is adopted during Q-value update to mitigate the estimation bias of Q-learning towards future values, thereby enhancing the stability of routing strategies. In addition, the protocol adopts two optimization methods, namely Q-value initialization and adaptive learning parameters, to accelerate the learning process. Simulation results based on NS-3 demonstrate that WDQAR outperforms the comparative protocols in terms of packet delivery rate, end-to-end delay, energy consumption, and routing overhead.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-18
DOI
https://doi.org/10.1038/s41598-026-81001-2
Primary Topic
Mobile Ad Hoc Networks
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

WDQAR: a weighted double Q-learning based adaptive routing protocol for flying ad hoc networks

Yijie Bai, Daojie Yu, Qi Zhang, Fan Meng et al.
3 links
Scientific Reports
Mobile Ad Hoc Networks
article

WDQAR: a weighted double Q-learning based adaptive routing protocol for flying ad hoc networks

Yijie Bai, Daojie Yu, Qi Zhang, Fan Meng, Qingzhong Ni
article en

Abstract

Reliable routing in flying ad hoc networks (FANETs) faces challenges such as frequent topology changes and unstable links, which are caused by the rapid mobility of UAVs. Existing Q-learning-based protocols still suffer from biased Q-value estimation, slow convergence, and high control overhead. This paper proposes WDQAR, an adaptive routing protocol based on weighted double Q-learning. WDQAR consists of two stages: neighbor discovery and routing. In the neighbor discovery phase, each UAV dynamically adjusts the HELLO message broadcast interval based on link duration and speed, promptly perceiving topology changes and reducing control overhead. In the routing phase, WDQAR applies a sector-based candidate neighbor filtering mechanism to narrow down the next-hop candidate set, thereby accelerating the selection of the optimal path. The reward function comprehensively considers residual communication energy, link duration, geographical progress, and valid neighbor ratio. Then, a weighted double Q-learning framework is adopted during Q-value update to mitigate the estimation bias of Q-learning towards future values, thereby enhancing the stability of routing strategies. In addition, the protocol adopts two optimization methods, namely Q-value initialization and adaptive learning parameters, to accelerate the learning process. Simulation results based on NS-3 demonstrate that WDQAR outperforms the comparative protocols in terms of packet delivery rate, end-to-end delay, energy consumption, and routing overhead.

Scientific Reports
PLA Information Engineering University (CN)
Industry, innovation and infrastructure
Mobile Ad Hoc Networks
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon