UAV Relay in VANETs Against Smart Jamming with Reinforcement Learning

Xiao, Liang; Zhuang, Wei; Zhou, Sheng; Chen, Cailian

doi:10.1007/978-3-030-01731-6_5

Cited by 69 publications

(126 citation statements)

References 21 publications

Supporting

Mentioning

124

Contrasting

Unclassified

Order By: Relevance

“…While the main purpose of utilizing city map or radio map is to learn the channel indirectly or directly, another useful technique is to learn and adapt to the environment by directly interacting with it, for which reinforcement learning emerges as a powerful tool [175]. Reinforcement learning has been used in UAV networks for various purposes, e.g., navigation [176], anti-jamming [177], and communication rate maximization [178]. Specifically, the authors in [176] applied the deep reinforcement learning (DRL) technique for autonomous UAV navigation in complex environment to guide the UAV flying from a given initial location to the destination, using only sensory information, such as the UAV's orientation angle and the distances to obstacles and the destination.…”

Section: F Uav-assisted Communication Via Intelligent Learningmentioning

confidence: 99%

Accessing From the Sky: A Tutorial on UAV Communications for 5G and Beyond

2019

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) have found numerous applications and are expected to bring fertile business opportunities in the next decade. Among various enabling technologies for UAVs, wireless communication is essential and has drawn significantly growing attention in recent years. Compared to the conventional terrestrial communications, UAVs' communications face new challenges due to their high altitude above the ground and great flexibility of movement in the threedimensional (3D) space. Several critical issues arise, including the line-of-sight (LoS) dominant UAV-ground channels and resultant strong aerial-terrestrial network interference, the distinct communication quality of service (QoS) requirements for UAV control messages versus payload data, the stringent constraints imposed by the size, weight and power (SWAP) limitations of UAVs, as well as the exploitation of the new design degree of freedom (DoF) brought by the highly controllable 3D UAV mobility. In this paper, we give a tutorial overview of the recent advances in UAV communications to address the above issues, with an emphasis on how to integrate UAVs into the forthcoming fifth-generation (5G) and future cellular networks. In particular, we partition our discussions into two promising research and application frameworks of UAV communications, namely UAVassisted wireless communications and cellular-connected UAVs, where UAVs serve as aerial communication platforms and users, respectively. Furthermore, we point out promising directions for future research and investigation. KeywordsUnmanned aerial vehicle (UAV), wireless communication, cellular network, channel model, antenna model, energy efficiency, air-ground interference, 3D placement and trajectory, optimization. TechnologyDescription Advantages Disadvantages Direct linkDirect point-to-point communication with ground node Simple, low cost Limited range, low data rate, vulnerable to interference, non-scalable Satellite Communication and Internet access via satellite Global coverage Costly, heavy/bulky/energyconsuming communication equipment, high latency, large signal attenuation Ad-hoc network Dynamically self-organizing and infrastructure-free network Robust and adaptable, support for high mobility Costly, low spectrum efficiency, intermittent connectivity, complex routing protocol Cellular network Enabling UAV communications by using cellular infrastructure and technologies Almost ubiquitous accessibility, cost-effective, superior performance and scalability Unavailable in remote areas, potential interference with terrestrial communications and high reliability/throughput for both air-to-air and airto-ground wireless communications in the three-dimensional (3D) space. Towards this end, four candidate communication technologies are listed and compared in Table III, including i) direct link; ii) satellite; iii) ad-hoc network; and iv) cellular network. In the following, we discuss the advantages as well as limitations of each of these technologies in detail.1) Direct Link: Due to its simpli...

show abstract

Section: F Uav-assisted Communication Via Intelligent Learningmentioning

confidence: 99%

Accessing From the Sky: A Tutorial on UAV Communications for 5G and Beyond

2019

View full text Add to dashboard Cite

show abstract

“…In this article, we have elaborated the bi-directional mission offloading framework in SAGIN, which makes full use of the complementary advantages from the space-air networks and ground 15 networks. The overall architecture of agile mission offloading, and the enabling network reconfiguration framework based on NFV and SFC, have been introduced and validated with a case study, which demonstrates the substantial performance gain in reliability and cost reduction.…”

Section: Discussionmentioning

confidence: 99%

Bidirectional Mission Offloading for Agile Space-Air-Ground Integrated Networks

Zhou

Wang

Zhang

et al. 2019

IEEE Wireless Commun.

Self Cite

View full text Add to dashboard Cite

Space-air-ground integrated networks (SAGIN) provide great strengths in extending the capability of ground wireless networks. On the other hand, with rich spectrum and computing resources, the ground networks can also assist space-air networks to accomplish resource-intensive or power-hungry missions, enhancing the capability and sustainability of the space-air networks. Therefore, bi-directional mission offloading can make full use of the advantages of SAGIN and benefits both space-air and ground networks. In this article, we identify the key role of network reconfiguration in coordinating heterogeneous resources in SAGIN, and study how network function virtualization (NFV) and service function chaining (SFC) enable agile mission offloading. A case study validates the performance gain brought by bi-directional mission offloading. Future research issues are outlooked as the bi-directional mission offloading framework opens a new trail in releasing the full potentials of SAGIN.

show abstract

“…To quickly achieve the optimal relay policy for the UAV, the DQL based on CNN is then adopted. The simulation results in [115] show that the proposed DQL scheme takes only 200 time slots to converge to the optimal policy, which is 83.3% less than that of the relay scheme based on Q-learning [116]. Moreover, the proposed DQL scheme reduces the BER of the user by 46.6% compared with the hill climbing-based UAV relay scheme [117].…”

Section: A Network Securitymentioning

confidence: 99%

Applications of Deep Reinforcement Learning in Communications and Networking: A Survey

Luong

Hoang

Gong

et al. 2019

IEEE Commun. Surv. Tutorials

1,476

481

View full text Add to dashboard Cite

This paper presents a comprehensive literature review on applications of deep reinforcement learning in communications and networking. Modern networks, e.g., Internet of Things (IoT) and Unmanned Aerial Vehicle (UAV) networks, become more decentralized and autonomous. In such networks, network entities need to make decisions locally to maximize the network performance under uncertainty of network environment. Reinforcement learning has been efficiently used to enable the network entities to obtain the optimal policy including, e.g., decisions or actions, given their states when the state and action spaces are small. However, in complex and large-scale networks, the state and action spaces are usually large, and the reinforcement learning may not be able to find the optimal policy in reasonable time. Therefore, deep reinforcement learning, a combination of reinforcement learning with deep learning, has been developed to overcome the shortcomings. In this survey, we first give a tutorial of deep reinforcement learning from fundamental concepts to advanced models. Then, we review deep reinforcement learning approaches proposed to address emerging issues in communications and networking. The issues include dynamic network access, data rate control, wireless caching, data offloading, network security, and connectivity preservation which are all important to next generation networks such as 5G and beyond. Furthermore, we present applications of deep reinforcement learning for traffic routing, resource sharing, and data collection. Finally, we highlight important challenges, open issues, and future research directions of applying deep reinforcement learning.

show abstract

UAV Relay in VANETs Against Smart Jamming with Reinforcement Learning

Cited by 69 publications

References 21 publications

Accessing From the Sky: A Tutorial on UAV Communications for 5G and Beyond

Accessing From the Sky: A Tutorial on UAV Communications for 5G and Beyond

Bidirectional Mission Offloading for Agile Space-Air-Ground Integrated Networks

Applications of Deep Reinforcement Learning in Communications and Networking: A Survey

Contact Info

Product

Resources

About