Article
Computer Science, Information Systems
Atefeh Hajijamali Arani, M. Mahdi Azari, Peng Hu, Yeying Zhu, Halim Yanikomeroglu, Safieddin Safavi-Naeini
Summary: Integrating UAVs as aerial base stations into terrestrial cellular networks is an effective solution for coverage and communication service enhancement. This study proposes a novel trajectory design mechanism for rotary-wing UAV-BSs to improve energy efficiency using reinforcement learning. Simulation results show significant performance gains in terms of network throughput and energy efficiency.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Aerospace
Xingling Shao, Yi Xia, Zewei Mei, Wendong Zhang
Summary: This paper proposes a model-guided reinforcement learning algorithm for UAVs to enclose a maneuverable target within a fixed time. The algorithm ensures collision-free behaviors and reinforced tracking capability. It introduces a fixed-time enclosing controller, reward functions, and a training procedure to enhance efficiency and generalization.
AEROSPACE SCIENCE AND TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Deepti Saraswat, Ashwin Verma, Pronaya Bhattacharya, Sudeep Tanwar, Gulshan Sharma, Pitshou N. Bokoro, Ravi Sharma
Summary: This paper introduces the application of blockchain-based federated learning in UAV networks. By training data on local nodes, data privacy can be protected and network communication can be improved. In addition, a blockchain-based federated learning scheme is proposed to achieve trusted data exchange among UAV swarms and ground stations. The survey also provides a logistics case study of blockchain-based federated learning-oriented UAV networks.
Article
Computer Science, Information Systems
Tingting Yuan, Christian Esteve Rothenberg, Katia Obraczka, Chadi Barakat, Thierry Turletti
Summary: This paper proposes using multiple 5G UAVs to enhance fairness in network resource allocation among vehicles, leveraging deep reinforcement learning to determine UAV position and improve allocation efficiency. Simulation results demonstrate that this approach can improve network resource allocation based on targeted fairness objectives.
IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT
(2021)
Article
Engineering, Electrical & Electronic
Da Liu, Liqian Dou, Ruilong Zhang, Xiuyun Zhang, Qun Zong
Summary: This paper studies the coordinated dynamic task allocation (CDTA) problem for heterogeneous unmanned aerial vehicles (UAVs) in the presence of environment uncertainty. A CDTA strategy for heterogeneous UAVs is proposed through the proposer-responser mechanism and prioritized experience replay. The CDTA algorithm considers the uncertainty of dynamic tasks and has high scalability, effectively reducing the burden of online calculation and increasing the speed of online operation. Experimental results prove the effectiveness of the proposed algorithm, with scalability verified within 10-180 UAVs through comparison simulations with game theory-based and reinforcement learning-based methods.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Computer Science, Hardware & Architecture
Haifa Touati, Amira Chriki, Hichem Snoussi, Farouk Kamoun
Summary: With the advancement of drone technology, drones are being increasingly utilized in various military and civilian missions. Communication mechanisms and spectrum resource sharing are key issues in drone swarm communication, with Cognitive Radio Network technology promising to meet the spectrum requirements of drone networks.
Article
Engineering, Electrical & Electronic
Wen Wang, Liang Wang, Junfeng Wu, Xianping Tao, Haijun Wu
Summary: In this paper, the flocking and navigation control of large-scale UAV swarms are formulated as Markov Decision Processes (MDPs), and multi-agent reinforcement learning methods are used to solve the problem. The proposed method addresses the challenges of scalability and partial observations, and achieves good performance compared to traditional local observation methods.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Automation & Control Systems
Qiyu Sun, Jinbao Fang, Wei Xing Zheng, Yang Tang
Summary: This paper proposes a curiosity-driven reinforcement learning method for quadrotor aggressive flight tasks. By introducing a curiosity module and a branch structure exploration strategy, the training procedure is accelerated and the policy's robustness is ensured. The algorithm performs well in both simulation experiments and real-world experiments.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Engineering, Electrical & Electronic
Mohammad Neinavaie, Joe Khalife, Zaher M. Kassas
Summary: This paper proposes a receiver architecture that can cognitively extract navigation observables from 5G NR signals. The receiver only requires knowledge of the frame duration and carrier frequency of the signal to estimate all the RSs and derive navigation observables. The receiver operates in two stages, acquisition and tracking, using sequential detection and tracking loops. Extensive experimental results show the capabilities of the receiver, including the first navigation results with real 5G signals on an unmanned aerial vehicle (UAV) with a position RMSE of 4.35 m.
IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING
(2022)
Article
Engineering, Electrical & Electronic
Guillem Reus-Muns, Kaushik Chowdhury
Summary: This paper proposes a deep learning method to detect the type/model of UAVs using RF signals and makes four main contributions: learning the preamble portion of data packets, introducing a pre-processing scheme for enhanced accuracy, developing a deep neural network architecture, and extending the model to a federated learning paradigm.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Computer Science, Information Systems
Jiequ Ji, Kun Zhu, Lin Cai
Summary: In this article, the authors investigate the content transmission in a heavy-crowded multiple access cellular network, and propose a novel optimization problem to minimize the sum content acquisition delay of users by optimizing the multiuser association, cache placement, UAV trajectory, and transmission power. They model the problem as a partially observable stochastic game and propose a Dual-Clip PPO-based algorithm to solve it. The proposed algorithm outperforms the standard PPO-based deep reinforcement learning algorithm, and the joint design scheme achieves a dramatic reduction of content acquisition delay compared with benchmark schemes.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Chemistry, Multidisciplinary
Wenlong Zhao, Zhijun Meng, Kaipeng Wang, Jiahui Zhang, Shaoze Lu
Summary: Active tracking control is crucial for UAVs in GPS-denied environments, and this paper proposes an end-to-end high-level control method that leverages deep reinforcement learning to map raw images to high-level action commands. By unifying perception and decision-making stages with a novel high-level controller architecture, encoding spatial and temporal features of dynamic targets, and introducing auxiliary segmentation and motion-in-depth losses for denser training signals, the UAVs achieved significantly better performance in active tracking tasks than traditional three-stage methods.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Electrical & Electronic
Mohammad Mozaffari, Xingqin Lin, Stephen Hayes
Summary: This article provides an overview of the key barriers and design considerations for the widespread commercial use of flying UAVs, along with potential solutions. It discusses how cellular networks can support UAVs by utilizing advanced features, network intelligence, key enabling technologies for beyond 5G and 6G, and new tools from machine learning towards limitless connectivity in 6G.
IEEE COMMUNICATIONS MAGAZINE
(2021)
Article
Computer Science, Information Systems
Yu Zhang, Zhiyu Mou, Feifei Gao, Ling Xing, Jing Jiang, Zhu Han
Summary: The emerging backscatter communication technology holds promise for solving the battery problem of IoT devices, but its transmission range is limited. To address this challenge, a multi-UAV-aided data collection scenario was proposed to minimize total flight time. The algorithms effectively handle multiple boundary scenarios for UAV flying regions.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Chemistry, Analytical
Wonseok Lee, Young Jeon, Taejoon Kim, Young-Il Kim
Summary: A network of unmanned aerial vehicles (UAVs) serving as base stations, known as UAV-BS network, is becoming an important component in next-generation communication systems. A novel deep Q-network (DQN)-based learning model has been proposed for optimal deployment of a UAV-BS, optimizing trajectory to maximize mean opinion score (MOS) for ground users moving on various paths. The model's accuracy has been validated by comparing results with a mathematical optimization solver.
Article
Engineering, Electrical & Electronic
Somayeh Aghashahi, Jamshid Abouei, Aliakbar Tadaion
Summary: This paper investigates the coordinated beamforming problem in a two-cell network to minimize the sum transmit power of the system. An algorithm using random matrix theory is proposed to design coordinated beamforming vectors for scenarios where base stations and users have massive MIMO antennas. Simulation results show that this algorithm accurately tracks the optimal sum transmit power minimization beamforming approach in the massive MIMO scenario, while reducing computational complexity.
IRANIAN JOURNAL OF SCIENCE AND TECHNOLOGY-TRANSACTIONS OF ELECTRICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Fatemeh Taghizadeh-Marvast, Jamshid Abouei, Ali Ghafoorzadeh-Yazdi
Summary: This article presents the design and performance analysis of a 15-GHz voltage-controlled ring oscillator using 130-nm CMOS technology that operates in a 1.8-voltage power supply for high-speed applications. The five-stage ring oscillator design includes replacing one inverter with a NAND gate to shut down the oscillator during inactive mode. The authors optimally designed the ring oscillator with respect to central frequency and power dissipation compared to other publications.
IEEE INDUSTRIAL ELECTRONICS MAGAZINE
(2021)
Article
Engineering, Electrical & Electronic
Hassan Khayatian, Farzad Parvaresh, Jamshid Abouei, Mohammad Saberali
Summary: This paper investigates the diversity-multiplexing tradeoff in two-hop parallel N-relay networks over Gamma-Gamma free-space optical channels with identical average received signal to noise ratios, considering both local and global channel state information. The study explores the impact of local and global CSI on network performance, as well as optimizing relay listening and transmitting times. The optimal DMT is achieved using static and dynamic quantize map and forward strategies, with relay scheduling in the dynamic strategy determined by local channel conditions.
IET COMMUNICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Arash Rasti-Meymandi, Ali Madahian, Jamshid Abouei, Ali Mirvakili, Zohreh HajiAkhondi-Meybodi, Arash Mohammadi, Murat Uysal
Summary: This paper presents an innovative and smart Visible Light Communication-based Intelligent Transportation System using vehicle LED headlamps as a remote controller. The system features an innovative optical interference cancellation mechanism and has been experimentally evaluated for Bit Error Rate performance at different link ranges.
AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Mohammad Mehdi Pishrow, Jamshid Abouei, Hesam Ghaferi
Summary: This paper focuses on designing matched filters with low peak sidelobe level and mismatched filters with low loss in processing gain and peak sidelobe level for phase codes. An algorithm based on genetic algorithm and a method based on semidefinite programming are proposed to solve the resulting optimization problems. Simulation results demonstrate the superiority of the proposed methods in peak sidelobe level and integrated sidelobe level for binary and polyphase codes.
Article
Telecommunications
Mina Taghavi, Jamshid Abouei
Summary: This paper discusses the importance of user coverage and QoS in wireless networks, and proposes the use of drones as mobile DBSs to address this issue. By modeling the problem of optimizing DBS positioning as a P-median optimization problem, the optimal locations of DBSs are determined.
WIRELESS PERSONAL COMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Hassan Khayatian, Farzad Parvaresh, Jamshid Abouei, S. Mohammad Saberali, Murat Uysal
Summary: In this study, two-hop parallel networks with half-duplex relays over quasi-static Rayleigh fading channels are considered, and optimal communication schemes in terms of diversity multiplexing trade-off (DMT) are obtained. It is shown that dynamic and static quantize-map-and-forward communication strategies achieve the optimal DMT at different ranges of multiplexing gain. The DMT of the proposed schemes matches the upper bounds under arbitrary average signal-to-noise ratios, and optimal transmitting/listening times of relays are derived for different communication strategies in the general setting.
PHYSICAL COMMUNICATION
(2021)
Article
Computer Science, Information Systems
Zeinab Askari, Jamshid Abouei, Muhammad Jaseemuddin, Alagan Anpalagan
Summary: This article proposes a two-tier scheduling algorithm for real-time monitoring of patients' vital signs, addressing issues such as energy efficiency, interference, delay, emergency conditions, and reliability. The algorithm considers channel state, energy consumption, and delay, and prioritizes certain information using an emergency index.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Mohammad Mehdi Pishrow, Jamshid Abouei
Summary: This study focuses on the joint design of discrete phase transmit sequences and receive filters in radar systems to achieve low peak sidelobe level and integrated sidelobe level. An iterative method based on alternate optimization approach is proposed to address the non-convex optimization problem. The proposed framework leads to better peak sidelobe ratio and integrated sidelobe ratio for discrete phase sequences compared to the optimal mismatched filters design approach.
IET RADAR SONAR AND NAVIGATION
(2022)
Article
Engineering, Electrical & Electronic
Mohsen Kazemian, Jamshid Abouei, Alagan Anpalagan
Summary: The enhanced-NOMA (E-NOMA) scheme proposed in this study outperforms FFT-NOMA by approximately 4.3 dB and 9.5 dB in terms of PAPR and BER, respectively. Additionally, E-NOMA achieves at least 56% computational complexity reduction compared to an SLM-based NOMA method.
PHYSICAL COMMUNICATION
(2021)
Article
Engineering, Electrical & Electronic
Donatella Darsena, Francesco Verde
Summary: This paper studies the effects of a jamming attack on the beam alignment (BA) procedure in millimeter-wave multiple-input multiple-output (MMW MIMO) communications. A countermeasure based on randomized probing is proposed to reject the jamming signal and improve performance and quality-of-service.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2022)
Article
Computer Science, Information Systems
Donatella Darsena, Giacinto Gelli, Ivan Iudice, Francesco Verde
Summary: This article investigates anti-jamming communication for UAV-aided WSNs operating over doubly selective channels in the downloading phase. A blind physical-layer technique is proposed to jointly detect the UAV and jammer symbols through serial disturbance cancelation based on symbol-level post-sorting of the detector output, to suppress high-power jamming signals.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Review
Engineering, Electrical & Electronic
Donatella Darsena, Giacinto Gelli, Ivan Iudice, Francesco Verde
Summary: This article presents a taxonomy and review of sensing technologies based on the Internet of Things (IoT) for real-time crowd analysis in public transportation systems. It introduces a reference architecture for crowd management using modern information and communication technologies (ICTs) to monitor and predict crowding events, implement crowd-aware policies, and inform users of the crowding status in real-time.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Information Systems
Seyed Mohammad Sheikholeslami, Fahimeh Fazel, Jamshid Abouei, Konstantinos N. Plataniotis
Summary: This paper investigates the issue of indoor localization and handover based on Visible Light Communication (VLC), proposing a Convolutional Neural Network (CNN) algorithm that utilizes both offline and online modes. The algorithm demonstrates superior performance compared to traditional methods through simulation results in a smart environment.
Article
Computer Science, Information Systems
Zohreh Hajiakhondi-Meybodi, Arash Mohammadi, Jamshid Abouei
Summary: This paper addresses the urgent need for secure communication systems due to the COVID-19 pandemic by developing a connection scheduling framework based on deep Q-networks, which trains ground users through reinforcement learning to handle requests in order to minimize user access delay and optimize energy consumption.
