Junlong Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Junlong Wang has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Hongtao Tao
School of Mechanical Engineering, University of Science and Technology Beijing, Beijing 100083, China
1 shared publication - Li Zhao
School of Electrical and Information Engineering, Yunnan Minzu University, Kunming, China
1 shared publication - Shigeru Shimamoto
Department of Computer Science and Communications Engineering, Faculty of Science and Engineering, Waseda University, Tokyo, Japan
1 shared publication - Tu Dac Ho
Department of Information Security and Communication Technology, Norwegian University of Science and Technology, Trondheim, Norway
1 shared publication - Wen Zhao
State Key Laboratory for Fine Exploration and Intelligent Development of Coal Resources, China University of Mining & Technology, Beijing, China
1 shared publication - Wen Zhao
Global Information and Telecommunication Institute, Waseda University, Shinjuku, Japan
1 shared publication
Latest Publications
- Open Access
- 21 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 66054-66068
ResNet-Based Deep Learning for OTFS Channel Estimation With Pilot Superimposition
The rapid growth of the information industry imposes strict demands on future wireless communication systems, especially in high-mobility vehicular scenarios. To address these challenges, we employ orthogonal time frequency space (OTFS) modulation to mitigate Doppler effects and enhance robustness in high-speed environments. In this paper, we examine a superimposed pilot technique for OTFS channel estimation and introduce an optimized two-stage ResNet-based architecture at the receiver. Specifically, the ResNet core leverages residual connections to extract the 2D macroscopic structural envelope of the delay-Doppler channel, effectively decoupling the severe pilot-data interference inherent in superimposed schemes. Furthermore, historical channel estimation data are integrated as a locally quasi-stationary reference to prevent noise overfitting and increase accuracy in subsequent tracking tasks. The system performance is comprehensively evaluated using Normalized Mean Square Error (NMSE) and Bit Error Rate (BER) metrics. Simulation results demonstrate that our proposed method significantly lowers the estimation error floor and outperforms existing traditional and deep learning-based approaches under various high-mobility conditions.
- 21 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 66054-66068
Open AccessResNet-Based Deep Learning for OTFS Channel Estimation With Pilot Superimposition
The rapid growth of the information industry imposes strict demands on future wireless communication systems, especially in high-mobility vehicular scenarios. To address these challenges, we employ orthogonal time frequency space (OTFS) modulation to mitigate Doppler effects and enhance robustness in high-speed environments. In this paper, we examine a superimposed pilot technique for OTFS channel estimation and introduce an optimized two-stage ResNet-based architecture at the receiver. Specifically, the ResNet core leverages residual connections to extract the 2D macroscopic structural envelope of the delay-Doppler channel, effectively decoupling the severe pilot-data interference inherent in superimposed schemes. Furthermore, historical channel estimation data are integrated as a locally quasi-stationary reference to prevent noise overfitting and increase accuracy in subsequent tracking tasks. The system performance is comprehensively evaluated using Normalized Mean Square Error (NMSE) and Bit Error Rate (BER) metrics. Simulation results demonstrate that our proposed method significantly lowers the estimation error floor and outperforms existing traditional and deep learning-based approaches under various high-mobility conditions.
- Open Access
- 29 January 2026
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 98 (1) , 12
LLM-Powered Obstacle Map Generation for UAV Path Planning in Complex Weather Conditions
- 29 January 2026
- Published by Springer Nature in Journal of Signal Processing Systems
- Vol. 98 (1) , 12
Open AccessLLM-Powered Obstacle Map Generation for UAV Path Planning in Complex Weather Conditions
- Open Access
Research on Disaster Environment Map Fusion Construction and Reinforcement Learning Navigation Technology Based on Air–Ground Collaborative Multi-Heterogeneous Robot Systems
The primary challenge that robots face in disaster rescue is to precisely and efficiently construct disaster maps and achieve autonomous navigation. This paper proposes a method for air–ground collaborative map construction. It utilizes the flight capability of an unmanned aerial vehicle (UAV) to achieve rapid three-dimensional space coverage and complex terrain crossing for rapid and efficient map construction. Meanwhile, it utilizes the stable operation capability of an unmanned ground vehicle (UGV) and the ground detail survey capability to achieve precise map construction. The maps constructed by the two are accurately integrated to obtain precise disaster environment maps. Among them, the map construction and positioning technology is based on the FAST LiDAR–inertial odometry 2 (FAST-LIO2) framework, enabling the robot to achieve precise positioning even in complex environments, thereby obtaining more accurate point cloud maps. Before conducting map fusion, the point cloud is preprocessed first to reduce the density of the point cloud and also minimize the interference of noise and outliers. Subsequently, the coarse and fine registrations of the point clouds are carried out in sequence. The coarse registration is used to reduce the initial pose difference of the two point clouds, which is conducive to the subsequent rapid and efficient fine registration. The coarse registration uses the improved sample consensus initial alignment (SAC-IA) algorithm, which significantly reduces the registration time compared with the traditional SAC-IA algorithm. The precise registration uses the voxelized generalized iterative closest point (VGICP) algorithm. It has a faster registration speed compared with the generalized iterative closest point (GICP) algorithm while ensuring accuracy. In reinforcement learning navigation, we adopted the deep deterministic policy gradient (DDPG) path planning algorithm. Compared with the deep Q-network (DQN) algorithm and the A* algorithm, the DDPG algorithm is more conducive to the robot choosing a better route in a complex and unknown environment, and at the same time, the motion trajectory is smoother. This paper adopts Gazebo simulation. Compared with physical robot operation, it provides a safe, controllable, and cost-effective environment, supports efficient large-scale experiments and algorithm debugging, and also supports flexible sensor simulation and automated verification, thereby optimizing the overall testing process.
Open AccessResearch on Disaster Environment Map Fusion Construction and Reinforcement Learning Navigation Technology Based on Air–Ground Collaborative Multi-Heterogeneous Robot Systems
The primary challenge that robots face in disaster rescue is to precisely and efficiently construct disaster maps and achieve autonomous navigation. This paper proposes a method for air–ground collaborative map construction. It utilizes the flight capability of an unmanned aerial vehicle (UAV) to achieve rapid three-dimensional space coverage and complex terrain crossing for rapid and efficient map construction. Meanwhile, it utilizes the stable operation capability of an unmanned ground vehicle (UGV) and the ground detail survey capability to achieve precise map construction. The maps constructed by the two are accurately integrated to obtain precise disaster environment maps. Among them, the map construction and positioning technology is based on the FAST LiDAR–inertial odometry 2 (FAST-LIO2) framework, enabling the robot to achieve precise positioning even in complex environments, thereby obtaining more accurate point cloud maps. Before conducting map fusion, the point cloud is preprocessed first to reduce the density of the point cloud and also minimize the interference of noise and outliers. Subsequently, the coarse and fine registrations of the point clouds are carried out in sequence. The coarse registration is used to reduce the initial pose difference of the two point clouds, which is conducive to the subsequent rapid and efficient fine registration. The coarse registration uses the improved sample consensus initial alignment (SAC-IA) algorithm, which significantly reduces the registration time compared with the traditional SAC-IA algorithm. The precise registration uses the voxelized generalized iterative closest point (VGICP) algorithm. It has a faster registration speed compared with the generalized iterative closest point (GICP) algorithm while ensuring accuracy. In reinforcement learning navigation, we adopted the deep deterministic policy gradient (DDPG) path planning algorithm. Compared with the deep Q-network (DQN) algorithm and the A* algorithm, the DDPG algorithm is more conducive to the robot choosing a better route in a complex and unknown environment, and at the same time, the motion trajectory is smoother. This paper adopts Gazebo simulation. Compared with physical robot operation, it provides a safe, controllable, and cost-effective environment, supports efficient large-scale experiments and algorithm debugging, and also supports flexible sensor simulation and automated verification, thereby optimizing the overall testing process.