Zhipeng Lin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Zhipeng Lin has a total of 121 co-authors. Here are the top co-authors by number of shared publications:
- Tiejun Lv
Beijing University of Posts and Telecommunications
21 shared publications - Qiuming Zhu
Nanjing University of Aeronautics and Astronautics
16 shared publications - Qihui Wu
Nanjing University of Aeronautics and Astronautics, Nanjing, China
10 shared publications - Jie Zeng
Beijing Institute of Technology
8 shared publications - Wei Ni
CSIRO
7 shared publications - Pingmu Huang
School of Artificial Intelligence, Beijing University of Posts and Telecommunications (BUPT), Beijing 100876, China
6 shared publications
Latest Publications
- 25 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 18655-18668
Bayesian Learning-Based Spectrum Mapping With UAV Path Dynamic Optimization Under 3-D Unknown Environments
Spectrum mapping (SM) visualizes spectrum information across a geographical area, constructing radio environment maps (REMs), which serve as a foundation for spectrum monitoring, management, and security. Most existing SM schemes rely on spatially distributed sensors or vehicle-mounted equipment, and assume prior environmental knowledge, limiting their applicability in dynamic or unknown 3D environments. In this paper, we propose a Bayesian learning-based three-dimensional (3D) SM framework that enables accurate REM construction through adaptive UAV sampling in complex and unknown environments. First, a mutual-information-driven UAV path planner is designed by integrating an enhanced sampling-based optimization scheme, enabling efficient data collection according to the maximum mutual information criterion and recent sensing data. Second, a semi-deterministic channel dictionary, refined with sampled field data, is established to model the correlation between observed spectrum values and environmental features. Based on this dictionary, a Bayesian learning-based recovery algorithm reconstructs the spectrum distribution at unsampled positions, producing the corresponding 3D REM. Experimental results on open simulated and measured datasets demonstrate that the proposed framework reduces the mean absolute error by over 60% compared with CS-based methods and by 35% with data-driven interpolation. It also improves sampling efficiency by up to 70% for a given recovery accuracy, highlighting the effectiveness in unknown 3D environments.
- 25 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 18655-18668
Bayesian Learning-Based Spectrum Mapping With UAV Path Dynamic Optimization Under 3-D Unknown Environments
Spectrum mapping (SM) visualizes spectrum information across a geographical area, constructing radio environment maps (REMs), which serve as a foundation for spectrum monitoring, management, and security. Most existing SM schemes rely on spatially distributed sensors or vehicle-mounted equipment, and assume prior environmental knowledge, limiting their applicability in dynamic or unknown 3D environments. In this paper, we propose a Bayesian learning-based three-dimensional (3D) SM framework that enables accurate REM construction through adaptive UAV sampling in complex and unknown environments. First, a mutual-information-driven UAV path planner is designed by integrating an enhanced sampling-based optimization scheme, enabling efficient data collection according to the maximum mutual information criterion and recent sensing data. Second, a semi-deterministic channel dictionary, refined with sampled field data, is established to model the correlation between observed spectrum values and environmental features. Based on this dictionary, a Bayesian learning-based recovery algorithm reconstructs the spectrum distribution at unsampled positions, producing the corresponding 3D REM. Experimental results on open simulated and measured datasets demonstrate that the proposed framework reduces the mean absolute error by over 60% compared with CS-based methods and by 35% with data-driven interpolation. It also improves sampling efficiency by up to 70% for a given recovery accuracy, highlighting the effectiveness in unknown 3D environments.
- Open Access
Soft Threshold Denoising-Based Environmental Adaptive UAV Signal Modulation Recognition for Small-Sample Scenarios
As a key technology for wireless signal identification, modulation recognition plays an important role in the fields of unmanned aerial vehicle (UAV) communications, low-altitude spectrum management, etc. However, the accuracy of modulation recognition often cannot be guaranteed in scenarios with serious noise interference when a few samples are available. In this paper, we propose an intelligent modulation recognition method for UAV signals based on small-sample augmentation and soft threshold denoising. We first propose a new dual-driven dataset expansion method by combining the UAV air–ground channel propagation model with the received data samples. Then, we construct a background learning-based long short-term memory (BL-LSTM) model to extract the environmental background features embedded in the UAV signal, including Line-of-Sight (LoS) state, multi-scale fading parameters and Doppler shift characteristics. We integrate environmental background information into the data training model and optimize the authenticity of data distribution. As a result, the model adaptability can be enhanced. Finally, we construct a deep residual shrinkage network based on the soft threshold function (STF-DRSN). By leveraging the capability of the soft threshold that resists noise interference, we integrate it into each residual block of the deep residual shrinkage network. Simulation results show that compared with the state of the art, our method can improve the modulation recognition accuracy of UAV signals in small-sample scenarios.
Open AccessSoft Threshold Denoising-Based Environmental Adaptive UAV Signal Modulation Recognition for Small-Sample Scenarios
As a key technology for wireless signal identification, modulation recognition plays an important role in the fields of unmanned aerial vehicle (UAV) communications, low-altitude spectrum management, etc. However, the accuracy of modulation recognition often cannot be guaranteed in scenarios with serious noise interference when a few samples are available. In this paper, we propose an intelligent modulation recognition method for UAV signals based on small-sample augmentation and soft threshold denoising. We first propose a new dual-driven dataset expansion method by combining the UAV air–ground channel propagation model with the received data samples. Then, we construct a background learning-based long short-term memory (BL-LSTM) model to extract the environmental background features embedded in the UAV signal, including Line-of-Sight (LoS) state, multi-scale fading parameters and Doppler shift characteristics. We integrate environmental background information into the data training model and optimize the authenticity of data distribution. As a result, the model adaptability can be enhanced. Finally, we construct a deep residual shrinkage network based on the soft threshold function (STF-DRSN). By leveraging the capability of the soft threshold that resists noise interference, we integrate it into each residual block of the deep residual shrinkage network. Simulation results show that compared with the state of the art, our method can improve the modulation recognition accuracy of UAV signals in small-sample scenarios.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74 (00906778) , 7090-7105
Variational Bayesian Multi-Source Localization in Complex Multipath Environments
Radiation source (RS) localization is crucial in electromagnetic environmental monitoring. Existing methods require prior knowledge of the environment and have overlooked the impact of multipath effects. Their positioning accuracy is penalized in practical applications. This paper presents an environment cognition-based variational Bayesian positioning method, which can achieve precise multi-source localization in uncooperative multipath environments using only the measurements of received signal strength (RSS). We begin by leveraging the signal propagation properties to design a model-data integrated unsupervised learning network, which partitions the positioning area according to the transmission states of different multipath signals. To estimate the number of RSs and mitigate the multipath effect, we identify the data samples of multipath RSS and divide them into groups, each corresponding to an RS. Then, a new variational Bayesian positioning algorithm is developed to operate without prior channel information and locate multiple RSs accurately across varying signal propagation models. Simulations show that our positioning method can effectively improve the accuracy of multi-RS localization by 10% in uncooperative multipath environments, compared to the state-of-the-art RSS-based localization techniques.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74 (00906778) , 7090-7105
Variational Bayesian Multi-Source Localization in Complex Multipath Environments
Radiation source (RS) localization is crucial in electromagnetic environmental monitoring. Existing methods require prior knowledge of the environment and have overlooked the impact of multipath effects. Their positioning accuracy is penalized in practical applications. This paper presents an environment cognition-based variational Bayesian positioning method, which can achieve precise multi-source localization in uncooperative multipath environments using only the measurements of received signal strength (RSS). We begin by leveraging the signal propagation properties to design a model-data integrated unsupervised learning network, which partitions the positioning area according to the transmission states of different multipath signals. To estimate the number of RSs and mitigate the multipath effect, we identify the data samples of multipath RSS and divide them into groups, each corresponding to an RS. Then, a new variational Bayesian positioning algorithm is developed to operate without prior channel information and locate multiple RSs accurately across varying signal propagation models. Simulations show that our positioning method can effectively improve the accuracy of multi-RS localization by 10% in uncooperative multipath environments, compared to the state-of-the-art RSS-based localization techniques.
- 17 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74 (00906778) , 5412-5427
Measurement-Driven Cluster Power Generation Method for a Hybrid A2G Channel Model
Drones are expected to be promising aerial platforms in air-to-ground (A2G) integrated communication networks, where the A2G propagation channel is fundamental for reliable communication links. This paper proposes a hybrid parameter generation framework combining the deterministic and statistical methods for a cluster-based A2G channel model. In this framework, the map-based deterministic method is used to generate delay and angle parameters, which can achieve great scenario consistency. However, it is difficult for users to provide precise material information of scatterers, which would cause deviation of power parameters. To tackle this issue, a measurement-driven power generation method is proposed. Firstly, a bandwidth-dependent clustering method is developed to group the rays into clusters. Then, the cluster power is generated by measurement-driven statistical models with a power-decomposition idea. It decomposes the power parameter into several parts that are less dependent on the scenario. Moreover, it can avoid massive measurement campaigns and is more robust when applied in unmeasured scenarios. Finally, a new channel measurement campaign in a street canyon scenario is performed for validations. The proposed method is also compared with a ray-tracing (RT) method and the 3rd Generation Partnership Project (3GPP) channel model. It is shown that the proposed framework and power generation method are great alternatives for accurate and robust modeling requirements under specific A2G communication scenarios.
- 17 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74 (00906778) , 5412-5427
Measurement-Driven Cluster Power Generation Method for a Hybrid A2G Channel Model
Drones are expected to be promising aerial platforms in air-to-ground (A2G) integrated communication networks, where the A2G propagation channel is fundamental for reliable communication links. This paper proposes a hybrid parameter generation framework combining the deterministic and statistical methods for a cluster-based A2G channel model. In this framework, the map-based deterministic method is used to generate delay and angle parameters, which can achieve great scenario consistency. However, it is difficult for users to provide precise material information of scatterers, which would cause deviation of power parameters. To tackle this issue, a measurement-driven power generation method is proposed. Firstly, a bandwidth-dependent clustering method is developed to group the rays into clusters. Then, the cluster power is generated by measurement-driven statistical models with a power-decomposition idea. It decomposes the power parameter into several parts that are less dependent on the scenario. Moreover, it can avoid massive measurement campaigns and is more robust when applied in unmeasured scenarios. Finally, a new channel measurement campaign in a street canyon scenario is performed for validations. The proposed method is also compared with a ray-tracing (RT) method and the 3rd Generation Partnership Project (3GPP) channel model. It is shown that the proposed framework and power generation method are great alternatives for accurate and robust modeling requirements under specific A2G communication scenarios.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 15, 5105-5109
Uncertainty-Aware GP Bayesian Optimization for UAV-Based Closed-Loop RSSD Interference Localization
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Wireless Communications Letters
- Vol. 15, 5105-5109
Uncertainty-Aware GP Bayesian Optimization for UAV-Based Closed-Loop RSSD Interference Localization
- Open Access
An Intelligent Passive System for UAV Detection and Identification in Complex Electromagnetic Environments via Deep Learning
With the rapid proliferation of unmanned aerial vehicles (UAVs) and the associated rise in security concerns, there is a growing demand for robust detection and identification systems capable of operating reliably in complex electromagnetic environments. To address this challenge, this paper proposes a deep learning-based passive UAV detection and identification system leveraging radio frequency (RF) spectrograms. The system employs a high-resolution RF front-end comprising a multi-beam directional antenna and a wideband spectrum analyzer to scan the target airspace and capture UAV signals with enhanced spatial and spectral granularity. A YOLO-based detection module is then used to extract frequency hopping signal (FHS) regions from the spectrogram, which are subsequently classified by a convolutional neural network (CNN) to identify specific UAV models. Extensive measurements are carried out in both line-of-sight (LoS) and non-line-of-sight (NLoS) urban environments. The proposed system achieves over 96% accuracy in both detection and identification under LoS conditions. In NLoS conditions, it improves the identification accuracy by more than 15% compared with conventional full-spectrum CNN-based methods. These results validate the system’s robustness, real-time responsiveness, and strong practical applicability.
Open AccessAn Intelligent Passive System for UAV Detection and Identification in Complex Electromagnetic Environments via Deep Learning
With the rapid proliferation of unmanned aerial vehicles (UAVs) and the associated rise in security concerns, there is a growing demand for robust detection and identification systems capable of operating reliably in complex electromagnetic environments. To address this challenge, this paper proposes a deep learning-based passive UAV detection and identification system leveraging radio frequency (RF) spectrograms. The system employs a high-resolution RF front-end comprising a multi-beam directional antenna and a wideband spectrum analyzer to scan the target airspace and capture UAV signals with enhanced spatial and spectral granularity. A YOLO-based detection module is then used to extract frequency hopping signal (FHS) regions from the spectrogram, which are subsequently classified by a convolutional neural network (CNN) to identify specific UAV models. Extensive measurements are carried out in both line-of-sight (LoS) and non-line-of-sight (NLoS) urban environments. The proposed system achieves over 96% accuracy in both detection and identification under LoS conditions. In NLoS conditions, it improves the identification accuracy by more than 15% compared with conventional full-spectrum CNN-based methods. These results validate the system’s robustness, real-time responsiveness, and strong practical applicability.
- 25 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 4032-4044
Nonlinearity Approximation-Based Spectrum Map Fusion With Environmentally Adaptive Regularization
Fusing spectral data of multiple sensors can improve the construction accuracy of spectrum environment maps (SEMs) compared to single-sensor systems. However, most existing SEM fusion methods are inapplicable to electromagnetic environments with different change rates of electromagnetic strength. This paper presents a new nonlinearity approximation-based SEM fusion method that can dynamically adjust the regularization coefficients for SEM fusion according to the complexity of electromagnetic environments. We first estimate the nonlinearity of spectrum data by developing multi-dimensional local linear approximation (MD-LOLA), with which the spectrum variation intensity of the environment can be assessed. Then, we construct a new SEM fusion model with environmentally adaptive regularization. This model combines the nonlinearity of spectrum data with the least absolute shrinkage selection operator (LASSO) and Tikhonov regularization terms, and therefore, can adjust the regularization constraint strength according to the spectrum variation intensity of the environment, improving the SEM construction accuracy. Simulation results show that our proposed SEM fusion method can effectively eliminate outliers caused by noises and measurement errors, and achieve an accuracy improvement of more than 25% compared to state-of-the-art fusion methods in SEM construction.
- 25 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 4032-4044
Nonlinearity Approximation-Based Spectrum Map Fusion With Environmentally Adaptive Regularization
Fusing spectral data of multiple sensors can improve the construction accuracy of spectrum environment maps (SEMs) compared to single-sensor systems. However, most existing SEM fusion methods are inapplicable to electromagnetic environments with different change rates of electromagnetic strength. This paper presents a new nonlinearity approximation-based SEM fusion method that can dynamically adjust the regularization coefficients for SEM fusion according to the complexity of electromagnetic environments. We first estimate the nonlinearity of spectrum data by developing multi-dimensional local linear approximation (MD-LOLA), with which the spectrum variation intensity of the environment can be assessed. Then, we construct a new SEM fusion model with environmentally adaptive regularization. This model combines the nonlinearity of spectrum data with the least absolute shrinkage selection operator (LASSO) and Tikhonov regularization terms, and therefore, can adjust the regularization constraint strength according to the spectrum variation intensity of the environment, improving the SEM construction accuracy. Simulation results show that our proposed SEM fusion method can effectively eliminate outliers caused by noises and measurement errors, and achieve an accuracy improvement of more than 25% compared to state-of-the-art fusion methods in SEM construction.
- 28 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 1664-1677
UAV-Aided Efficient Informative Path Planning for Autonomous 3D Spectrum Mapping
Three-dimensional (3D) radio environment maps (REMs) can visually represent the spectrum situation over the geographical maps, which facilitates the management of spectrum resources. Recently, the utilization of unmanned aerial vehicles (UAVs) for 3D REM construction has emerged as a promising solution. In this paper, we address the UAV sampling path planning problem by maximizing the information collection along dynamically optimized paths. On this basis, we propose an efficient informative path planning (IPP)-based scheme for autonomous spectrum mapping in unknown environments. Firstly, an effective spectrum data spatial distribution prediction method based on the scalable Gaussian process is introduced. It generates estimated REMs and quantifies associated uncertainty, which serve as the basis for UAV path planning. Then, we develop a hierarchical path planning method. Specifically, an uncertainty-aware target decision strategy is initially designed, which selects informative targets by maximizing the negative integrated posterior variance utility. Then, a local path planner driven by the upper confidence bound criterion is introduced, which utilizes the dynamic programming method to refine the UAV’s path toward the target. The proposed method is validated by a simulated dataset in the campus scenario. Compared to state-of-the-art approaches, simulation results demonstrate that the proposed scheme achieves reductions of 20.37%, 14.85%, and 57.29% in mean absolute error, mean negative log loss, and mean uncertainty, respectively. Moreover, the time consumption is reduced by 81.65% compared to conventional IPP-based method.
- 28 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 1664-1677
UAV-Aided Efficient Informative Path Planning for Autonomous 3D Spectrum Mapping
Three-dimensional (3D) radio environment maps (REMs) can visually represent the spectrum situation over the geographical maps, which facilitates the management of spectrum resources. Recently, the utilization of unmanned aerial vehicles (UAVs) for 3D REM construction has emerged as a promising solution. In this paper, we address the UAV sampling path planning problem by maximizing the information collection along dynamically optimized paths. On this basis, we propose an efficient informative path planning (IPP)-based scheme for autonomous spectrum mapping in unknown environments. Firstly, an effective spectrum data spatial distribution prediction method based on the scalable Gaussian process is introduced. It generates estimated REMs and quantifies associated uncertainty, which serve as the basis for UAV path planning. Then, we develop a hierarchical path planning method. Specifically, an uncertainty-aware target decision strategy is initially designed, which selects informative targets by maximizing the negative integrated posterior variance utility. Then, a local path planner driven by the upper confidence bound criterion is introduced, which utilizes the dynamic programming method to refine the UAV’s path toward the target. The proposed method is validated by a simulated dataset in the campus scenario. Compared to state-of-the-art approaches, simulation results demonstrate that the proposed scheme achieves reductions of 20.37%, 14.85%, and 57.29% in mean absolute error, mean negative log loss, and mean uncertainty, respectively. Moreover, the time consumption is reduced by 81.65% compared to conventional IPP-based method.