Rong Fan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rong Fan has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Yishan Su
Tianjin University
17 shared publications - Zhigang Jin
School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
8 shared publications Yishan Su
6 shared publications- Sidan Yang
School of Electrical and Information Engineering, Tianjin University, Tianjin, China
5 shared publications - Azzedine Boukerche
University of Ottawa
3 shared publications - Pan Pan
School of Electrical and Information Engineering, Tianjin University, Tianjin, China
3 shared publications
Latest Publications
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 25238-25254
Secure Bandwidth Allocation in UASNs via Spatially Correlated Shadowing Compensation and Geometric-Consistency Constraints
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (16) , 25238-25254
Secure Bandwidth Allocation in UASNs via Spatially Correlated Shadowing Compensation and Geometric-Consistency Constraints
- 2 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. PP (19430663) , 1
A Real-Time Embedded System for Underwater Automatic Modulation Recognition Based on a Lightweight Network
Reliable automatic modulation recognition (AMR) drives the advancement of intelligent underwater acoustic communication (UAC). However, deploying deep learning-based AMR on edge platforms remains challenging due to strict computational constraints. This letter presents an end-to-end real-time embedded UAC-AMR system via hardware–algorithm co-design, jointly addressing weak-signal acquisition, timing stability, and edge inference. A hierarchically decoupled architecture with multi-level buffering isolates sampling-critical acquisition from backend processing. The system integrates an STM32F413- and Raspberry Pi Zero 2W-based acquisition subsystem for deterministic acquisition and bridged transmission, and a Raspberry Pi 5 with Hailo-8 NPU for embedded inference. A lightweight MobileNetV4-based AMR neural network with stage-wise block allocation is adopted to support memory-efficient and hardware-friendly deployment. Field experiments over nine modulation types yield 97.92% accuracy after INT8 quantization with a single-frame latency of 6.47 ms, demonstrating practical embedded UAC-AMR.
- 2 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Embedded Systems Letters
- Vol. PP (19430663) , 1
A Real-Time Embedded System for Underwater Automatic Modulation Recognition Based on a Lightweight Network
Reliable automatic modulation recognition (AMR) drives the advancement of intelligent underwater acoustic communication (UAC). However, deploying deep learning-based AMR on edge platforms remains challenging due to strict computational constraints. This letter presents an end-to-end real-time embedded UAC-AMR system via hardware–algorithm co-design, jointly addressing weak-signal acquisition, timing stability, and edge inference. A hierarchically decoupled architecture with multi-level buffering isolates sampling-critical acquisition from backend processing. The system integrates an STM32F413- and Raspberry Pi Zero 2W-based acquisition subsystem for deterministic acquisition and bridged transmission, and a Raspberry Pi 5 with Hailo-8 NPU for embedded inference. A lightweight MobileNetV4-based AMR neural network with stage-wise block allocation is adopted to support memory-efficient and hardware-friendly deployment. Field experiments over nine modulation types yield 97.92% accuracy after INT8 quantization with a single-frame latency of 6.47 ms, demonstrating practical embedded UAC-AMR.
- 2 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1-17
Attack-Resilient Secure Localization for Underwater Acoustic Networks via Frequency-Diverse DRSS
Differential received signal strength (DRSS)-based localization is increasingly preferred for resource-constrained underwater acoustic sensor networks due to its low measurement complexity and ease of implementation. However, DRSS measurements are inherently sensitive to power perturbations induced by malicious anchor attacks, and these effects are further aggravated by severe ambient noise and strong multipath propagation, leading to degraded localization accuracy. To address these challenges, this paper proposes a multi-frequency DRSS-based underwater secure localization scheme that explicitly accounts for power-attack-induced distortions. By integrating a multipath propagation loss model with multi-frequency DRSS measurements, a robust localization model is established to suppress both power attack effects and multipath-induced uncertainties. Frequency diversity is exploited to replace spatial redundancy, enabling three-dimensional localization using only two anchors while maintaining localization reliability. Furthermore, channel impulse response simplifies the localization model by establishing relative multipath constraints, converting it into a worst-case uncertainty-aware optimization problem to handle power attacks, with a semidefinite programming (SDP) estimator proposed to turn this into a convex SDP problem for accurate sensor position estimates. Simulation results and field experiments demonstrate that the proposed scheme achieves improved localization accuracy and reliability under power attacks and other adverse measurement conditions.
- 2 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. PP (15361233) , 1-17
Attack-Resilient Secure Localization for Underwater Acoustic Networks via Frequency-Diverse DRSS
Differential received signal strength (DRSS)-based localization is increasingly preferred for resource-constrained underwater acoustic sensor networks due to its low measurement complexity and ease of implementation. However, DRSS measurements are inherently sensitive to power perturbations induced by malicious anchor attacks, and these effects are further aggravated by severe ambient noise and strong multipath propagation, leading to degraded localization accuracy. To address these challenges, this paper proposes a multi-frequency DRSS-based underwater secure localization scheme that explicitly accounts for power-attack-induced distortions. By integrating a multipath propagation loss model with multi-frequency DRSS measurements, a robust localization model is established to suppress both power attack effects and multipath-induced uncertainties. Frequency diversity is exploited to replace spatial redundancy, enabling three-dimensional localization using only two anchors while maintaining localization reliability. Furthermore, channel impulse response simplifies the localization model by establishing relative multipath constraints, converting it into a worst-case uncertainty-aware optimization problem to handle power attacks, with a semidefinite programming (SDP) estimator proposed to turn this into a convex SDP problem for accurate sensor position estimates. Simulation results and field experiments demonstrate that the proposed scheme achieves improved localization accuracy and reliability under power attacks and other adverse measurement conditions.
- 13 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (11) , 23722-23735
Diffusion Model-Based Data Augmentation With Mel Spectrogram for Automatic Modulation Recognition in Underwater Acoustic Communication
Automatic modulation recognition (AMR) in underwater acoustic (UWA) communications is limited by scarce training data and channel impairments such as multipath propagation, Doppler shifts, and ambient noise. To address these challenges, we propose Mel-Diffusion, a diffusion model-based data augmentation framework. The method first converts time-domain signals into Mel spectrograms for preprocessing. The proposed Mel-Diffusion model, which is a denoising diffusion probabilistic model (DDPM)-based diffusion framework with a Transformer backbone, is then applied for feature enhancement. Finally, the Griffin–Lim algorithm is used to reconstruct the time-domain signals. Building upon this processing pipeline, Mel-Diffusion effectively captures modulation features and latent channel characteristics. In addition, a depthwise–pointwise feed-forward network (DPFFN) integrated into the diffusion model reduces model complexity, while denoising diffusion implicit model (DDIM) sampling improves computational efficiency. Experiments on datasets from the reservoir and the sea demonstrate that Mel-Diffusion generates realistic and diverse signals, improving AMR accuracy by up to 5.51% and 5.27% under data-scarce conditions. After data augmentation, the highest AMR accuracies reach 98.66% and 95.81%, respectively, on the reservoir and sea datasets. These results highlight the effectiveness of Mel-Diffusion as a practical solution for accurate and robust AMR in complex UWA environments.
- 13 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (11) , 23722-23735
Diffusion Model-Based Data Augmentation With Mel Spectrogram for Automatic Modulation Recognition in Underwater Acoustic Communication
Automatic modulation recognition (AMR) in underwater acoustic (UWA) communications is limited by scarce training data and channel impairments such as multipath propagation, Doppler shifts, and ambient noise. To address these challenges, we propose Mel-Diffusion, a diffusion model-based data augmentation framework. The method first converts time-domain signals into Mel spectrograms for preprocessing. The proposed Mel-Diffusion model, which is a denoising diffusion probabilistic model (DDPM)-based diffusion framework with a Transformer backbone, is then applied for feature enhancement. Finally, the Griffin–Lim algorithm is used to reconstruct the time-domain signals. Building upon this processing pipeline, Mel-Diffusion effectively captures modulation features and latent channel characteristics. In addition, a depthwise–pointwise feed-forward network (DPFFN) integrated into the diffusion model reduces model complexity, while denoising diffusion implicit model (DDIM) sampling improves computational efficiency. Experiments on datasets from the reservoir and the sea demonstrate that Mel-Diffusion generates realistic and diverse signals, improving AMR accuracy by up to 5.51% and 5.27% under data-scarce conditions. After data augmentation, the highest AMR accuracies reach 98.66% and 95.81%, respectively, on the reservoir and sea datasets. These results highlight the effectiveness of Mel-Diffusion as a practical solution for accurate and robust AMR in complex UWA environments.
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (5) , 9782-9794
A Secure-Enhanced Resource Allocation Scheme for OFDMA-Based Underwater Acoustic Sensor Networks
Physical layer security (PLS), as a key technology for ensuring communication security, is widely applied in both point-to-point and multiuser communication scenarios due to its low computational and communicational overhead. However, executing physical layer security in underwater acoustic sensor networks (UASNs) poses considerable challenges in multiuser scenarios where hybrid attackers are present. To enhance the security performance of orthogonal frequency division multiple access (OFDMA)-based UASNs, we first model the secrecy capacity under the scenario of hybrid attackers. Then, a joint subcarrier and power allocation algorithm named MWOM to maximize the total secrecy capacity of multiple users is proposed. In addition, an overlapping window-based dynamic clustering scheme is applied to reduce the feedback delay of the communication link. Finally, to reduce the complexity of the proposed allocation algorithm, we map the resource allocation problem to a bipartite graph model and derive the suboptimal solution of the problem by leveraging graph theory. Simulation results verify that the proposed algorithm outperforms conventional algorithms in security performance, with negligible impact on communication performance. More importantly, we conducted actual sea trials in real underwater environments, which consistently confirmed these results.
- 18 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (5) , 9782-9794
A Secure-Enhanced Resource Allocation Scheme for OFDMA-Based Underwater Acoustic Sensor Networks
Physical layer security (PLS), as a key technology for ensuring communication security, is widely applied in both point-to-point and multiuser communication scenarios due to its low computational and communicational overhead. However, executing physical layer security in underwater acoustic sensor networks (UASNs) poses considerable challenges in multiuser scenarios where hybrid attackers are present. To enhance the security performance of orthogonal frequency division multiple access (OFDMA)-based UASNs, we first model the secrecy capacity under the scenario of hybrid attackers. Then, a joint subcarrier and power allocation algorithm named MWOM to maximize the total secrecy capacity of multiple users is proposed. In addition, an overlapping window-based dynamic clustering scheme is applied to reduce the feedback delay of the communication link. Finally, to reduce the complexity of the proposed allocation algorithm, we map the resource allocation problem to a bipartite graph model and derive the suboptimal solution of the problem by leveraging graph theory. Simulation results verify that the proposed algorithm outperforms conventional algorithms in security performance, with negligible impact on communication performance. More importantly, we conducted actual sea trials in real underwater environments, which consistently confirmed these results.
- 21 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (1) , 37-53
An Underwater Secure Localization Scheme Based on Physical Layer Cryptographic Learning
In open underwater environments, ensuring accurate positions of sensors while protecting private information of localization systems presents a significant challenge. The physical channel differences between terrestrial and underwater networks render most existing privacy protection schemes designed for terrestrial networks inapplicable underwater. Moreover, limited research on underwater privacy protection has led to high implementation complexity and communication expenses. In this paper, to reduce the complexity of privacy protection, a secure mobile localization scheme using autonomous underwater vehicles (AUVs) as anchors is proposed for underwater sensor networks, based on adversarial neural cryptography utilizing acoustic channel features. Depending on whether eavesdroppers show interest in keys, two adversarial cryptography models are proposed to protect transmission of legitimate localization information and to actively counter eavesdroppers with learning capabilities in real time. Furthermore, to obtain effective keys and minimize unnecessary key transmission, random physical layer channel features are dynamically utilized as real-time keys for the cryptography system, and a synchronous channel probing protocol is designed for key generation. Simulation and experimental results demonstrate that, compared to other approaches, the proposed secure localization scheme effectively prevents the leakage of position information and maintains localization accuracy while operating at lower implementation complexity and communication expenses.
- 21 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (1) , 37-53
An Underwater Secure Localization Scheme Based on Physical Layer Cryptographic Learning
In open underwater environments, ensuring accurate positions of sensors while protecting private information of localization systems presents a significant challenge. The physical channel differences between terrestrial and underwater networks render most existing privacy protection schemes designed for terrestrial networks inapplicable underwater. Moreover, limited research on underwater privacy protection has led to high implementation complexity and communication expenses. In this paper, to reduce the complexity of privacy protection, a secure mobile localization scheme using autonomous underwater vehicles (AUVs) as anchors is proposed for underwater sensor networks, based on adversarial neural cryptography utilizing acoustic channel features. Depending on whether eavesdroppers show interest in keys, two adversarial cryptography models are proposed to protect transmission of legitimate localization information and to actively counter eavesdroppers with learning capabilities in real time. Furthermore, to obtain effective keys and minimize unnecessary key transmission, random physical layer channel features are dynamically utilized as real-time keys for the cryptography system, and a synchronous channel probing protocol is designed for key generation. Simulation and experimental results demonstrate that, compared to other approaches, the proposed secure localization scheme effectively prevents the leakage of position information and maintains localization accuracy while operating at lower implementation complexity and communication expenses.
- 22 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 26009-26018
A Mobile-Anchor-Based Underwater Node Localization Mechanism With Improved Particle Swarm Optimization Algorithm
To mitigate the impact of the stratification effect, clock asynchrony, and measurement noise on underwater node localization while overcoming the restriction of the limited energy of unlocalized nodes, we propose a mobile-anchor-based underwater node localization mechanism. Specifically, we first utilize the information periodically broadcast by vertically mobile anchor nodes to calculate the clock skew and the time difference of arrival (TDoA). Then, we use the analytical expression of propagation delay based on an isogradient sound speed profile to compute the estimated TDoA, leading to the formulation of an optimization problem for node localization. Finally, we design an adaptive particle swarm optimization algorithm based on fitness value (APSO-F) to obtain the position of unlocalized nodes efficiently. Simulation results demonstrate that, compared with other localization solutions, the proposed mechanism effectively alleviates the influence of various unfavorable factors, ensuring high-precision node localization by one-way communication with anchor nodes, and APSO-F exhibits a superior convergence rate compared to the ordinary adaptive particle swarm optimization algorithm.
- 22 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (13) , 26009-26018
A Mobile-Anchor-Based Underwater Node Localization Mechanism With Improved Particle Swarm Optimization Algorithm
To mitigate the impact of the stratification effect, clock asynchrony, and measurement noise on underwater node localization while overcoming the restriction of the limited energy of unlocalized nodes, we propose a mobile-anchor-based underwater node localization mechanism. Specifically, we first utilize the information periodically broadcast by vertically mobile anchor nodes to calculate the clock skew and the time difference of arrival (TDoA). Then, we use the analytical expression of propagation delay based on an isogradient sound speed profile to compute the estimated TDoA, leading to the formulation of an optimization problem for node localization. Finally, we design an adaptive particle swarm optimization algorithm based on fitness value (APSO-F) to obtain the position of unlocalized nodes efficiently. Simulation results demonstrate that, compared with other localization solutions, the proposed mechanism effectively alleviates the influence of various unfavorable factors, ensuring high-precision node localization by one-way communication with anchor nodes, and APSO-F exhibits a superior convergence rate compared to the ordinary adaptive particle swarm optimization algorithm.
- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (2) , 924-938
Secure Localization for Underwater Wireless Sensor Networks via AUV Cooperative Beamforming With Reinforcement Learning
In harsh underwater environments, the localization of network nodes faces severe challenges due to open deployment environments. Most existing underwater localization methods suffer from privacy leaks. However, privacy protection schemes applied in terrestrial networks are not viable for underwater acoustic networks due to stratification effects and multipath complexities. In this paper, we introduce a secure localization scheme for underwater wireless sensor networks (UWSNs) utilizing cooperative beamforming among mobile underwater anchor nodes. With this scheme, the underwater sensor communicates and ranges with mobile anchor nodes to perform self-localization via time difference of arrival (TDOA) algorithm. However, the presence of eavesdroppers poses a threat by intercepting information emitted by the anchors. To avoid localization information leakage, then we model the secure localization requirement as a multi-anchors multi-objective dual joint optimization problem to enhance both security and energy performance. The deep reinforcement learning (DRL)-based multi-agent deep deterministic policy gradient (MADDPG) algorithm is applied to solve the optimization problem. Both simulation and field experimental results robustly validate the efficiency and accuracy of the proposed secure localization scheme.
- 2 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (2) , 924-938
Secure Localization for Underwater Wireless Sensor Networks via AUV Cooperative Beamforming With Reinforcement Learning
In harsh underwater environments, the localization of network nodes faces severe challenges due to open deployment environments. Most existing underwater localization methods suffer from privacy leaks. However, privacy protection schemes applied in terrestrial networks are not viable for underwater acoustic networks due to stratification effects and multipath complexities. In this paper, we introduce a secure localization scheme for underwater wireless sensor networks (UWSNs) utilizing cooperative beamforming among mobile underwater anchor nodes. With this scheme, the underwater sensor communicates and ranges with mobile anchor nodes to perform self-localization via time difference of arrival (TDOA) algorithm. However, the presence of eavesdroppers poses a threat by intercepting information emitted by the anchors. To avoid localization information leakage, then we model the secure localization requirement as a multi-anchors multi-objective dual joint optimization problem to enhance both security and energy performance. The deep reinforcement learning (DRL)-based multi-agent deep deterministic policy gradient (MADDPG) algorithm is applied to solve the optimization problem. Both simulation and field experimental results robustly validate the efficiency and accuracy of the proposed secure localization scheme.