Qi Jiang

Qi Jiang

Auto-generated profile, based on publicly available information

Qi Jiang

School of Electronic Science and Engineering, Nanjing University, Nanjing, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 April 2024 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
ZCZhenxin Cai
Zhenxin Cai
YWYu Wang
Yu Wang
QJQi Jiang
Qi Jiang
GGGuan Gui
Guan Gui
JSJin Sha
Jin Sha
The inherent flexibility of small unmanned aerial vehicles (UAVs) enables their deployment across various emerging markets. Unauthenticated UAVs pose a significant threat if they intrude into aviation-sensitive areas. To address this issue, deep learning (DL)-based radio frequency fingerprint identification (RFFI) has been developed as a promising approach for identifying illegal UAVs. However, these commonly used DL-based methods demand high computation and storage requirements, which are not suitable for the deployment of RFFI. In this paper, we propose an efficient and low-complexity RFFI method for UAV identification. Specifically, we design a lightweight backbone network consisting of lightweight multi-scale convolution (LMSC) blocks that can significantly reduce the model size and enhance the feature extraction ability. The simulation results indicate that our proposed UAV RFFI method outperforms other state-of-the-art and popular DL-based RFFI methods in terms of both identification performance and complexity. The identification accuracy surpasses that of all other methods at low signal-to-noise ratios (SNRs) and achieves nearly 100% accuracy at high SNRs. To further enhance model efficiency, we employ data truncation in our experimental simulations, demonstrating that a sample length of 2000 is sufficient to retain high identification performance. Additionally, we incorporate the Mixup regularization strategy, which improves accuracy without increasing the complexity, especially as sample length decreases.

19 July 2023 in IEEE Transactions on Wireless Communications
Verified
Journal Article
Closed access
QJQi Jiang
Qi Jiang
JSJin Sha
Jin Sha
The explosive growth of maritime vessels imposes high demands on the security of automatic identification system (AIS). Radio frequency fingerprinting identification (RFFI) as a physical-layer authentication method offers a new perspective on security solutions for wireless communication systems. However, RFFI for AIS satellite component suffers from poor performance under low signal-to-noise ratio (SNR) scenarios. To this end, this paper proposes a data-driven RFFI method with high performance and resilience for AIS satellite links. The bivariate variational mode decomposition (VMD) enables adaptive signal decomposition for effective denoising. The raw I/Q samples after bivariate VMD are directly taken as input to the RF fingerprinting feature extraction network without any transformation process. The complex-valued neural network used for feature extraction provides a superior representation capability compared to traditional real-valued neural networks. Furthermore, channel attention mechanisms are embedded in the feature extraction network to shed light on the correlation between channels within the signal. The following integrated spatial attention mechanisms are employed to reveal the correlations between signals. Numerical results show that the proposed RFFI method can achieve over 90.37% accuracy on real-world datasets with SNRs greater than 6 dB, and achieve a balance between complexity and performance.

5 April 2023 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
QJQi Jiang
Qi Jiang
JSJin Sha
Jin Sha
data exchange system (VDES) is explored as a navigational safety guarantor for ships worldwide. As the number of jammers and spoofers targeting VDES is on the rise, its security authentication emerges as an urgent issue. Radio frequency fingerprinting identification (RFFI) as a noncryptographic physical-layer security solution can effectively defend against increasingly sophisticated attacks. Nevertheless, the power consumption of frequent RFFI is not negligible for VDES satellite components with energy constraints. To this end, spiking neural networks (SNNs) are first employed to build an ultralow-power RFFI system for VDES, where the optimized spiking neurons can yield high accuracy with very few spikes. Moreover, multiple RF fingerprinting features are fused to highly integrate the information of the VDES signal in different dimensions, and the attention mechanism is utilized to further improve RFFI performance. Numerical results show that our SNN-based RFFI system can yield identification accuracy up to 92.59% at a signal-to-noise ratio (SNR) of 25 dB on VDES data sets and reduces power consumption by 64% compared to artificial neural networks (ANNs) of comparable accuracy on 45-nm CMOS process.

23 November 2022 in IEEE Wireless Communications Letters
Verified
Journal Article
Closed access
QJQi Jiang
Qi Jiang
JSJin Sha
Jin Sha
Low earth orbit (LEO) satellites promise to be a source of innovation in wireless communication and their physical layer security deserves attention. However, many physical layer security techniques are hampered by excessive power consumption and high latency when used for LEO satellites. This letter proposes an energy-efficiency radio frequency fingerprinting identification (RFFI) method for LEO satellite communication systems based on spiking neural networks (SNN). In addition, we investigate a channel-independent RFF feature transformation and data augmentation to enhance system robustness. Numerical results show that our proposed model can yield identification accuracy up to 95.26% at a signal-to-noise ratio (SNR) of 25dB on orthogonal frequency division multiplexing (OFDM) signals, and reduces power consumption by 63.3% compared to existing models of comparable accuracy on FPGA.

13 June 2024 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
QJQi Jiang
Qi Jiang
JSJin Sha
Jin Sha
Interference against global navigation satellite system (GNSS) is threatening its reliability. Radio frequency fingerprint identification (RFFI) emerges as a physical-layer security solution that can effectively identify genuine transmitters. However, external noise in the transmission is not conducive to maintaining the robustness of the RFFI, and the deep neural networks (DNNs) or large datasets for boosting robustness will consume excessive resources. To this end, this letter proposes a lightweight RFFI scheme based on variational autoencoder (VAE) and long short-term memory (LSTM) for real-field GPS signals collected in Nuremberg. The VAE aims to denoise and reconstruct the RFFs, thereby improving the identification accuracy and reducing feature dimensionality. LSTMs can extract the RFF features without any pre-transformation and avoid the problem of gradient vanishing or gradient exploding. Numerical results demonstrate that our model can yield an identification accuracy of up to 95.68% on post-correlation GPS data at low complexity.

2 September 2024 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
QJQi Jiang
Qi Jiang
JSJin Sha
Jin Sha
Radio frequency fingerprint identification (RFFI) offers a paradigm for the security authentication of the satellite component in automatic identification system (SAT-AIS). This paper proposes a data-driven RFFI solution for SAT-AIS with multiple communication scenarios. Conditional generative adversarial networks (CGANs) embedded with channel estimation can generate numerous realistic samples for data augmentation. Both the generator and the discriminator in CGANs employ deep neural networks to adaptively extract RFF features thus enhancing the robustness of this system. Moreover, in light of the dilemma of collecting large datasets, Bayesian meta-learning via variational autoencoder is performed for model training to enable few-shot learning, where only a small number of samples are required to complete the inference for efficient RFFI in multi-scenario communications. Experimental results show that the proposed RFFI framework outperforms existing methods by achieving up to 97.30% accuracy on datasets with signal-to-noise ratios (SNRs) of 10dB.