JSJiayao Sun

Jiayao Sun

*This is an algorithmically generated author record based on publicly available information.

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


Jiayao Sun has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Fading-Induced Phase-Fluctuation Reduction and SNR Enhancement in ϕ -OTDR Using Video Block-Matching 3-D Collaborative Processing

    by Jiayao Sun, Deyu Xu, Jingming Zhang, Qi Mu, Xiaobing Shi, Peide Zhang, Yuanhang Zhuo, Liyang Shao

  • Adaptive 1-bit Compressed Sensing for High-Speed and High-compression Φ-OTDR Systems

    by Jiayao Sun, Deyu Xu, Jingming Zhang, Qi Mu, Nong Tian, Xiaobing Shi, Peide Zhang, Shanfan Lu, Chaoxian Qiu, Liyang Shao

    Massive data generation in distributed acoustic sensing (DAS) systems based on phase-sensitive optical time-domain reflectometry (Φ-OTDR) poses significant challenges for data transmission and storage. To address this issue, this paper proposes an adaptive 1-bit compressed sensing (CS) algorithm for DAS signal compression and reconstruction. In the compression stage, signal sparsity is adaptively estimated from spectral-energy statistics and morphological constraints. In the reconstruction stage, an adaptive reconstruction scheme based on the fast iterative shrinkage-thresholding algorithm (FISTA) is employed to suppress quantization-induced distortion and improve reconstruction stability and efficiency. Experiments were conducted on a laboratory standard 500 Hz sinusoidal signal, three known real-world intrusion scenarios, and one unseen subway-operation scenario. For the laboratory signal, the proposed method achieved a compression ratio of 148.58, a runtime of 6.83 ms, and a Pearson correlation coefficient (PCC) of 0.99. For the three known intrusion scenarios, the average compression ratio remained 143.27, with an average runtime of 9.45 ms, an average signal-to-noise ratio (SNR) of 75.40 dB, and an average PCC of 0.86. For the unseen subway-operation signal, the proposed method still achieved a compression ratio of 145.65, an SNR of 72.57 dB, a PCC of 0.86, and a runtime of 9.25 ms. Compared with Orthogonal Matching Pursuit (OMP) and Binary Iterative Hard Thresholding (BIHT), the proposed method provides substantially higher compression efficiency and lower reconstruction time. Compared with the lightweight deep-learning-based 1-bit adaptive delta modulation convolutional neural network (1-ADM-CNN method), it maintains a much higher compression ratio and better adaptability under unseen complex field conditions. These results demonstrate that the proposed method provides an effective and practical solution for real-time DAS monitoring in complex engineering environments.

  • Label-Efficient Urban Traffic Monitoring Using Dark-Fiber Distributed Acoustic Sensing

    by Jingming Zhang, Yinghuan Li, Yuanhang Zhuo, Jiayao Sun, Jinjun Wang, Shanfan Lu, Xiaobing Shi, Peide Zhang, Weihao Lin, Alan Pak Tao Lau, et al.

  • Ultra-High Compression and Rapid Reconstruction of DAS Data Based on Adaptive Compressed Reconstruction Algorithm

    by Jiayao Sun, Deyu Xu, Jingming Zhang, Qi Mu, Zetian Sang, Huaxin Gu, Xiaobing Shi, Peide Zhang, Ke Yi, Liyang Shao

    Distributed acoustic sensing (DAS) systems face critical challenges in data transmission, storage, and real-time processing due to the extremely high density of signal acquisition in large-scale deployments. Conventional compressed sensing methods often rely on manually predefined parameters, exhibit poor robustness under complex noise conditions, and suffer from low reconstruction accuracy and high computational cost. To overcome these limitations, this study proposes an adaptive compressive sampling pursuit (ACoSaMP) algorithm that dynamically estimates signal sparsity from the frequency-band energy distribution, thereby eliminating manual tuning. Integrated into the CoSaMP framework, the proposed approach reduces computational complexity while enhancing compression efficiency. Experimental validation in both laboratory experiments and real-world scenarios involving three representative external disturbances (rammer operation, manual excavation, and vehicle/subway vibration) demonstrates that ACoSaMP achieves ultra-low compression ratios of 1.1%, reconstruction times of 25 ms, 85 ms, and 81 ms, and Pearson correlation coefficients (PCC) consistently above 0.9. Compared with orthogonal matching pursuit (OMP), ACoSaMP improves compression efficiency by a factor of 19.2 and accelerates reconstruction speed by 18.5 times. In addition, support vector machine (SVM) classification of the reconstructed signals achieves 95.3% accuracy, exceeding that obtained with uncompressed signals by 5.5%.


  • Laser-Frequency-Drift Compensation in Φ-OTDR Using Linear Fitting Intercepts Method

    by Qi Mu, Deyu Xu, Feihong Yu, Jiayao Sun, Jingming Zhang, Zetian Sang, Xingwei Chen, Huaxin Gu, Ke Yi, Liyang Shao

    With many significant advantages, the phase-sensitive optical time-domain reflectometry (Φ-OTDR) system has become a popular monitoring instrument, especially in geophysical research, such as seismic wave observation. In some scenarios, low-frequency vibration detection provides important clues for analyzing changes in target behavior. However, the ever-present frequency drift in lasers poses a challenge for sub-hertz vibration detection. In this paper, we propose a linear fitting method for taking intercepts to reduce the influence of laser frequency drift in heterodyne Φ-OTDR systems. Experimental results show that the method can effectively improve the signal-to-noise ratio (SNR) of 0.5 Hz sinusoidal signals by 23.909 dB over a 5 km long sensing fiber. Additionally, the variance of the compensation results at 40 different positions is 0.666 (dB)2, which is 15.6 times better than the variance of 10.397 (dB)2 obtained using the conventional compensation method under the same conditions, indicating good stability in compensation at different positions. Using this method, we also succeeded in improving the SNR of the perturbed signals of 0.1 Hz and 0.05 Hz by 8.575 dB and 6.056 dB, respectively, at 2 km, effectively reducing the distortion of the original demodulation curves. Benefiting from this, the proposed method can further improve the low-frequency detection performance of the Φ-OTDR system.


  • Open Access

    Explainable Riemannian Manifold Learning for Application Scene Generalization With Distributed Acoustic Sensing System

    by Puyu Han, Jingming Zhang, Xingwei Chen, Guanfang Shao, Zixing Huang, Jiayao Sun, Xiaobing Shi, Xiongji Yang, Liyang Shao

    Intrusion event detection in distributed fiber optic sensing systems plays a vital role in urban emergency management and public safety. However, current AI-based systems face high false positive and false negative rates, as well as limited generalizability, due to variations in dataset quality, processing methods, and model architectures. To address these issues, we propose the Riemannian-Transformer-OTDR (RTO) model, an explainable deep-learning approach based on Riemannian manifolds. By exploiting the manifold space, our method identifies stable latent representations of dynamic data distributions from similar events, embedding intrusion detection within this latent manifold. The RTO model accurately identifies identical event types across diverse scenarios, reducing misclassification rates and enhancing model stability. Field tests using a distributed fiber optic sensing system validated the effectiveness of RTO: it achieved 98.43% accuracy in a subway construction scenario and demonstrated its cross-scenario generalization ability by maintaining 94.21% accuracy in gas pipeline monitoring, with 80.79% accuracy in a 100% noisy environment. Additionally, signal structure analysis of misclassified samples revealed insights into the model’s learning process. Our analysis of performance degradation across scenarios and noise levels further clarifies the RTO mechanisms and explainability. Ultimately, the RTO enables precise, retraining-free intrusion event classification across multiple scenarios, offering a practical solution for rapid event detection deployment in various domains.


  • Listening to the Underwater Acoustic Based on Fiber-Optic Tweezers Technology

    by Jiayao Sun, Lun Yan, Chunlei Jiang, Yunkai Wang, Tao Wang, Taiji Dong, Yu Sun, Peng Chen, Xiufang Wang, Hongwei Liang, et al.

    We have explored a new way to achieve ultrasensitive detection of acoustic waves in an aqueous environment based on fiber-optic tweezers technology. Silicon dioxide particle trapped by the pico-Newton scale’s light force is used as underwater acoustic wave sensors. When the acoustic signal propagates in water, it will cause microscale vibration of the particle in the light trap, resulting in small changes in the distance between the particle and the tip of the fiber optic and a change in the light range difference between the reflected beams, and the original underwater acoustic wave signals are obtained based on the resulting double-beam interference effect. Experiments show that, by collecting the movement information of particle in the optical trap under the influence of acoustic waves, it is possible to listen to acoustic signals with sound power as low as -67.135dB and real-time monitoring of acoustic signals within the frequency range of 5Hz-4kHz, with a signal-to-noise ratio of up to 72.073dB. The new underwater acoustic sensing mode proposed in this paper can overcome the existing acoustic sensing technology’s limitations and provide a new direction for the sensing field of optical fiber tweezers technology.


  • Fiber-Optic Probes for Ring-Shaped Multiparticle Capture

    by Tao Wang, Linzhi Yao, Bingkun Gao, Chunlei Jiang, Yunkai Wang, Zhaoi Qan, Jiayao Sun, Taiji Dong, Yu Sun, Penghui Dai

    Conventional single-fiber optical tweezers usually capture particles at the front or side end of the tip of a fiber probe, thus enabling optical manipulation of particles in a limited range. In this paper, we design and fabricate a novel single-fiber probe, which is prepared by integrating a common single-mode fiber (SMF) and a silica capillary microtubular fiber (COF), and the excitation of LP 21 higher-order modes in fiber-optic probes, and its output light field has multiple focused light traps, which enhances the intensity of the light field at the edge of the light probe, and capable of capturing and manipulating multiple 3μm diameter PS particles simultaneously in a circular multi-directional manner thus breaking the limitation of the direction of the light capture, and expanding the spatial range of the light capture. Simulation and experimental results demonstrate the feasibility of the fiber probe. The excitation and application of multimode beams in this new fiber probe improves the capture efficiency of single fiber probes and is easy to prepare, it enriches the functionality of a single fiber optic probe, which is expected to be applied in the field of micro-objects research, such as the collection of multiple cells in biomedicine and the manipulation of multiple particles.


  • Micro-Nano Fiber Flexible Multimodal Sensors for Fingerprint Recognition

    by Yunkai Wang, Xianli Yu, Chunlei Jiang, Tao Wang, Taiji Dong, Yu Sun, Xu Liu, Lun Yan, Yan Lu, Jiayao Sun

    Multimodal biometric sensing and processing systems can significantly improve the success rate of identification and authentication compared to traditional unimodal biometric authentication techniques. We propose a Flexible micro-nanofiber (MNF) multimodal sensor for fingerprint recognition. We used PDMS as the sensor substrate and placed the MNF in s-shape on the PDMS. The surface of the PDMS is covered with a film with a thickness of only 2 μm to package the MNF. A ridge-shaped sensing area is formed on the sensor surface. Such structure protects the NMF from damage at the same time. It also greatly preserves the sensitivity of the NMF. This method realizes the simultaneous measurement of the fingerprint signal and the touch dynamics signal. We tested the stability of the sensors when measuring friction and pressure as well as temperature changes. We performed the recognition of 400 sets of fingerprint signals with an accuracy of 95.75%. Such method has great potential for applications in communication, computing, commercial purposes, and financial transactions.