Miao Ma

Ms. Miao Ma

orcid

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
5 June 2023 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZXZhong xiang
Zhong xiang
YSYujia Shen
Yujia Shen
MMMiao Ma
Miao Ma
MQMiao Qian
Miao Qian
Automatic defect detection is a critical stage of quality control in the textile industry. Because defects vary in texture structure, size, and spatial distribution, existing defect detection methods have difficulty in achieving a perfect trade-off between detection efficiency, accuracy, and generalizability. This paper presents an advanced fabric defect detection model based on a deep learning algorithm. A novel parallel dilated attention module was designed: this is nested in the deep layers of the neural network backbone to establish global channel dependencies and capture multiscale contextual information. Further, a hook-shaped feature pyramid network was developed for multiscale context aggregation: this directs the network to focus on lower-level features and is helpful for further improving detection efficiency without sacrificing detection accuracy. In addition, the Alpha-GIoU loss is used to improve the accuracy of bounding box regression because of its modulated power parameter α. The advantages and effectiveness of the proposed method were statistically analyzed using data from several public datasets. The code is available at https://github.com/aabb605/HookNet.

26 June 2025 in Journal of Lightwave Technology
Verified
Journal Article
Closed access
TZTao Zhang
Tao Zhang
WNWangsheng Niu
Wangsheng Niu
WMWeichao Ma
Weichao Ma
JWJiahao Wang
Jiahao Wang
MMMiao Ma
Miao Ma
MXMingjun Xia
Mingjun Xia
Anapole resonance has recently attracted increasing attention due to its non-radiating characteristics. However, suitable nanostructures supporting anapole resonance still need to be found, and its practical applications remain largely unexplored. At the same time, the development of miniaturized, high-performance devices for terahertz-band signal processing, such as filters and modulators, remains limited. This lack of progress significantly hinders advancements in fields such as terahertz communications and imaging. In this study, we designed a novel metasurface structure (capacitor mesh) that supports anapole resonance. This structure exhibits strong polarization independence, high incident angle insensitivity, topological connectivity, and tunable resonance frequency through geometric parameter adjustments. Using this metasurface, we proposed and implemented a terahertz bandpass filter with a designed center frequency of 0.4 THz. Due to its anapole resonance properties, the device offers significant advantages in minimizing reflection loss. It was fabricated and characterized, with transmittance spectra revealing low energy loss (28%) at the center frequency of 0.39THz and a bandwidth of 60 GHz. Additionally, we demonstrated the metasurface's potential for use in terahertz modulators. Our work underscores the significant potential of anapole resonance in terahertz field manipulation and terahertz devices.

15 May 2025 in Photonics
Verified
Journal Article
Open access
PDF file
HZHangfeng Zhou
Hangfeng Zhou
MMMiao Ma
Miao Ma
CZChenwei Zhang
Chenwei Zhang
XZXinlong Zhao
Xinlong Zhao
WMWeichao Ma
Weichao Ma
MXMingjun Xia
Mingjun Xia
With the increasing communication frequencies in 6G networks, high-speed terahertz (THz) modulation signal generation has become a critical research area. This study first proposes an on-chip high-speed THz modulation signal generation system based on lithium niobate (LN), which integrates a pair of racetrack resonator-integrated Mach–Zehnder modulators (RMZMs) with a chirped periodically poled lithium niobate (CPPLN) waveguide. The on-chip system combines near-infrared electro-optic modulation and cascaded difference-frequency generation (CDFG) for high-speed THz modulation signal generation. At 300 K, utilizing two input optical waves at frequencies of 193.55 THz and 193.14 THz, this on-chip system enables high-speed THz modulation signal generation at 0.41 THz, with a 1 Gbit/s modulation rate and a 0.25 V modulation voltage. During the simulation, when the intensity of the input optical waves is 1000 MW/cm2, the generated 0.41 THz signal reaches a peak intensity of 21.24 MW/cm2. Furthermore, based on theoretical analysis and subsequent simulation, the on-chip system is shown to support a maximum modulation signal generation rate of 7.75 Gbit/s. These results demonstrate the potential of the proposed on-chip system as a compact and efficient solution for high-speed THz modulation signal generation.

1 September 2026 in Journal of Lightwave Technology
Verified
Journal Article
Closed access
TZTao Zhang
Tao Zhang
YWYingjie Wang
Yingjie Wang
CZChenwei Zhang
Chenwei Zhang
WMWeichao Ma
Weichao Ma
MMMiao Ma
Miao Ma
MXMingjun Xia
Mingjun Xia

13 July 2026 in Applied Sciences
Verified
Journal Article
Open access
PDF file
MMMiao Ma
Miao Ma
DTDebang Tang
Debang Tang
ZGZenan Gan
Zenan Gan
XGXinyi Geng
Xinyi Geng
XZXinlong Zhao
Xinlong Zhao
HPHaipeng Pan
Haipeng Pan
Reliable and efficient methods to identify Parkinson’s disease (PD) from EEG signals are highly valuable for studying early disease markers and supporting timely intervention research. In this work, we propose a lightweight deep learning framework for PD detection using electroencephalogram (EEG) signals. The framework employs a novel spatial–frequency multiscale feature fusion strategy to characterize both local neural activity and global inter-channel connectivity. To achieve high accuracy with low computational cost, we design a hybrid network combining the DCFConv lightweight convolution module and the OSGAttn orthogonal global attention module. Experiments on the public Iowa PD-EEG dataset show that our method achieves 99.81% accuracy, 100% precision, and 99.62% sensitivity, while maintaining only 0.16 M parameters and 44.08 M FLOPs. The results demonstrate that the proposed approach provides an effective, efficient, and reliable solution for PD EEG signal classification.