Xinyu Xie
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xinyu Xie has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
Hui Liang
1 shared publication- Jian Lan
Guangli Ruihong Electronic Technology Co., Ltd
1 shared publication - Jie Zhang
College of Electrical and Information Engineering, Zhengzhou University of Light Industry, Zhengzhou, China
1 shared publication - Jun Shi
The School of Electrical and Information Engineering, Zhengzhou University of Light Industry, Zhengzhou, China
1 shared publication - Jun Shi
College of Electrical and Information Engineering, Zhengzhou University of Light Industry, Zhengzhou 450000, P. R. China
1 shared publication Wanyong Liang
1 shared publication
Latest Publications
- 1 June 2026
- Published by IOP Publishing in Engineering Research Express
- Vol. 8 (11) , 115409
Hybrid signal-neural processing: MA-Fourier dual-branch forecasting for efficient time series prediction
Time series forecasting plays a critical role in industrial prognostics and health management, where both prediction accuracy and computational efficiency are essential. In practical industrial scenarios, forecasting models are often deployed under limited computational resources and must respond rapidly to dynamic operating conditions. This imposes significant challenges in achieving reliable predictions without incurring excessive computational cost. To address this challenge, we propose LeftNet, a lightweight hybrid signal-neural forecasting model with a dual-branch architecture. Specifically, a trend modeling module is designed to capture smooth temporal evolution through moving average filtering and multilayer perceptrons, while a high-frequency pattern module leverages Fourier-based spectral filtering to model short-term fluctuations and transient dynamics. By explicitly decoupling temporal patterns at different frequency scales, the proposed architecture enables efficient and interpretable feature extraction. Extensive experiments on both an industrial dataset (DST) and a public benchmark (ETT) demonstrate that LeftNet achieves competitive predictive performance with reduced computational cost. In particular, the proposed method maintains low prediction errors while achieving faster inference compared to representative baseline models. These results highlight the effectiveness of LeftNet in achieving a favorable trade-off between prediction accuracy and computational efficiency, making it well suited for real-time and resource-constrained forecasting applications.
- 1 June 2026
- Published by IOP Publishing in Engineering Research Express
- Vol. 8 (11) , 115409
Hybrid signal-neural processing: MA-Fourier dual-branch forecasting for efficient time series prediction
Time series forecasting plays a critical role in industrial prognostics and health management, where both prediction accuracy and computational efficiency are essential. In practical industrial scenarios, forecasting models are often deployed under limited computational resources and must respond rapidly to dynamic operating conditions. This imposes significant challenges in achieving reliable predictions without incurring excessive computational cost. To address this challenge, we propose LeftNet, a lightweight hybrid signal-neural forecasting model with a dual-branch architecture. Specifically, a trend modeling module is designed to capture smooth temporal evolution through moving average filtering and multilayer perceptrons, while a high-frequency pattern module leverages Fourier-based spectral filtering to model short-term fluctuations and transient dynamics. By explicitly decoupling temporal patterns at different frequency scales, the proposed architecture enables efficient and interpretable feature extraction. Extensive experiments on both an industrial dataset (DST) and a public benchmark (ETT) demonstrate that LeftNet achieves competitive predictive performance with reduced computational cost. In particular, the proposed method maintains low prediction errors while achieving faster inference compared to representative baseline models. These results highlight the effectiveness of LeftNet in achieving a favorable trade-off between prediction accuracy and computational efficiency, making it well suited for real-time and resource-constrained forecasting applications.
- Open Access
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Frequency-spatiotemporal synergistic network for high-precision semiconductor wafer dicing quality prediction
Semiconductor wafer dicing quality prediction requires micrometer-level precision, yet faces three critical challenges: scarce labeled data, complex spatiotemporal dependencies, and fine-grained feature preservation. Traditional physics-based modeling approaches suffer from extensive parameter calibration and fail to capture nonlinear dynamics, limiting their applicability in practical manufacturing scenarios. To address these challenges, this paper propose FreStyle, the first deep learning framework specifically designed for semiconductor wafer dicing quality prediction. To capture fine-grained features, the framework constructs a frequency-spatiotemporal synergistic dual-branch architecture that overcomes the over-smoothing problem of conventional methods. Specifically, the spatiotemporal branch employs depthwise separable 3D convolutions and adaLN-Zero Transformer to model long-range spatiotemporal dependencies, while the frequency branch introduces complex-valued convolutions, constructs analytic signals via Hilbert transform and applies Fast Fourier Transform, with an Adaptive Radial Energy Partition (AREP) module designed to separate low- and high-frequency components and perform differentiated attention processing. To effectively fuse heterogeneous features, a Cross-Domain Consistency-Driven Bidirectional Interaction (CDCI) block is constructed to achieve deep complementary enhancement through an interaction mechanism where spatiotemporal dynamics guide frequency-domain aggregation and frequency-domain priors calibrate spatiotemporal responses. Moreover, we construct the first public semiconductor wafer dicing dataset (SCD) meeting industrial precision standards for realistic evaluation. Experimental results on the SCD dataset show that FreStyle achieves an MSE of 0.0028 and an SSIM of 0.8961, corresponding to a 26.3% reduction in MSE and a 1.3% improvement in SSIM compared to the best baseline method. Extensive experiments on WeatherBench, TrafficBJ, and NSE datasets further validate the generalization capability of the proposed framework.
- 9 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessFrequency-spatiotemporal synergistic network for high-precision semiconductor wafer dicing quality prediction
Semiconductor wafer dicing quality prediction requires micrometer-level precision, yet faces three critical challenges: scarce labeled data, complex spatiotemporal dependencies, and fine-grained feature preservation. Traditional physics-based modeling approaches suffer from extensive parameter calibration and fail to capture nonlinear dynamics, limiting their applicability in practical manufacturing scenarios. To address these challenges, this paper propose FreStyle, the first deep learning framework specifically designed for semiconductor wafer dicing quality prediction. To capture fine-grained features, the framework constructs a frequency-spatiotemporal synergistic dual-branch architecture that overcomes the over-smoothing problem of conventional methods. Specifically, the spatiotemporal branch employs depthwise separable 3D convolutions and adaLN-Zero Transformer to model long-range spatiotemporal dependencies, while the frequency branch introduces complex-valued convolutions, constructs analytic signals via Hilbert transform and applies Fast Fourier Transform, with an Adaptive Radial Energy Partition (AREP) module designed to separate low- and high-frequency components and perform differentiated attention processing. To effectively fuse heterogeneous features, a Cross-Domain Consistency-Driven Bidirectional Interaction (CDCI) block is constructed to achieve deep complementary enhancement through an interaction mechanism where spatiotemporal dynamics guide frequency-domain aggregation and frequency-domain priors calibrate spatiotemporal responses. Moreover, we construct the first public semiconductor wafer dicing dataset (SCD) meeting industrial precision standards for realistic evaluation. Experimental results on the SCD dataset show that FreStyle achieves an MSE of 0.0028 and an SSIM of 0.8961, corresponding to a 26.3% reduction in MSE and a 1.3% improvement in SSIM compared to the best baseline method. Extensive experiments on WeatherBench, TrafficBJ, and NSE datasets further validate the generalization capability of the proposed framework.