AUTHOR=Qin Zhenkai , Wei Baozhong , Gao Caifeng , Chen Xiaolong , Zhang Hongfeng , In Wong Cora Un 

TITLE=SFDformer: a frequency-based sparse decomposition transformer for air pollution time series prediction

JOURNAL=Frontiers in Environmental Science

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/environmental-science/articles/10.3389/fenvs.2025.1549209

DOI=10.3389/fenvs.2025.1549209

ISSN=2296-665X

ABSTRACT=IntroductionWith the rapid advancement of industrialization and the prevalent occurrence of haze weather, PM2.5 contamination has emerged asa significant threat to public health and environmental sustainability. The concentration of PM2.5 exhibits intricate dynamic attributes and is profoundly correlated with meteorological conditions as well as the concentrations of other pollutants, thereby substantially augmenting the complexity of predictive endeavors.MethodsA novel predictive methodology has been developed. It integrates time seriesfrequency domain analysis with the decomposition of deep learning models. This approach facilitates the capture of interdependencies among high - dimensional features through time series decomposition, employs Fourier Transform to mitigate noise interference, and incorporates sparse attention mechanisms to selectively filter critical frequency components, thereby enhancing time - dependent modeling. Importantly, this technique effectively reduces computational complexity from O(L2) to O(L⁡log⁡L).ResultsEmpirical findings substantiate that this methodology yields notably superior predictive accuracy relative to conventional models across a diverse array of real-world datasets.DiscussionThis advancement not only offers an efficacious resolution for PM2.5 prediction tasks but also paves the way for innovative research and application prospects in the realm of complex time series modeling.