Conference paper · 2026 · Open access

Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning

Liyu Wang

International Conference on Advanced Algorithms and Neural Networks (AANN), 447–451 · Published 1 August 2026 · DOI 10.1109/aann71145.2026.11709062

Open full textCite this paper

In brief

“Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning” is a 2026 conference paper by Liyu Wang, published in International Conference on Advanced Algorithms and Neural Networks (AANN). OpenAlex counts 0 citations as of 20 November 2026. It is open access under a CC BY licence: the full text is free to read at International Conference on Advanced Algorithms and Neural Networks (AANN).

Citations
0
References
12
Authors
1

Abstract

Urban surface waterlogging under short-duration heavy rainfall can change within minutes, whereas hydrodynamic models and cloud-based deep learning systems often fail to provide low-latency and reliable on-site warnings. This paper proposes an Edge Spatial-Temporal Lightweight Network (Edge-STLN) for multi-sensor urban flood early warning on edge neural processing units. Rainfall and water-level sequences are aligned by dynamic time warping, while surveillance images are downsampled and converted to HSV space. A temporal branch uses multi-scale dilated one-dimensional convolutions and gated linear units, and a visual branch combines depthwise separable convolution, channel shuffle, and learnable channel shrinkage. Cross-modal features are fused through an outer product and Tucker decomposition. Quantization-aware training and memory rearrangement are then applied for INT8 deployment. On a 5-TOPS edge module, Edge-STLN achieves 50-ms latency, 3.2-MB storage, 2.3-W power consumption, 96.7% recall, a 6.2% false-alarm rate, and only 0.51% accuracy drift after 12 h of continuous operation.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Wang, L. (2026). Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning. International Conference on Advanced Algorithms and Neural Networks (AANN), 447–451. https://doi.org/10.1109/aann71145.2026.11709062

MLA 9

Wang, Liyu. “Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning.” International Conference on Advanced Algorithms and Neural Networks (AANN), 2026, pp. 447–451. https://doi.org/10.1109/aann71145.2026.11709062.

Chicago

Wang, Liyu. 2026. “Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning.” International Conference on Advanced Algorithms and Neural Networks (AANN): 447–451. https://doi.org/10.1109/aann71145.2026.11709062.

Harvard

Wang, L. (2026) ‘Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning’, International Conference on Advanced Algorithms and Neural Networks (AANN), pp. 447–451. doi:10.1109/aann71145.2026.11709062.

IEEE

L. Wang, “Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning,” International Conference on Advanced Algorithms and Neural Networks (AANN), pp. 447–451, 2026, doi: 10.1109/aann71145.2026.11709062.

BibTeX

@inproceedings{wang2026urban,
  title     = {{Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning}},
  author    = {Wang, Liyu},
  booktitle = {International Conference on Advanced Algorithms and Neural Networks (AANN)},
  year      = {2026},
  pages     = {447--451},
  doi       = {10.1109/aann71145.2026.11709062},
  url       = {https://doi.org/10.1109/aann71145.2026.11709062}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Environmental Science › Global and Planetary Change › Flood Risk Assessment and Management

  • sensor fusion
  • edge computing
  • lightweight deep learning
  • dynamic time warping
  • surveillance video analysis
  • cross-modal feature fusion
  • Tucker decomposition
  • quantization-aware training

Work with this paper

Most-cited papers in Environmental Science

All highly cited Environmental Science papers

Questions about this paper

How do I cite this paper in APA style?

Wang, L. (2026). Urban Flood Multi-Sensor Data Fusion Early Warning Method Based on Edge Intelligence and Lightweight Deep Learning. International Conference on Advanced Algorithms and Neural Networks (AANN), 447–451. https://doi.org/10.1109/aann71145.2026.11709062

How many times has this paper been cited?

0 times, according to OpenAlex on 20 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at International Conference on Advanced Algorithms and Neural Networks (AANN): sign in and open the full text from this page.

Who wrote this paper?

Liyu Wang.

What is the DOI of this paper?

10.1109/aann71145.2026.11709062, permanently at https://doi.org/10.1109/aann71145.2026.11709062. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 20 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.