EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HAPQ: A Hardware-Aware Pruning and Quantization Pipeline for Event-Based SNN Detection

Zhengyinan Li, Jing Wu

Sensors · 06.05.2026 · 26 (9) · 2910

DOI: 10.3390/s26092910 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.05.2026. Verfasst von Zhengyinan Li und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Automotive Engineering, Wuhan University of Technology, Wuhan 430070, China. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Autonomous driving perception demands low latency, high temporal resolution, and stringent hardware efficiency. While event-based spiking neural networks (SNNs) offer bio-inspired sparse computation, their deployment on edge field-programmable gate arrays (FPGAs) is obstructed by irregular execution patterns and temporal state storage overhead. To address this, we propose HAPQ, a unified hardware-aware pruning and quantization pipeline for compact event-based object detection. Starting from an end-to-end adaptive sampling SNN detector (EAS-SNN), HAPQ conducts hardware-aware configuration search within discrete digital signal processor (DSP) and block RAM (BRAM) budgets, applies single-instruction-multiple-data (SIMD)-aligned structured pruning for computational regularity, and jointly quantizes synaptic weights and membrane potentials via a shift-friendly fixed-point recurrence. Evaluation on the Prophesee Gen1 dataset and an FPGA accelerator shows that HAPQ improves detection accuracy from 0.284 to 0.425 in mean average precision (mAP50:95) and achieves 0.722 mAP50. Hardware implementation reveals a reduction in lookup table (LUT) usage to 1680, complete DSP elimination, and a maximum operating frequency of 920.81 MHz at 0.630 W. These results confirm that effective temporal SNN deployment requires joint optimization of model architecture, state precision, and hardware-aligned workload organization.

Schlagwörter

Stichwörter der Autor:innen

FPGAevent-based object detectionhardware-aware co-designmembrane potential quantizationstructured pruning

Autor:innen und Institutionen

  • School of Automotive Engineering, Wuhan University of Technology, Wuhan 430070, China.
  • Hubei Provincial Key Laboratory of Modern Auto Parts Technology, Wuhan University of Technology, Wuhan 430070, China.

Zitieren

APA

Li, Z., & Wu, J. (2026). HAPQ: A Hardware-Aware Pruning and Quantization Pipeline for Event-Based SNN Detection. Sensors, 26(9), 2910. https://doi.org/10.3390/s26092910

Vancouver

Li Z, Wu J. HAPQ: A Hardware-Aware Pruning and Quantization Pipeline for Event-Based SNN Detection. Sensors. 2026;26(9):2910. doi:10.3390/s26092910

BibTeX
@article{Li2026,
  author = {Li, Zhengyinan and Wu, Jing},
  title = {HAPQ: A Hardware-Aware Pruning and Quantization Pipeline for Event-Based SNN Detection},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {9},
  pages = {2910},
  doi = {10.3390/s26092910}
}
Kostenlos starten →