基于动态适应性算法的多域感知与异构信息融合方法

侯泽宇; 徐君逸

doi:10.12382/bgxb.2025.0673

您当前的位置：

首页 >

文章列表页 >

基于动态适应性算法的多域感知与异构信息融合方法

更新时间：2026-04-24

- 基于动态适应性算法的多域感知与异构信息融合方法
- Multidomain Perception and Heterogeneous Information Fusion Method Based on Dynamic Adaptive Algorithm
- 兵工学报 2026年
- 作者机构：
  
  1. 中国矿业大学（北京）应急管理与安全工程学院,北京,100083
  2. 中兵智能创新研究院有限公司,北京,100072
  3. 中国人民大学公共管理学院,北京,100872
- 作者简介：
- 基金信息：
  
  中央高校基本科研业务费新进教师项目（2026XJJSAQ07）
- DOI：10.12382/bgxb.2025.0673
  中图分类号： TP391
- 收稿：2025-07-21，
  
  网络首发：2026-04-24，
- 稿件说明：
移动端阅览
侯泽宇，徐君逸. 基于动态适应性算法的多域感知与异构信息融合方法[J/OL]. 兵工学报, 2026(2026-04-24). https://doi.org/10.12382/bgxb.2025.0673.

HOU Z Y, XU J Y. Multidomain perception and heterogeneous information fusion method based on dynamic adaptive algorithm[J/OL]. Acta Armamentarii, 2026(2026-04-24). https://doi.org/10.12382/bgxb.2025.0673. (in Chinese)
侯泽宇，徐君逸. 基于动态适应性算法的多域感知与异构信息融合方法[J/OL]. 兵工学报, 2026(2026-04-24). https://doi.org/10.12382/bgxb.2025.0673. DOI：

HOU Z Y, XU J Y. Multidomain perception and heterogeneous information fusion method based on dynamic adaptive algorithm[J/OL]. Acta Armamentarii, 2026(2026-04-24). https://doi.org/10.12382/bgxb.2025.0673. (in Chinese) DOI：

摘要

针对复杂场景中高维异构数据融合面临的环境适应性差与边缘端计算开销大的难题，提出一种基于动态适应性算法的多域感知与异构信息融合框架。通过引入跨域协同机制、动态权重分配策略以及轻量化的边缘计算模型，有效应对传统方法在处理高异构性数据时所面临的融合效率低、交互成本高及跨域协同困难等瓶颈问题。实验结果表明，该方法在智慧城市场景中的多域融合目标检测准确率（mAP@0.5）提升至94.2%，在20%高噪声干扰的网络安全态势感知中异常检测F1分数仍保持84.3%，同时通过边缘云协同机制将边缘设备的能耗降低了40%、推理延迟缩短至28.7ms，为实现高效、鲁棒的多域感知与信息融合提供了新的技术路径。

Abstract

To address the challenges of poor environmental adaptability and high computational overhead on edge devices faced by high-dimensional heterogeneous data fusion in complex scenarios

this paper proposes a multi-domain perception and heterogeneous information fusion framework based on a dynamic adaptive algorithm. By introducing a cross-domain collaboration mechanism

a dynamic weight allocation strategy

and a lightweight edge computing model

it effectively addresses the bottleneck problems faced by traditional methods when processing highly heterogeneous data

such as low fusion efficiency

high interaction costs

and difficulties in cross-domain collaboration. Experimental resultsshow that the proposed method achieves a multi-domain fusion object detection accuracy (mAP@0.5) of 94.2% in smart city scenarios

and maintains an anomaly detection F1 score of 84.3% in network security situational awareness even under 20% high noise interference. Meanwhile

through the edge-cloud collaboration mechanism

it reduces the energy consumption of edge devices by 40% and shortens the inference latency to 28.7 ms

providing a new technical path for achieving efficient and robust multi-domain perception and information fusion.

关键词

Keywords

references

余拥军. 自动驾驶汽车多传感器融合技术研究[D].上海:同济大学,2019.

YU Y J.Research on multi-sensor fusion technology for autonomous vehicles[D].Shanghai: Tongji University, 2019.(in Chinese)

AHMED A A, MIHAELA A C, ALLAN J B, et al.Data quality challenges in largescale cyberphysical systems:a systematic review[J].Information Systems, 2022, 105.DOI:10.1016/j.is.2021.101951.

IPPA S, PRATEEK C, GOWTHAM R R, et al. Autonomous multi-sensor fusion techniques for environmental perception in self-driving vehicles[C]//Proceedings of 2024 International Conference on Communication, Computer Sciences and Engineering. Gautam Buddha Nagar, India: IEEE, 2024: 1146-1151.

SAHU A, MAO Z Y, WLAZLO P, et al. Multi-source MultiDomain data fusion for cyberattack detection in power systems[J].IEEE Access, 2021, 9:119118-119138.

DENG X H, ZHANG J J, ZHANG H G, et al. Deep-reinforcement-learning-based resource allocation for cloud gaming via edge computing[J]. IEEE Internet of Things Journal, 2023, 10(6):5364-5377.

刘爔炀. 面向物联网的隐私保护的数据共享协议研究[D]. 成都: 电子科技大学, 2024.

LIU X Y. Research on privacy-preserving data sharing protocols for the internet of things[D].Chengdu: University of Electronic Science and Technology of China, 2024.(in Chinese)

XUE H H, HUANG B, QIN M M, et al. Edge computing for internet of things: a survey[C]//Proceedings of 2020 International Conferences on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing(CPSCom) and IEEE Smart Data(SmartData) and IEEE Congress on Cybermatics(Cybermatics). Rhodes, Greece:IEEE, 2020:755-760.

SHAFER G.A mathematical theory of evidence[M]. Princeton, NJ, US: Princeton University Press, 1976.

LIU B L, LI Q, ZHENG Z H, et al. A review of multi-source data fusion and analysis algorithms in smart city construction: facilitating real estate management and urban optimization[J]. Algorithms 2025, 18(1): 30.

SHEN G Q , CHEN W C, ZHU B C, et al. DRL based binary computation offloading in wireless powered mobile edge computing[J].IET Communications, 2023,17(15):1837-1849.

ABDELLATIF A A, SAMARA L, MOHAMED A, et al.MEdge-chain:leveraging edge computing and blockchain for efficient medical data exchange[J]. IEEE Internet of Things Journal, 2021,8(21):15762-15775.

BREIMAN L. Random forests[M].Berlin, Germany: Machine Learning, 2001,45:5-32.

ZHANG R B, TAO T, WANG J B, et al.Graph neural network model under multi-modal fusion[C]//Proceedings of 2024 International Conference on Integrated Intelligence and Communication Systems. Kalaburagi, India:IEEE, 2024:1-5.

MALL P K, PSINGH R K, SRIVASTAV S, et al.A comprehensive review of deep neural networks for medical image processing:recent developments and future opportunities[J].Healthcare Analytics, 2023,4:100216.

JOSE R K, NIGAM C, KIRUBASRI G, et al. Real-time object detection on edge devices using mobile neural networks[C]//Proceedings of 2024 International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics. Bangalore, India: IEEE, 2024:1-4.

VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J].Advances in Neural Information Processing Systems, 2017, 30.

ZHOU J, CUI G Q, HU S D, et al. Graph neural networks: a review of methods and applications[J].AI Open, 2020, 1:57-81.

SENEL N, KEFFERPṺTZ K, DOYCHEVA K, et al. Multi-sensor data fusion for real-time multi-object tracking[J]. Processes, 2023, 11(2):501.

WANG J, YU L, TIAN S W. Crossattention interaction learning network for multimodel image fusion via transformer[J].Engineering Applications of Artificial Intelligence, 2025,139(Part A):109583.

DEVLIN J, CHANG M W, LEE K, et al. Bert: pre-training of deep bidirectional transformers for language understanding:arXiv:1819.04805[R].Ithaca,NY,US:Cornell University, 2018: 1819.04805.

LIU Y, ZHANG Y, WANG Y X, et al. A survey of visual transformers[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(6):7478-7498.

SHANKAR V. Edge AI: a comprehensive survey of technologies, applications, and challenges[C]//Proceedings of the 2024 1st International Conference on Advanced Computing and Emerging Technologies. Ghaziabad, India:IEEE, 2024:1-6.

LIU Q X.Application research and improvement of weighted information fusion algorithm and kalman filtering fusion algorithm in multi-sensor data fusion technology[J]. Sens Imaging,2023, 24.DOI:10.1007/s11220-023-00448-z.

MURPHY K P.Machine learning:a probabilistic perspective[M].Cambridge, MA,US: MIT Press, 2021.

浏览量

下载量

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据