Hierarchical decomposition of CNN for resource-constrained mechanical vibration WSN edge computing

被引：0
|
作者：
Fu H. [1 ]
Deng L. [1 ]
Tang B. [1 ]
Li Z. [1 ]
Wu Y. [1 ]
机构：
[1] State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing
来源：
Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument | 2024年 / 45卷 / 03期
关键词：
CNN; edge computing; MCU; mechanical vibration; resource constrain;
D O I：
10.19650/j.cnki.cjsi.J2312181
中图分类号：
学科分类号：
摘要：
The microcontroller of wireless sensor network (WSN) nodes used for mechanical vibration monitoring requires intricate edge computing, yet face limitations in hardware resources. Convolutional neural network (CNN), as a high-performance and commonly used deep learning algorithm, can enhance the computational capabilities of edge WSN nodes when run on microcontroller units (MCUs). This paper proposes a hierarchical decomposition method for CNN models without modification, addressing the challenge of running non-lightweight CNN on resource-constrained MCU and enhancing the computational capabilities of mechanical vibration WSN nodes. First, a file structure is designed to decompose and store CNN model parameters. Subsequently, a memory management method is proposed, and the consumption process of random-access memory is derived. Finally, a parameter localization method is introduced to accurately and efficiently retrieve model parameters. Experiments demonstrated that with only 1.76 KB of RAM and 2.14 KB of Flash, high-precision edge computing recognition tasks can be accomplished within 3.15 ms. © 2024 Science Press. All rights reserved.
引用
收藏
页码：94 / 105
页数：11
相关论文
共 22 条
  • [1] ZHU L L, TANG B P, HUANG Y, Et al., Transmit power optimization control method for a mechanical vibration wireless sensor network with a lot of transmission data, Vibration and Shock, 39, 17, pp. 275-280, (2020)
  • [2] XIAO X, TANG B P, DENG L, Et al., Accumulated synchronous acquisition error control method based on cross-layer design for mechanical vibration wireless sensor networks, Chinese Journal of Mechanical Engineering, 55, 15, pp. 202-207, (2019)
  • [3] SUN CH X, YANG L, WANG X P, Et al., Optimized Deployment method of edge computing network service function chain delay combined with deep reinforcement learning, Journal of Electronics and Information, pp. 1-10
  • [4] XUE J Q, SHI Y J, LI B., Overview of edge computing technology for unmanned cluster [J], Acta Ordnance Engineering, 44, 9, pp. 2546-2555, (2023)
  • [5] SUN B, LIU Y, WANG T, Et al., Review on optimization of federated learning efficiency in mobile edge networks, Journal of Computer Research and Development, 59, 7, pp. 1439-1469, (2022)
  • [6] LIU P K, LI B Q, QIN L, Et al., Application of deep learning object detection algorithm in insulator defect detection of overhead output circuits, High Voltage Technology, 49, 9, pp. 3584-3595, (2019)
  • [7] HUANG T, ZHANG Q, TANG X, Et al., A novel fault diagnosis method based on CNN and LSTM and its application in fault diagnosis for complex systems [J], Artificial Intelligence Review, 55, pp. 1289-1315, (2022)
  • [8] CHOUDHARY A, MISHRA R K, FATIMA S, Et al., Multi-input CNN based vibro-acoustic fusion for accurate fault diagnosis of induction motor, Engineering Applications of Artificial Intelligence, 120, (2023)
  • [9] SUN H, ZHAO Z, FU X, Et al., Limited feedback double directional massive MIMO channel estimation: From low-rank modeling to deep learning [C], 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp. 1-5, (2018)
  • [10] CAI G, LI J, LIU X, Et al., Learning and compressing: low-rank matrix factorization for deep neural network compression, Applied Sciences, 13, 4, (2023)
123