Article Details

Abstract

According to the monitoring requirements of the near-ground environment in the desert fringe area, a two-level data fusion algorithm based on multi-sensor collected data is designed. Firstly, the Kalman filtering algorithm is adopted to locally fuse the data collected by the same type of sensors, effectively suppressing noise and random errors, and improving the accuracy and stability of single-parameter measurement. Then, the BP neural network is utilized to globally fuse and nonlinearly model the multi-source heterogeneous environmental parameters, achieving the desertification level evaluation. Simulation experiments show that the Kalman filter can effectively reduce data fluctuations and measurement errors, while the neural network model performs well in the identification of non-desertification and severe desertification levels, with room for improvement in overall classification accuracy. This method enhances the reliability of the monitoring system through multi-sensor information fusion, providing an effective technical solution for dynamic assessment and early warning of desertification.

Keywords

environmental monitoring; Kalman filter; data Fusion; BP Neural Network; desertification level evaluation

References

1. [1] Liu Lei, Zhou Hu, Han Meilin, etc Multi sensor data fusion algorithm in ecological environment monitoring [J]. Journal of Hubei Institute of Technology, 2021, 37 (6): 21-25
2. [2] Liu Zhihao, Rong Lihong, Li Juan, etc A data fusion algorithm for pigsty environment based on wireless sensor network [J]. Journal of Qingdao Agricultural University (Natural Science Edition), 2025, 42 (1): 65-71, 78
3. [3] Zhang Longxiang, Feng Quanyuan Research on Gas Environment Monitoring Technology Based on Information Fusion [J]. Instrument Technology and Sensors, 2024 (7): 110-115
4. [4] Li Jianhong, Jing Wei, Liu Gongning Tunnel deformation prediction and safety evaluation based on monitoring data and RBF neural network [J]. Experimental Technology and Management, 2025, 42 (11): 154-160
5. [5] Liu Hui, Zhang Xiaoli, Huang Tianchen, etc Research on Coal Mine Underground Safety Monitoring Based on Improved RBF Data Fusion Algorithm [J]. Computer Measurement and Control, 2023, 31 (11): 173-180
6. [6] Wang Shuai Design of Water Vapor Detection System Based on Multi Sensor Fusion [D]. Nanjing: Nanjing University of Information Science and Technology, 2023
7. [7] Mi Chun, Li Siying, Mou Jiayi, etc Research on Collaborative Fusion Method of Urban Tunnel Environmental Monitoring Data Based on Multi Sensor [J]. Modern Tunnel Technology, 2023, 60 (5): 177-185
8. [8] Li Jinguang, Li Dan Design and Experiment of Data Fusion Algorithm for Soybean in Plant Factory Based on ZigBee Technology [J]. Changjiang Information and Communication, 2024, 37 (2): 50-52
9. [9] Li Wenchen Wireless Environmental Monitoring System Based on Multi Sensor Data Fusion [D]. Nanjing: Nanjing University of Science and Technology, 2014
10. [10] Li Kailin, Meng Fanpu, Ren Wanlong, etc A Water Quality Data Processing System Based on Neural Network Assisted Kalman Filter Algorithm [J]. Journal of Qilu University of Technology, 2023, 37 (5): 28-33
11. [11] Wang Yongfang Ecological risk assessment of desertification in Horqin Sandy Land based on multi-source data fusion and DPSIR model [D]. Changchun: Northeast Normal University, 2016
12. [12] Yan Yongfeng, Wu Xuan, Tian Cheng, etc Comprehensive Pipe Gallery Environmental Assessment Method Based on Information Fusion Technology [J]. Journal of Underground Space and Engineering, 2019, 15 (S2): 508-512

Funding

2023 Gansu Province University Innovation Fund Project (2023B-291); Special Research Project on Curriculum Textbooks for Primary, Secondary and Tertiary Schools in Gansu Province in 2024 (GSJC-Y2024181)

About the authors

Ma Lihong (1988.05-), female, Han ethnicity, from Qinhuangdao, Hebei Province, lecturer, master's student, research direction: intelligent control.