Bayesian Approach for Multi-Sensor Data Fusion Based on Compressed Sensing for Wireless Structural Damage Signal

Sai Ji,Zhen-Yu Chen,Ping Guo,Ya-Jie Sun,Jian Shen,Jin Wang,Chin-Feng Lai
DOI: https://doi.org/10.6138/jit.2016.17.7.20160415a
2016-01-01
Abstract:To meet the requirements of data compression and data fusion in structural health monitoring (SHM) in wireless sensor networks (WSNs), this paper proposes a novel method of multi-sensor data fusion based on compressed sensing (CS) for wireless structural damage signal, which can realize data fusion and reconstruct sparse signals. First, the damage signals of aviation aluminum plate are measured onto the linear measurement data through inner products with random Gaussian matrix. Next, data fusion of measurement data is realized by the Bayesian algorithm. Finally, the damage signals can be reconstructed by the CS method. The experiment results show that, compared with the existing methods, the proposed approach can save the network bandwidth and energy according to a good data fusion performance, anti-noise property and a better data compression effect. The proposed approach can also realize the damage identification accurately on the aviation aluminum plate and keep the detection error within 0.82 mm.
What problem does this paper attempt to address?