Multiple SHM signals separation based on IUKF in wireless sensor network

作者：Zheng, Hengbin^*; Yan, Quansheng; Wang, Weifeng

来源：Journal of Information and Computational Science, 2011, 8(14): 2991-3001.

• Instantaneous mixtures
• Iterated unscented Kalman filter
• Non-linear model
• Non-stationary sources
• Nonstationary
• Separation problems
• Sigma point
• Structural health
• Wireless sensor

An iterated unscented Kalman filter (IUKF) based method is presented for multiple Structural Health Monitoring (SHM) signals separation problem in wireless sensor network (WSN). The IUKF-based method is easy to handle non-stationary sources in nonlinear models. By conducting experiments on instantaneous mixtures of several real SHM signals, the proposed method is compared with a traditional non-stationary blind source separation (BSS) algorithm. The simulation results show that our method has better performance. Considering the convenience to transmit sigma points between sensor nodes, the IUKF-based one can cast to multiple SHM signals collaborative processing in distributed WSN.

• 出版日期2011-12
• 单位华南理工大学