An Information Fusion Identification Method for Multisensor Autoregressive Moving Average Signals with White Measurement Noise and Sensor Bias

作者：Liu Jinfang; Deng Zili^*

来源：Sensor Letters, 2011, 9(4): 1443-1447.

DOI：10.1166/sl.2011.1654

• Multisensor ARMA Signal
• Sensor Bias
• Information Fusion
• Identification

For the multisensor autoregressive moving average (ARMA) signals with white common disturbance noise, white measurement noise and sensor bias, a multi-stage information fusion identification method is presented, when all the model parameters, noise variances and sensor bias are unknown. In the first stage, the estimators of autoregressive (AR) parameters and sensor bias are obtained based on the recursive extended least squares (RELS) algorithm. In the second stage, the estimators of common disturbance noise variance and measurement noise variances are obtained based on the correlation method. Then using Gevers-Wouters algorithm with dead band, the estimators of moving average (MA) parameters and process noise variance are obtained in the third stage. Both the local and fused estimators have strong consistency, where the fused estimators are obtained by taking the average of local estimators, so they have higher credibility. A simulation example shows the effectiveness of this method.

• 出版日期2011-8
• 单位黑龙江大学