The filtering based parameter identification for bilinear-in-parameter systems |
| |
Authors: | Xuehai Wang Feng Ding |
| |
Institution: | 1. School of Mathematics and Statistics, Xinyang Normal University, Xinyang 464000, PR China;2. College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, PR China;3. School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, PR China |
| |
Abstract: | This paper discusses the parameter estimation for a class of bilinear-in-parameter systems with colored noise. By utilizing the filtering technique, we derive the relationship between the filtered output and the measurement output and obtain two linear regressive sub-models. A filtering based multi-innovation stochastic gradient algorithm is derived for interactively identifying each sub-model. The proposed algorithm avoids the estimation of correlated noise and improves the parameter estimation accuracy by making full use of the measurement data. The numerical simulation results indicate that the proposed algorithm has higher estimation accuracy than the hierarchical multi-innovation stochastic gradient algorithm. |
| |
Keywords: | Corresponding author |
本文献已被 ScienceDirect 等数据库收录! |
|