An optimal filtering algorithm for systems with multiplicative/additive noises

作者：Zhang Ling^*; Zhang Xian Da

来源：IEEE Signal Processing Letters, 2007, 14(7): 469-472.

DOI：10.1109/LSP.2006.891331

• innovation process
• multiplicative noise
• nonlinear measurement model
• optimal filtering
• projection theorem

In this letter, we propose a new optimal filtering algorithm for systems with multiplicative noise process under correlated additive noises. Different from the traditional derivation method of constructing innovation process, we derive the new algorithm based on the well-known projection theorem, which is simple in principle, convenient for recursive estimation and flexible for application. The new recursive filter is superior to the conventional Kalman filter in flexibility and applicability.

• 出版日期2007-7
• 单位清华大学