Designing a Fuzzy Gain Lyapunov Adaptive Filter Algorithm
Man Zhihong, Seng Kah Phooi, Hong Ren Wu · Studies in fuzziness and soft computing · 2002
A new approach in fuzzy adaptive filtering is presented in this paper. The proposed fuzzy filter is constructed from a set of fuzzy IF-THEN rules. Based on the observations of input signal and a collection of desired response, the filter parameters are updated by the Lyapunov sense fuzzy rules so that the error can asymptotically converge to zero. This scheme is the extension of the idea of Lyapunov theory-based adaptive filtering (LAF), thus it possesses the properties of the LAF and fuzzy logic. The stability is guaranteed by the Lyapunov theory. The design is independent of the signals’ stochastic properties. The effectiveness and robustness of the proposed filter are demonstrated in the simulation examples to support the theoretical analysis.