Combining Genetic Algorithm and Simulated Annealing to Design H_2/H_∞ Deconvolution Filter with Missing Observations

摘要

In this paper, we combine GA (genetic algorithm) and SA (simulated annealing) to approach H_2/H_∞ deconvolution filter with missing observations. The missing observations model is based on a probabilistic structure. The probability of occurrence of missing data is unknown prior. The aim of H_2/H_∞rncriterion is to achieve the H_2 optimal reconstruction and subject to the H_∞ normrnconstraint on the transfer function from the channel input to the filter error. In this situation, the design deconvolution filter becomes a complicated nonlinear estimation problem. In this paper, we combine the selected features form GA and SA to achieve weak dependence on initial parameters and fast convergence to treat the signal reconstruction problem with missing observations. Finally, a numerical example is presented to illustrate the design procedure and confirm the robustness performance of the proposed method.