Maximum entropy spectrum estimation with imprecise autocorrelation function lags by fuzzy set theory

The author provides a novel method and a viewpoint based on fuzzy set theory which lead to the algorithm by solving two sets of linear inequalities by linear programming. The algorithm based on the method converges in finite steps and to the optimal solution if some regular conditions are satisfied. Simulation results show that considerable improvement has been obtained for the frequency resolution. The method can be extended to the multidimensional case and to related fields directly.<
>