In 'vecfitcpu_vortex_simfits.m', I don't change any setting but set ‘dz = -300’ ( or dz = 300). The estimator's performance is not as good as the one in the article Supplementary Information Figure 1(j). The parameter estimation results have many outliers and incorrect results so that the variable ‘thetastd’ is far greater than CRLB. How to make the estimator's performance convert to CRLB when ‘dz’ is >200nm or <200nm, which has been showed in the article Supplementary Information Figure 1(j).
In 'vecfitcpu_vortex_simfits.m', I don't change any setting but set ‘dz = -300’ ( or dz = 300). The estimator's performance is not as good as the one in the article Supplementary Information Figure 1(j). The parameter estimation results have many outliers and incorrect results so that the variable ‘thetastd’ is far greater than CRLB. How to make the estimator's performance convert to CRLB when ‘dz’ is >200nm or <200nm, which has been showed in the article Supplementary Information Figure 1(j).