Thank you very much! The LMI is just a example in my problem(I don't consider its meaning). And i found another method to solve it in the paper " Vibration control for active seat suspension systems via dynamic output feedback with limited frequency characteristic ". I have another question. I found the results are different when I calculate the robust controller with "robust control toolbox" and 'yalmip toolbox', as follows,

I care about the gamma and K in a controller, and how can treat the different results with the same Rho when I use two kinds of toolbox? Thank you!