Dear Taranjitsingh,

concerning the question is it a high-pass filter. I think so and to refine the question comprehensively here are links and attached files.

- Control of a Spring-Mass-Damper System Using Mixed mu-Synthesis ...

www.mathworks.com › ... › Robust Controller Design

- Mixed-Sensitivity Loop Shaping - MATLAB & Simulink - MathWorks

www.mathworks.com › ... › Loop-Shaping Synthesis

- H∞ mixed-sensitivity synthesis method for robust control loopshaping ...

www.mathworks.com › ... › Loop-Shaping Synthesis

Best regards

Hi Taranjit,

Constant sensitivity at 0dB is not a good (robust) design. Since sensitivity reflects modeling uncertainty, a gain as large as 0dB corresponds to large uncertainty. It should have some high-pass filter behavior.

Your closed loop sensitivity depends on how you choose your weighting transfer functions. You should choose a LPF like weighting transfer function to ensure sensitivity behaves like a HPF.

Dear Dr. Lafifi, 

Thank you very much for answer. The links are helpful. However, I have designed the filters following the same approach. I have still a problem.

Dear  Mr. Baranwal, 

Thank you very much for your reply. I have followed the same approach to design the weighting filters. I chose a LPF to ensure bound on the sensitivity. No matter how I allow the filter to be (being relaxed or harsh), it still gives the closed loop sensitivity constant at 0db. At first, I though it would be a wrong notion, but the I used the function 'loopsens' which provides the same result. 

So, my question is, what are the scenarios that the sensitivity is constant at 0db. Is it the fact that I am linearizing the non-linear plant and it doesn't consider the valuable informations ? 

Regards, 

Taranjit