Reducing S11 in lumped ports

Hello,
I am trying to simulate the flow of energy through a dielectric slab excited by a lumped port. I use 2 conductors (copper sheets 0f dimension 2 cm x 2 cm x .02 cm) and a rectangle to connect them, forming a lumped port. The dimension of dielectric is 10x10x80 cm. When I simulate it in the frequency range of 100KHz to 50 MHz, the distance propagated by the signal is increasing with frequency. Can you please tell me where I am going wrong?

Also, S11 is either 1 or 0.99998. Can you tell me a way to improve this? Would appreciate any suggestions.

Thanks,
Meenu

?
What's wrong with this?

Hi FvM, Thanks for responding. Actually the attenuation is calculated as the product of angular frequency and sq rt of permeability and permittivity. So isn't the attenuation suppose to be high for higher frequencies? But I am experiencing an opposite effect. I am unable to justify it. Please help.

Thanks again