Help!Help!Help! For LNA design

I have designed LNA, that contains three stages and bandpass filter. First stage is used ATF-36077 PHEMT GaAsFET. Following second stage with Minicircuits ERA-3 MMIC. Microtrip bandpass filter take out amplified outband signals and thrid stage, again with ERA-3, amplify final signal.
For PCB, I have used ROGERS 5880 material (Er=2.2; thickness = 1.575mm ).
My amplifier desire parameter as following:
Frenqency range: (2.65 -:- 3.15)GHz
Gain: (30 -:- 32)dB
Noise figure: <2dB
Supply voltage: +5V; -5V
Connectors type: N

I have calculated for LNA by hand. Therefore, my LNA is without accuracy and mis-matching. I achieved is about 10dB (too low).
Anybody has software to design LNA. Please! You design and calculate LNA in softwave. I wonder if you can to give me your LNA layout, so I make reference to it.
Thank you for your help!!!!!
I am looking forward your reply.

Hi,
I have doen an LNA for 1.5GHz so may not be useful for you. But check out the agiltent website for APpcad and also a matching game simulator. These shall defintely help you.
BRM

I am a freshman in rf design!
but i have design a lna in 4ghz with atf36077, and the tool is serenade.
the nf is 0.8,
but it still had some problems

Are you sure in microstrip filter characteristic? Remove this filter and check gain again.

Anybody has done it using CMOS process using SpectreRf... please let me know

T Zul

I suggest you to check with RF probe at different interval. As you have cascaded them, you need to find out which component cause problem before you make any changes...

for cascaded amplifiers, the inter-stage match networks must be designed cafully, otherwise, the performance will degrade.

What are:
Noise Figure + Gain + Output IP3 for each part of your LNA?

What are also for the system:
Temperature + Input Power + Noise Bandwidth + S/N sensitivity?

If calculation for output is not what you expected then must do some magic like matching if possible or you must choose something else.