Receiver Design problem.

Dear all,
Now, I am doing the UWB receiver design. I just design in the first frequency 3-band from 3GHz - 5GHz. I have combined the Rx/Tx switch + LNA + Down mix-1 + Down mix-2. I have a problem in NF problem. For the first and third frequency band, the receiver could give me ~ 4.5dB NF value. However, the second frequency band gives me ~ 6.5dB NF value. My specification need to be 4.5dB around. I still don't know why it is happy. My receiver deisgn is zero-IF. Does any RF design expert give me some suggestions to fix it or find the problem out?

Thanks

wccheng

would u please send the block digram of ur receiver , and the spec's of each block , so we can check the system spec's ,

but it seems that ur LNA , have higher NF in the second band

khouly

Most likely your Tx/Rx switch insertion loss increases in Rx mode at higher frequency or the LNA noise optimum is matched to lower band. 4.5 dB NF requirement does not sound so tough yet for 5 GHz design. You must use such process in selected semiconductor that provide low noise. Like GaAs PHEMT. Check LNAs from Avago or RFMD.

Check the insertion loss of your Tx/Rx switch over the whole band with network analyzer and measure your LNA NF with noise figure meter over the band. This should clarify where the NF degradation comes.

Karote

However, my NF is so strange. The first and third band is okay (~4.5dB). The middle band of frequency is higher (~6.5dB)

It is hard to tell as there is not enough info about your circuit of LNA (broadband vs narrowband with frequency switching) and mixer. One thing you can do is to isolate LNA and Mixer, see which one dominates the NF.

I think there may be a dip in the Insertion loss charecterstics of your Tx/Rx switch or in the NF charecterstic of LNA.
Measure both of the above parameters individually, so that you can try to match or to replace that particular device.