Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Question about noise in complex(polyphase) filter

Status
Not open for further replies.

moh_hafez

Newbie level 1
Newbie level 1
Joined
Mar 10, 2009
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,291
complex polyphase filter

I have made a 3rd order complex filter , when performing noise analysis using cadence , i put noise sources at both the inputs of I and Q channel , and saw the output noise at the I or Q channel , it was 45nV/sqrt(hz) . when i measured the input referred noise at the I or Q channel , it was about 100nV/sqrt(hz) although the gain is 0dB .
So does this may be bec. of Cadence when calculating Noise , as it doesn't see the phase difference bet. the noise sources of the I and Q channel (as it doesnt care for the data in ac sources when performing noise analysis) and hence it doesnt see the 0dB gain which depends on the phase shift bet. I and Q inputs .
So is the solution that i get the output noise using cadence and then calculate the input referred noise based on my understanding that the input will be the same as the output .
 

Hello hafez,
I am designing the polyphase filter and am stuck at the exactly same problem as yours.
Cadence recommends following way to do the noise simulation in Complex filters: -

Problem
You are trying to do a Spectre noise simulation of a circuit with multiple inputs and/or outputs.
However, the noise setup has the following restriction: only one port/source and only one output node/port can be specified.Is it possible to use SpectreRF in this case? If yes, how?

Solution
The answer is - yes: you can solve your problem using "principle of superposition".
For example, you have a complex filter with two inputs Iin, Qin and two outputs Iout and Qout in your LowIF receiver:
According to principle of superposition you can calculate the output spectra as follows.
1. Set ac-amplitudes of both input ports: portI = 1, portQ = 0.
2. Using AC analysis calculate the I- and Q - output voltages and save as the transfer functions HII(f) and HIQ(f) respectively.
3. Set ac-amplitudes of both input ports: portI = 0, portQ = 1.
4. Using AC analysis calculate both I- and Q - output voltages and save as the transfer functions HQI(f) and HQQ(f) respectively.
5. Calculate both I- and Q- output spectra:
Iout(f) = HII(f) * Iin(f) + HQI(f) * Qin(f);
Qout(f) = HQQ(f) * Qin(f) + HIQ(f) * Iin(f).

After step 2 you can calculate output noise (V**2/Hz) when only portI is active:
- at Iout: Ioutn_I(f);
- at Qout: Qoutn_I(f).

Note: Spectre calculates 'SSBnoise' = 2 * 'DSB noise'.

After step 3 you can calculate output noise (V**2/Hz) when only portQ is active:

- at Iout: Ioutn_Q(f);
- at Qout: Qoutn_Qf).

Now you can calculate the total otput noise (V**2/Hz) at both outputs:
Ioutn(f) = Ioutn_I(f) + Ioutn_Q(f);
Qoutn(f) = Qoutn_I(f) + Qoutn_Q(f);
______________________________________________________________________________

Now even after doing all this I see that input referred noise is double that of output referred noise. I don't know why this is happening. Please let me know if you have got the answer.
Thanks
Megh
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top