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.

Register Log in

REG: Spectrum Analyzer for determining the received envelope R

Status
Not open for further replies.
D

david87

Junior Member level 1
Joined
Apr 26, 2012
Messages
15
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,394
Hi,
I have to measure the fade depth from the received signal obtained from spectrum analyzer. The data set obtained from the spectrum analyzer is said to be a received envelope signal? I'm keeping it in zero span and measuring the signal. Please clarify my doubt.

Thank you,

Regards,
David.
 

I'm not sure what your doubt is, but normally a spectrum analyzer displays amplitude vs frequency. In zero span mode it displays amplitude vs time (more like an oscilloscope).
 

pstuckey said:
I'm not sure what your doubt is, but normally a spectrum analyzer displays amplitude vs frequency. In zero span mode it displays amplitude vs time (more like an oscilloscope).
Click to expand...

ya, i know in zero span mode, it is more like oscilloscope. I mean after taking the data set from spectrum analyzer and plotting the amplitude Vs time. Is the plot is said to be received Envelope(R) which is used to calculate the fade depth (R/Rrms)? or after taking data set should we have to do some procedure to get the received envelope?.
 

I see. What are the units of your amplitude? Normally when I think of envelope it is voltage vs time. The default for your spectrum analyzer is probably power (dBm) vs time.
 

pstuckey said:
I see. What are the units of your amplitude? Normally when I think of envelope it is voltage vs time. The default for your spectrum analyzer is probably power (dBm) vs time.
Click to expand...

ya it is in dbm. i have converted to linear unit. So that is the received envelope( voltage Vs time)??
 

Maybe. Please clarify exactly what linear unit? For instance dBm is a log unit, and Watts is linear, but they both are a measure of power.
 

pstuckey said:
Maybe. Please clarify exactly what linear unit? For instance dBm is a log unit, and Watts is linear, but they both are a measure of power.
Click to expand...

Please see the settings below.

"Mode;ANALYZER;
Center Freq;2400000000.000000;Hz
Freq Offset;0.000000;Hz
Span;0.000000;Hz
x-Axis;LIN;
Start;2400000000.000000;Hz
Stop;2400000000.000000;Hz
Ref Level;0.000000;dBm
Level Offset;0.000000;dB
Ref Position;100.000000;%
y-Axis;LOG;
Level Range;100.000000;dB
Rf Att;25.000000;dB
RBW;10000000.000000;Hz
VBW;1000000.000000;Hz
SWT;0.500000;s
Trace Mode;AVERAGE;
Detector;SAMPLE;
Sweep Count;0;
Trace 1:;;
x-Unit;s;
y-Unit;dBm;
Values;1251;"
 

Your y-axis (amplitude) is power, dBm.

If the signal is a simple tone, converting from power to voltage is straightforward.

Vpeak-peak = 2*sqrt(Prms*50 ohm*2)

If your input is a modulated signal then the conversion is more complicated. What is your input signal and how accurately do you need to know the envelope?
 

Ok ok. I'm transmitting the signal from signal generator and FREQ is 2.4GHz and level is 0dbm. I need to analyze the fade depth of the received signal envelope and need to determine level crossing rate from received signal envelope.
 

In zero span mode, the display showed total power in the selected IF bandwidth (Resolution Bandwidth), which is 10MHz according to the settings you posted. So, the display shows the total integrated power over 10MHz vs. Time. If you leave the units as dBm, then simply subtract the smallest signal magnitude from the largest, and it will give you the fade depth in dB. That's the beauty of logarithmic units - turns a division problem into subtraction...
 

w2aew said:
In zero span mode, the display showed total power in the selected IF bandwidth (Resolution Bandwidth), which is 10MHz according to the settings you posted. So, the display shows the total integrated power over 10MHz vs. Time. If you leave the units as dBm, then simply subtract the smallest signal magnitude from the largest, and it will give you the fade depth in dB. That's the beauty of logarithmic units - turns a division problem into subtraction...
Click to expand...

OK OK...But the data set which we obtain from spectrum analyzer as said before should be converted to linear values and it should be normalized to its RMS value and it can be converted to db. After that as you said the largest signal magnitude is subtracted with smallest signal mag, Which will give fade depth, right?.
 

david87 said:
OK OK...But the data set which we obtain from spectrum analyzer as said before should be converted to linear values and it should be normalized to its RMS value and it can be converted to db. After that as you said the largest signal magnitude is subtracted with smallest signal mag, Which will give fade depth, right?.
Click to expand...

In my mind, if you have the analyzer set to use the Average detector, and units in dBm, then I don't see the need to convert this to linear values and then back to dB again.
 

I agree with w2aew, for fade depth measurements just use the spectrum analyzer output in dBm, no conversion necessary.

If you also need to compute the average power over some period of time, then you do need to convert power to linear units (Watts), do the average using the linear units, then convert back to dBm.
 

w2aew said:
In zero span mode, the display showed total power in the selected IF bandwidth (Resolution Bandwidth), which is 10MHz according to the settings you posted. So, the display shows the total integrated power over 10MHz vs. Time. If you leave the units as dBm, then simply subtract the smallest signal magnitude from the largest, and it will give you the fade depth in dB. That's the beauty of logarithmic units - turns a division problem into subtraction...
Click to expand...

OK OK. I have attached a doc and in the table.1 they have calculated fading depth. Please see and tell whether they have calculated as you said. Please find doc attached.
 

Attachments

  • fulltext.pdf
    126.6 KB · Views: 92
I've looked through your paper. He definitely is not calculating the fade depth by max-min, or even average-min. He might be basing it of a probability of link availability (although then it is usually called fade margin). Unfortunately he doesn't define how he calculates it.
 

But the real radio channel is probability.
So the fade is also probability. what do you need?
 

I need to find fade depth, level crossing rate and average fade duration from the received envelope. Can you please tell me the procedure to find it from spectrum analyzer, it would be helpful for me.
 

Since the fade is varied by time, and Spectrum analysis only samples the signal in a very short time period for a fixed frequency, it maybe not a good idea to measrue fade by SA.
 

tony_lth said:
Since the fade is varied by time, and Spectrum analysis only samples the signal in a very short time period for a fixed frequency, it maybe not a good idea to measrue fade by SA.
Click to expand...

OK..But i have only spectrum analyzer. So after getting the received signal, is there any way to determine the fade?? In most of the paper they have used spectrum analyzer, but not described clearly about the measurements.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top