Prev: STUPIDENT::Re: Measuring SNR using output of ADC
Next: c2c vs r2c fft
From: commengr on 8 Nov 2009 13:02
I'm simulating DS-CDMA using Matlab and I have some observations and have
thought of a method to simulate it. I am caught by an issue, which I aim to
resolve with your help


For CDMA, is following "datagen-spread-modulate" procedure equivalent to
"datagen-modulate-spread"?

I am using 'multiplication over [1,-1] to solve XOR over [0,1]' used for
spreading and once I do 'datagen-spread-modulate', the data after spreading
is [1,-1] and such input is not compatible with Matlab functions for PSK
modulation.

Moreover, even when I convert the spread data from [1,-1], I still have to
divide it by sqrt(SF) to normalize the energy. Now, once the spread data is
reverse mapped to [0,1] so that I can use Matlab function pskmod.m to
modulate, I think the whole thing of normalized energy will be nullified

I can divide the QPSK (suppose QPSK modulated is used), by sqrt(SF) but
wouldn't that alter the orientation of QPSK symbols?


Any suggestions? Should I do "datagen-modulate-spread", but wouldn't have
similar affects?


Thanks to all who help
 | 
Pages: 1
Prev: STUPIDENT::Re: Measuring SNR using output of ADC
Next: c2c vs r2c fft