Is it the bandwidth ( BW ) of the individual pulse or should I generate a random process and calculate the BW from its PSD.
In my class we considered it the bandwidth of the raised cosine when we were studying the ISI. but then when we came to the modulation techniques we calculated the PSD of the random process ( and we considered that each bit is a rect not a raised cosine ).
What is the difference between the two approaches and which one is the transmission bandwidth?
\$\begingroup\$
\$\endgroup\$
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
5
If the transmitter uses pulse shaping, then it is definitely the PSD of the pulse (raised cosine or other), not a hypothetical rectangle.
-
\$\begingroup\$ Then what does the PSD of the random process indicate? \$\endgroup\$ Commented May 17, 2016 at 19:39
-
\$\begingroup\$ I am not asking about whether to use a rect or a raised cosine I am asking about whether to use the BW of the pulse - whatever its shape is - or to generate a random process and calculate its PSD \$\endgroup\$ Commented May 17, 2016 at 19:42
-
\$\begingroup\$ It would be indicative of information content. Generally pulse shaping is conducted on discrete signals. I suppose if the random process were analog, then you would have to convolve the two PSDs. \$\endgroup\$– MikePCommented May 17, 2016 at 19:42
-
\$\begingroup\$ you mean to convolve the PSD of the random process ( assuming rectangles ) with the FT of the raised cosine ? \$\endgroup\$ Commented May 17, 2016 at 19:50
-
\$\begingroup\$ Actually just multiply the fourier transforms (PSD is FT squared). Convolution would be the time domain. So the FT of the rectangle is a sync and the FT of the raised cosine is discussed here: math.stackexchange.com/questions/1419983/… Sorry for the confusion!!! \$\endgroup\$– MikePCommented May 18, 2016 at 14:18