Realizing modulation using complex data
in [DSP]
|
From: san_jack on 8 Jul 2010 01:55 Hey all, We intend to purchase the PSK modulator IP core. The core outputs the complex data vectors (I,Q of 10/14 bit width). To realize the standard PSK output, what operation should we perform with these output data (like addition)?
From: Tim Wescott on 8 Jul 2010 02:04 On 07/07/2010 10:55 PM, san_jack wrote: > Hey all, > We intend to purchase the PSK modulator IP core. The core outputs the > complex data vectors (I,Q of 10/14 bit width). To realize the standard PSK > output, what operation should we perform with these output data (like > addition)? I almost don't want to tell you, because if you have to ask then you're almost guaranteed to run into difficulty that you'll have trouble getting out of without external help. Fortunately you know how to find this list, and there are a good few of us who consult on this sort of thing. If this is really PSK at baseband, then just multiply the I and Q outputs by sine waves at your desired carrier frequency and 90 degrees different in phase, then add the results -- that's your output. If you're really just modulating raw ones and zeros into PSK then the core is a rip-off -- if you know how to use it, you know how to make one, and the time and glue logic to jam some general-purpose thing into your design will be larger than that for the PSK core itself. OTOH, if it's super high-speed, or squeezes some sparkling performance out of a FIFO, or if it's doing some fancy encoding before it modulates, or somehow does something beyond just turning ones and zeros into a quadrature PSK signal, then it may be worth the $$. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at http://www.wescottdesign.com/actfes/actfes.html
From: Steve Pope on 8 Jul 2010 03:23 Tim Wescott <tim(a)seemywebsite.com> wrote: >On 07/07/2010 10:55 PM, san_jack wrote: >> We intend to purchase the PSK modulator IP core. >If you're really just modulating raw ones and zeros into PSK then the >core is a rip-off -- if you know how to use it, you know how to make >one, and the time and glue logic to jam some general-purpose thing into >your design will be larger than that for the PSK core itself. Yeah, hard to say what's going on here. Perhaps this core performs some more elaborate function. Googling on PSK modulator IP cores produces hits on several products, but they in many cases perform something slightly elaborate, like OQPSK modulation compatible with some standard. Who knows, it might use something conceptually difficult like a CORDIC algorithm. S.
From: san_jack on 9 Jul 2010 01:37 Hi Tim Wescott, Thanks for the reply. Output of the PSK is already multiplied with the carrier and the data sheet mentions that the output is in complex form (a+ib). Whether the complex data will be in time domain? so that both I,Q outputs can be added and given to DAC to realize the standard PSK output. >On 07/07/2010 10:55 PM, san_jack wrote: >> Hey all, >> We intend to purchase the PSK modulator IP core. The core outputs the >> complex data vectors (I,Q of 10/14 bit width). To realize the standard PSK >> output, what operation should we perform with these output data (like >> addition)? > >I almost don't want to tell you, because if you have to ask then you're >almost guaranteed to run into difficulty that you'll have trouble >getting out of without external help. Fortunately you know how to find >this list, and there are a good few of us who consult on this sort of thing. > >If this is really PSK at baseband, then just multiply the I and Q >outputs by sine waves at your desired carrier frequency and 90 degrees >different in phase, then add the results -- that's your output. > >If you're really just modulating raw ones and zeros into PSK then the >core is a rip-off -- if you know how to use it, you know how to make >one, and the time and glue logic to jam some general-purpose thing into >your design will be larger than that for the PSK core itself. > >OTOH, if it's super high-speed, or squeezes some sparkling performance >out of a FIFO, or if it's doing some fancy encoding before it modulates, >or somehow does something beyond just turning ones and zeros into a >quadrature PSK signal, then it may be worth the $$. > >-- > >Tim Wescott >Wescott Design Services >http://www.wescottdesign.com > >Do you need to implement control loops in software? >"Applied Control Theory for Embedded Systems" was written for you. >See details at http://www.wescottdesign.com/actfes/actfes.html >
|
Pages: 1 Prev: Spectrogram HD for iPad Next: rectangular pulse in matlab |