How is the modulation technique used?

Is raised-cosine filter used in baseband modulation in real systems?

• I learned in class that when doing baseband modulation, a sinc function or a raised cosine function is used to modulate the digital signals into baseband waveforms. Is this true in the design of systems in real life or is it just textbook and we have something completely different? It would be nice to include one or two examples. Thanks in advance.

• Answer:

  Raised cosine filters are used in real systems. The digital bits going into a modulator are "pre distorted" by a root raised cosine filter. The purpose is to reduced intersymbol interference. Intersymbol interference occurs when the higher harmonics are smeared with respect to the fundamental due to group delay. Most transmission media is frequency dependent. Different frequencies take longer or shorter periods of time to transit the media. This effect is called "http://en.wikipedia.org/wiki/Group_delay_and_phase_delay" see image below. Perfect square pulses that transit the frequency dependent media come out all smeared because the phase relationship between the different harmonics is not preserved. A cosine filter reduces/removes the higher harmonics and enables the "zero crossing" of each pulse to be preserved. This is best illustrated by the "eye diagram" of digital pulses on oscilloscopes. As shown by the oscilloscope images above, the digital signal on the left has a narrow zero crossing and the eye is relatively wide, meaning the noise is low and the bit error rate for this digital bit stream would be low. The eye diagram on the right has an eye that is closing, indicating higher levels of noise, and worse, the zero crossing is smeared by group delay. This bit stream will have higher bit error rates due to the increase noise and uncertainty associated with sample timing.