1
\$\begingroup\$

Original wording of question: What is fractional network access?

I'm having trouble figuring out what this term actually means, though I can find a ton of whitepapers about how some-standard-or-another means to support it.

I know it's related to the fact that wimax makes use of it's operating on subchannels, but 3g takes up an entire band. How would that work with 3g, then?

\$\endgroup\$

2 Answers 2

2
\$\begingroup\$

Another way of looking at it would be to say that conventional signalling methods have a small number of discrete channels which will never interfere with each other, while CDMA has a large number of virtual channels that will occasionally, randomly, collide. Conventional signalling methods will have low error rates but have a hard capacity limits. CDMA does not have a hard capacity limit, but as the number of logical channels in use increases, so will error rates. CDMA can't achieve as high bandwidth as other technologies, but unlike other technologies it allows point-to-point communication without arbitration.

\$\endgroup\$
3
  • \$\begingroup\$ If you suggest that there is difference in degree of orthogonality between M-sequencies and Sinusoids, then it is wrong. Both must be infinite to be perfectly narrow. As both are not perfect, then informational, probabilistic (whatever) interference is bad/good to the same degree. \$\endgroup\$
    – user924
    Jul 16, 2011 at 17:23
  • \$\begingroup\$ @Rocket Surgeon: If some number of non-coordinated transmitters try to send data using different CDMA patterns, there will be a certain probability of each piece of the transmission colliding, but the probability will be essentially independent for 1each bit. If each transmitter modulates using a fixed randomly-selected center frequency, the probability of a collision will be essentially the same as it would be for each piece of a CDMA transmission. The probability of a completely collision-free transmission will be much higher than with CDMA, but... \$\endgroup\$
    – supercat
    Jul 16, 2011 at 21:21
  • \$\begingroup\$ @Rocket Surgeon: ...so will be the probability of a completely garbled transmission. Of course, what matters with CDMA is the probability that a transmission will be received with few enough collisions that forward error correction can fill in any gaps. That probability will depend in significant measure upon how much forward error correction is used. I don't know to what extent existing CDMA channels dynamically vary their use of FEC, but in some sense, despite the fact that it requires more digital circuitry than simple modulation, CDMA's degradation is more like analog broadcasting. \$\endgroup\$
    – supercat
    Jul 16, 2011 at 21:27
2
\$\begingroup\$

Why CDMA is able to use whole band, when Wimax is subchanneled?

Because code division is based on decorrelation (orthogonality) of signals as opposed to traditional separation across frequency spectrum and/or time. So CDMA allows to operate concurrently, the same time, in the same frequency band, in the same geographical spot.

To understand how it works, look at M-Sequencies.

From high level look at spectrum, code division, walsh patterns etc. as different types of orthogonality. They all are equally limited in terms of informational capacity with presence of noises. But frequency spectrum is simplier technically, because it is a scalar, not some "code space" thing.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.