Used to StackOverflow, this is my first question here. So please bear with me if the question is too simple and obvious to you as I am not good in Electrical Engineering stuff. I really need to understand this as part of a Computer Networking course:

I read that a composite signal can be thought of as a combination of simple sine waves and multiplexing is a method by which multiple analog message signals or digital data streams are combined into one signal over a shared medium. Now here are my confusions arising from this:

a) Can multiple composite signals be multiplexed, transmitted and then demultiplexed into the original composite signals? Or only non-composite, single-frequency signals be multiplexed and then demultiplexed into original non-composite signals? If the former is true, does demultiplexing entail separating the constituent frequencies of individual composite signals and then combining them to get the original composite signals? Is that a complex procedure?

b) Can a multiplexed signal be called a composite signal, more so if it is composed of separate composite signals? Can it certainly be considered a composite signal if it is composed of non-composite, single-frequency and separate signals?

c) Does data communication mostly involve composite signals (with many constituent frequencies), or are non-composite, single-frequency signals are also used in communication?

Your answers will be helpful.

  • \$\begingroup\$ Yes. The output of an FM detector is described as a composite signal, and it contains (a) many frequencies and (b) multiplexed stereo information. \$\endgroup\$
    – user207421
    Feb 1, 2015 at 0:19

3 Answers 3


It seems like your question is about data transmission, so I'll try to answer it in this context.

You can't transmit information on a single frequency, because technically "single frequency" implies that the sender is turned on forever. As soon as it is switched on/off (corner case of AM) additional frequencies are added. For example a 1 MHz carrier switched on/off (this is called on/off keying or OOK) at 1kHz causes two sharp spikes at 0.999 MHz and 1.001 MHz in the spectrum. The difference between the highest and the lowest frequency is usually called the bandwidth of your signal and in the case of OOK it is twice the modulation frequency. The max. data rate would be 2 kBit/s, because the highest frequency sequence that is possible is 1010101....

If you want to composite such transmission systems onto a shared medium (cable, air) you need to make sure that they don't overlap. in the above example you could place additional transmitters at .998 MHz, 1.002 MHz etc. A practical example is FM radio, the stations are separated by a couple of 100 kHz for this reason.

Multiplexing is about combining N serial bit streams into one with a N-times higher bitrate. Lets say that you want to multiplex 10 2kBit/s (example above) bit streams into one with 20kBit/s. The bandwidth of the signal is than 20kHz and the transmitter will occupy the same amount of spectrum on your shared medium as ten individual transmitters.

At this point, I recommend you to learn more about Fourier Transformation and Shannon Coding Theorem, it will help you to get the connection between time and frequency domain in a data transmission system.


You've asked 8 questions, of which I'll answer one: "Is a multiplexed signal same as a composite signal?:"

The answer is, it depends.

There are many multiplexing methods, of which one is "frequency division multiplexing", where the available bandwidth of a carrier is split into discrete channels by filters.

The information to be sent is shifted in frequency and band-limited to fit into a particular channel at the same time that all the other signals are being subjected to the same process and sent, each within its discrete channel, to a receiver, via the carrier.

At the receiving end, filters corresponding to the sent channels separate each channel's signal from that of all the others, and then shift the signal in frequency in order to restore it to "baseband" which will be a replica of the original signal.

In this instance, the information on each input can be - but doesn't have to be - composite, while the carrier itself will definitely be composite.

Other multiplexing schemes will have different rules.


The act of multiplexing two signals (in time) is one of sampling signal A then sampling signal B. The multiplexed signal is then a combo of A and B.

Demultiplexing is the act of separating A and B by sampling methods.

Providing you sample at a fast enough rate (see nyquist) you don't lose signal information.

Whether the signals are single channels or complex combinations of multiple channels or digital signals matters not one bit.

What you choose to call a signal is up to the individual. The complexity of the method is, again, one of semantics and not really answerable.

  • \$\begingroup\$ Does "single channel" signal mean a single, one-frequency wave, and a multiple-channel signal mean a composite wave? Further, I know that a (EM) wave has a single frequency, so can you give an idea in simple terms how a many of such waves combine to form a composite wave? What is the purpose of creating and using composite waves? \$\endgroup\$
    – Meathead
    Jan 31, 2015 at 12:06
  • \$\begingroup\$ Also, assuming a composite signal is composed of many frequencies/waves, how are the defining attributes of a composite signal derived from the constituent waves?What are the defining attributes of a composite wave as such? \$\endgroup\$
    – Meathead
    Jan 31, 2015 at 12:09
  • \$\begingroup\$ A useful signal comprises of more than one frequency. Consider a square wave - it comprises (theoretically) of an infinite number of individual frequencies. An audio signal comprises of a range of frequencies that can all occur simultaneously between 20Hz and 20kHz. An EM wave would be fairly useless if it only had one frequency because it would not be modulatable because to do so would produce an EM wave with a multitude of frequencies. \$\endgroup\$
    – Andy aka
    Jan 31, 2015 at 12:22
  • \$\begingroup\$ Thank you. It was very helpful. I am trying to digest it completely though :-) \$\endgroup\$
    – Meathead
    Jan 31, 2015 at 19:33

Your Answer

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

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