What is Multiplexing?

What is Multiplexing, and why should we care about it? Let’s find out.

Multiplexing and Demultiplexing

What is Multiplexing?

Imagine a group of girls singing in a chorus. What you’re hearing is the result of their combined efforts, where their song is ‘mixed’ in the air.

What if you want to hear each girl’s specific effort on its own? You can’t in the above example. What you need is each girl singing into her own microphone, and each audio channel is delivered to you separately.

The more the girls, the more the wires coming to your listening station. You begin to hope for a way to send all these signals through one cable, while still keeping the data separate (un-mixed).

That’s multiplexing or Muxing – the system by which many digital data streams are combined into one signal or file so they can be transmitted or delivered simply. Audio, video, and metadata is interleaved into one stream just for transport or delivery.

Not only does it reduce the amount of cabling, but it also reduces the usage of bandwidth. How’s that?

When you go out to get something from your local store, a family member or friend might say: “Dear, would you get me some ____ while you’re there?” The same happens with a delivery truck. If they are going to a particular area, they try to carry and deliver all the mail for that area in one go.

On the internet and airwaves, bandwidth is limited. Everyone wants his or her video to pass through very limited channels. Multiplexing tries to keep everyone happy.

You see multiplexing everywhere – telephone, digital television, the internet, and even video file systems. The device or software that performs the multiplexing is called a Multiplexer (MUX).

What is Demultiplexing?

The process of splitting the signals to its individual components is called Demultiplexing or Demuxing. It is the exact opposite of multiplexing.

The device or software that performs the demultiplexing is called a Demultiplexer (DEMUX).

Time-Division Multiplexing

There are many ways to combine signals. The most common method in digital video is called Time-Division Multiplexing. It sounds scary, but it’s pretty simple:

Time-division multiplexing
Image Courtesy: Tony R. Kuphaldt, Socratic Electronics website

If the above image confuses you, forget it. Imagine sprinters running on their own ‘channel’ (lane) of the track. Each sprinter has a camera just for herself.

If you want to watch each sprinter on your monitor for a second, and there are 10 sprinters, you use time-division multiplexing by making your monitor ‘switch’ every second.

Now, what if these 10 sprinters were forced to run on 5 lanes (channels) instead of 10? You’ll need to ‘switch’ faster, at half a second. A person sitting at home with a remote will believe all the channels are available at the same time. You can channel surf as fast as you like, the system is ready for it!

Now, this analogy isn’t completely accurate, but what you should take away is

  • Of the total time (bandwidth) available, each channel gets an equal share
  • You get the impression that every channel is available simultaneously, when in fact, they are just switching so fast you don’t notice they are switching.
  • At any given moment, only one signal is passing through the door (bandwidth).

 
In the case of modern technology, we go one step further, with Statistical time-division multiplexing. Simply put, it ‘studies’ the system to make the process even more efficient so you can squeeze more out of your limited bandwidth. It’s that precious.

All this isn’t as simple as I make it sound. Imagine the complexity – you need to interleave audio, video, metadata and subtitles, for both HDTV and SDTV, with varying bit rates, for hundreds of channels, over the same bandwidth.

It’s one the backbones of broadcasting.