background image
168 Chapter 6: Using ISDN and DDR Technologies
Once the data world was born, it became necessary to also transport data over long distances.
Because the digital facilities were already available, it seemed a natural extension to use those
facilities. However, it proved to be something of a painful experience for a time. The history of
voice and data transmission over T1 and E1 facilities followed the same growing pains.
T1 Framing
The first part of this section focuses on T1 framing and line coding. E1 framing and line coding
are discussed afterward.
T1 specifies the physical coding of the signal on the wire, and DS1 specifies the framing of
characteristics. So, T1 and DS1 are not the same thing after all, even though the two have been
used interchangeably for years and it's certain not to stop because of this book. But, now you
know.
A digital signal that is level 1 (DS-1) consists of 24 DS-0s. A DS-0 is a 64-kbps channel. This
channel is known as a timeslot. One DS0 represents one voice call. The timeslot is derived from
the Nyquist theorem. Nyquist said that f
s
= 2(BW). Because the voice world had decided that
0­4000 Hz would be the supported range for voice circuits, the number 4000 was plugged into
the formula in the BW position. Therefore, f
s
= 2(4000) = 8000, which is the number of samples
we should take of this analog wave per second.
To properly digitize (quantize) analog voice, it is necessary to take samples of the voice wave
over time. Sample too fast and you waste resources. Sample too slowly and you allow for
aliasing. Aliasing is a condition that occurs when two or more analog waves can match the
coordinate points set forth by the samples (a little beyond our scope).
Each of the 8000 samples per second is represented by an eight-bit code word. Without going
into too much depth, this code word simply defines the coordinates of the sample (polarity = 1
bit, segment = 3 bits, and step = 4 bits). Figure 6-13 depicts the sampling of the wave and the
resulting code word.
The resulting throughput of 8000 eight-bit samples per second is 64,000 bps, or one DS-0.
When one sample has been taken from each of the 24 timeslots, a T1 frame is created. Because
this is time division multiplexing (TDM), 8000 eight-bit samples are taken from each timeslot
every second. The result is 8000 T1 frames per second.
The telco provider specifies the type of framing that you should use when connecting to their
facilities. The choices with T1 are SuperFrame (SF, also known as D4 framing) and Extended
SuperFrame (ESF).
SF is the assembly of 12 T1 frames. Each of the T1 frames is separated from adjacent frames
by a single framing bit (8000 T1 frames per second, each with an additional bit that is an
additional 8000 bps of overhead).