T1 Line Service

How Is A T1 Line Signal Created?

Telephone networks have seen several innovations and changes over the last century or so. While most of the initial progress was made to allow multiple people to use the service, the real breakthrough came in the form of digitized services. With digital signals transferring voice calls, the whole concept changed both in terms of volume and the cost factor. In all these the building block was the T1 line.

One of the biggest challenges of moving to digital signals is the recovery of analog signals back. Once digitized the information from the signal is lost. Nyquist defined a theorem that ensured that the analog signal can be reliably recovered from digitized version if it is sampled at twice the rate of the highest frequency component.

When it comes to digitizing voice, the most important thing was to determine the frequency range of human voice. In most normal situations the human voice does not contain many components beyond 4kHz. Hence it is sampled at 8000 samples per second as per the Nyquist theorem. This gives 0.125 seconds for each sample and this is an important time duration in telephony.

Research has shown that using 256 levels is enough to capture the sudden variations in human voice. Hence this 8000 samples are placed in 256 levels and represented by 8 bits. This gives the base rate of a voice signal as 64kbps.

With digital signals, you get one advantage of value predictability. If you know that a digital signal pulse is of 125 microseconds, you are certain that for that time duration the pulse amplitude is not going to change. Hence if you look at the first microsecond or the last you will get the same value. This lets allows for multiplexing many voice signals in within 125 microseconds.

If we can multiplex many signals into one line, the same line can carry several calls rather than just one. This is what T1 lines achieved. These were designed to carry 24 voice channels in one line. This is truly a renaissance of sorts in the telephony world.

Each voice sample is represented by 8 bits, so 24 voice channels would have 192 bits. Now given the timing sensitivity of voice signals, it was important to maintain synchronization so the T1 standard added a framing bit to the T1 frame which meant the frame now had 193 bits. To align with the voice samples, these 193 bits had to be send in 125 microseconds giving a total line rate of 1.544Mbps for a T1 line.