T1 Line Voice and Data

Understanding A T1 Line Digital Signal 1

For many people in North America, the telephone lines are nothing but a privilege that you just get. However when the telephone was first invented, getting it to a single person was a big challenge. As the need to communicate grew there was growing demand to add more service. Telephone in its original form was not scalable. However, digitization changed all this. With the T1 line driving the network expansions, the growth was rapid and swift.

Signal digitization involves converting the analog signals into pulses. In this process not all the information in the analog signal can be captured. The issue with switching with signal digitization was to recover the original signal reliably. This is where Nyquist theorem comes into picture. T1 signals are closely based on the Nyquist theorem

With a few exceptions, the highest frequency component in any human voice rarely exceeds 4kHz. This is important information as it means that as per the Nyquist theorem we can sample the voice at 8000 samples in a second.

The next big step in voice digitization is to map these samples in a manner that can be transmitted over the network. To do this, the samples are mapped on to one of the 256 pre-defined levels. These 256 levels are represented by 8 bits. Thus with 8 bits per sample and 8000 samples per second the voice signals require 64kbps bandwidth.

Digital signals are typically pulses of constant amplitude for fixed duration. In case of voice samples this duration is 125 microseconds. So if you has a system that can work faster than 125 microseconds, it can then measure more than one signal in the time frame and thus multiplex many such signals.

The T1 standard was defined to carry 24 voice channels based on this multiplexing concept. Thus the T1 processing system could collect information from 24 channels within 125 microseconds and create one composite multiplexed signal of that duration. This allowed many users to share the medium rather than use dedicated channels.

Each voice channel as stated above contains 8 bits of information. Thus a frame with 24 voice channels would contain 192 bits, which is 8 times 24. To ensure proper synchronization the T1 standards also defined a framing bit giving the T1 frames a total of 193 bits. Since the voice signals are sampled at 8000 samples per second which means each sample is 125 microseconds long, for proper multiplexing the 193 T1 frame bits must be sent within this time. This gives the T1 line rate of 1.544Mbps.