Previous Table of Contents Next


Unchannelized T-carrier

Unchannelized T- carrier can be used to support bandwidth-intensive services that do not lend themselves to 64 Kbps channelization and standard framing conventions [8-8]. In other words, the traditional convention of 64 Kbps channels can be abandoned in favor of carving the T1 pipe into any segments of bandwidth of any usable size or increment. Additionally, any combination of bits can be transmitted, including an infinite number of zeros, without concern for the violation of the 1s density rules; in other words, a clear channel of 64 Kbps or more, rather than a 56 Kbps channel [8-5] and [8-6]. For instance, a very high-speed data communication or a full motion videoconference, might require a full T1 pipe. A less intensive communication might require 512 Kbps (8 channels, or one-third of a T1 facility).

Such services are supported through customer equipment in the form of highly intelligent TDM MUXs. In a private, dedicated leased-line network, this is easily accomplished. However, the carrier must be aware of the fact that such use will be made of the facility, in order that the entire facility can be properly allocated and managed.

Encoding

While T1 is a digital service, it supports the transmission of analog data as well. Voice and video, analog in their native forms, must be digitized through the use of codecs prior to being transmitted over a T1 circuit. The standard digitizing technique for voice, known as Pulse Code Modulation (PCM), was developed as an integral part of T-carrier. It also became the standard technique for digitizing voice in PBXs and other devices, for the obvious reason of providing seamless transmission between such devices and the network. The quantizing techniques typically employed include PCM and ADPCM; non-standard approaches included CVSD, VQL, VQC and HCV.

Pulse Code Modulation (PCM)

PCM is based on the Nyquist theorem developed by Harry Nyquist in 1928. Nyquist established the fact that the maximum signaling rate achievable over a circuit is twice the number of signal elements [8-9]. In consideration of the Nyquist theorem, PCM specifies that the analog voice signal be sampled at twice the highest frequency on the line. As a voice-grade analog line provides 4,000 Hz of bandwidth, the signal must be sampled 8,000 times per second. Each sample is a measurement of the amplitude of the sine wave; the frequency of signal change is automatically taken into account. The individual samples are encoded (quantized) into an 8-bit binary (digital) approximate value, based on a table of 255 standard values of amplitude. The individual samples then are transmitted in regular time slots over the T-carrier circuit, with the process reversed on the receiving end of the connection in order to reconstitute an approximation of the original analog voice signal (Figure 8.4). While the sampling rate and the 8-bit coding scheme were the subject of much debate, the yield is very high quality voice, although the bandwidth requirement is a bit excessive. It should be noted that sampling that is too infrequent results in a reconstructed analog voice signal that is less than smooth and accurate and, therefore, not pleasing. This phenomenon yields unacceptable levels of quantizing noise [8-2] and [8-5].


Figure 8.4  PCM encoding of analog voice signal, with reconstruction of approximate analog voice.

      4,000 Hz
        X 2
      8,000 Samples/Second
        X 8 Bits per Sample
      64 Kbps

As noted in the above calculation, 8,000 8-bit samples per second yields a bandwidth requirement of 64 Kbps for a PCM-encoded digital voice signal. As PCM was the first standard technique widely used in digital carrier systems, the 64 Kbps channel became standard, worldwide, for all forms of digital networking.

Differential Pulse Code Modulation (DPCM)

DPCM is more bandwidth efficient than PCM, as only the changes in signal level are encoded and transmitted. Based on the logical assumption that the change, or differential, in the voice signal occurs relatively slowly, fewer bits can be used to represent each sample; 4 bits are generally used in this form of compression, which yields a 2:1 compression ratio. DPCM generally provides voice quality comparable to that of PCM. However, noise (distortion) may result on the occasions in which the signal varies significantly from one sample to another (e.g., a 4,000 bps modem transmission over a T-carrier circuit).

Adaptive Differential Pulse Code Modulation (ADPCM)

ADPCM can be used to further improve on the quality of DCPM, without increasing the number of bits required. Through increasing the range of signal changes which can be represented by a 4-bit value, DPCM is adapted to provide higher transmission quality. As the result is a 32 Kbps voice stream, the bandwidth per conversation is halved; twice as many conversations (48 versus 24) can be carried over a T1 facility [8-5].

    4,000 Hz
      x 2
    8,000 Samples/Second
      x 4 Bits per Sample
    32 Kbps

As ADPCM will not interface with a CO exchange based on PCM, it is necessary that special equipment be used to insert two compressed voice conversations into a single PCM channel. A Bit Compression Multiplexer (BCM) generally is in the form of a printed circuit board which fits into the T1 MUX [8-5].

Digital Speech Interpolation (DSI)

DSI is rooted in a voice compression algorithm known as Time Assigned Speech Interpolation (TASI) developed by Bell Labs in the 1950s for transatlantic telephone cable systems [8-10]. DSI makes the legitimate assumption that there are predictable pauses in normal human speech; during those pauses, additional voice signals are inserted. This technique is sometimes referred to as silence suppression. As DSI works on the basis of statistical probabilities, it can be employed effectively only when there are a significant number of voice conversations supported. For instance, 72 channels yield additional compression of 1.5:1, and 96 channels yield an additional 2:1, for a total compression ratio of 8:1 (8 Kbps per voice conversation). Newer implementations can provide as much as 16:1 (4 Kbps).

DSI suffers the disadvantage of degradation of the signal quality during periods of heavy use. If the parties are speaking rapidly, with few pauses, the voice signal can be clipped. The more conversations supported, however, the lower the statistical probability of such degradation. DSI also suffers from high overhead. Regardless of the number of channels supported, as much as 96 Kbps of overhead is required.


Previous Table of Contents Next