PDH and Digital Voice Transmission
Before fiber-optic systems became widespread, communication networks needed a way to carry multiple telephone calls digitally over a copper pair. The Plesiochronous Digital Hierarchy (PDH) provided a family of regional standards for this purpose. During the 1970s and 1980s, these hierarchies supported the expansion of digital telephone networks.
PDH combines two operations: voice digitization with and the technique of combining multiple digital streams through .
Foundation 1: Digitizing Voice with PCM
Human speech is an analog signal. To transmit it digitally, it must be converted into a stream of bits using Pulse Code Modulation (PCM). This involves three steps:
- Sampling: The analog voice signal is measured at regular intervals. For telephony, this is done 8,000 times per second (8 kHz), based on the Nyquist-Shannon theorem.
- Quantization: In ITU-T G.711, the A-law or μ-law companding law maps each sample to one of 256 discrete quantization values.
- Coding: The selected value is represented by an 8-bit binary code word.
The Basic Digital Voice Channel
The result is a 64 kbit/s digital voice channel.
Calculation:
Foundation 2: Combining Channels with TDM
A standard 64 kbit/s digital voice channel can be combined with other channels on a higher-capacity link. PDH performs this operation with Time-Division Multiplexing (TDM).
In TDM, a multiplexer takes one sample (an 8-bit byte) from each incoming voice channel in a round-robin fashion and combines them into a single, repeating structure called a frame. A demultiplexer at the other end performs the reverse operation.
This process creates higher-rate digital streams. North American T1 and Japanese J1 each carry 24 64 kbit/s voice channels at 1.544 Mbit/s. Frame overhead accounts for 8 kbit/s. European E1 carries 30 64 kbit/s voice channels at 2.048 Mbit/s. In a CAS structure, TS0 and TS16 are reserved for framing and signaling. An E1 frame repeats 8,000 times per second, so its 256 bits occupy 125 microseconds. These regional variants form first-level PDH interfaces.
The Plesiochronous Clocking Model
The name PDH comes from the Greek word plesio, meaning "near" or "close." In a plesiochronous network, the clocks have closely matched frequencies and run independently rather than being phase-locked.
Each multiplexer and digital exchange in a PDH network has its own accurate clock. Small frequency differences remain between the clocks. When several streams with slightly different rates, called tributaries, are combined into one higher-rate stream, their bit timing varies.
Justification and Bit Stuffing
PDH compensates for this timing difference with a technique called or bit stuffing. The multiplexer for the higher-rate stream runs slightly faster than the sum of the nominal tributary rates. It uses the extra capacity to insert justification bits as needed before interleaving. Control bits tell the receiving demultiplexer whether an inserted bit belongs to the payload and should be retained or discarded.
Justification permits multiplexing tributaries whose clocks run independently. Its control information and variable bit positions increase implementation complexity and limit direct access to lower-rate channels.