As mentioned elsewhere, DDS uses a form of Bipolar, return-to-zero, Alternate Mark Inversion for transmission over a 4-Wire metallic pair circuit.
At a transmission rate of 56K, the Nyquist frequency is 28KHz. A binary zero is represented as zero volts on the line, while a binary one is transmitted in the form of a positive or negative pulse. Each binary one pulse is of opposite polarity from the previous pulse (Alternate Mark Inversion). The Alternate Mark Inversion prevents a build-up of the DC level on the line. Operation is similar to that of T1 AMI line encoding. HOWEVER, DDS/SW56K CIRCUITS EMPLOY UNIQUE BIPOLAR VIOLATION CODE SEQUENCES! THE STRUCTURE AND USE OF THESE SPECIAL BIPLOLAR VIOLATION SEQUENCES IS DISCUSSED LATER ON IN THIS DOCUMENT.
With DDS, a single control station can communicate to multiple tributary stations (Multi-point DDS). This is accomplished through the use of a Multi-point Junction Unit (MJU). The MJU can open the "reverse" channel path upon detection of channel data activity (space-bit detection), or upon sensing a change in channel line activity (from "Control Mode Idle" to "Data Mode"). In the case of the latter; like standard data modems, the Inactive to Active states are controlled by the state of the end-device's RTS lead. When RTS is low/off, the CSU/DSU reflects this state by transmitting a Control Mode Idle sequence to the network. When RTS is high/on, the CSU is actively transmitting data from the DTE device and is in "Data Mode".
DDS and SW56K codes are expressed in the following notations:
Return to the DDS page. 0 = Zero volts transmitted
B = Positive or Negative Pulse (Binary one)
V = Positive or Negative Pulse in violation of the AMI rule
X = Either a 0 or B, depending upon the required polarity of a
violation
N = Either a 0 or B is acceptable (Don't care)
The X and V bits are always separated by a 0, since consecutive pulses of the
same polarity could degrade performance.
Simplex DC current (Sealing current) is provided between the Transmit and Receive pairs. This current helps to prevent oxide corrosion in cable junctions and splices.
However, this simplex current also provides a loopback function. When NORMAL, the Transmit Pair (T1, R1) is kept positive with respect to the Receive Pair (T,R). When a CSU LOOPBACK is desired, this polarity is REVERSED. The CSU detects the current reversal and enters a loopback state back towards the connected CO.
The DC simplex current has the following characteristics:
The phone network usually relies on centralized diagnostic centers for DDS testing. This section details the diagnostic capabilities available to the test centers.
Proper operation is tested by the service provider through the use of loopbacks. Additionally, testing is usually consolidated in centralized "hub" locations. Through the use of special DS0 codes, loopbacks can be effected at Remote Office Channel Units (OCU). Upon receipt of a CSU loopback code, the Remote OCU can reverse the simplex power, resulting in a loopback in the customer's CSU equipment.
There are TWO types of loopbacks: Latching and Non-Latching. Standard DDS service utilizes Non-Latching loopbacks. DDS w/ Secondary Channel can use both Non-Latching and Latching loopbacks. A "new" service, 64K DDS, ONLY utilizes Latching loopbacks.
When a Latching loopback is invoked, a specific loopback activate and loopback deactivate sequence is sent.
When a Non-Latching loopback is invoked, the loopback is maintained as long as EVERY OTHER BYTE RECEIVED CONTAINS THE LOOPBACK CODE! Effective Bit Rate throughput for BERT tests during these loopbacks is actually 28 KBPS!