QUESTION:
Are E1 links used for data different from those used for voice?
ANSWER:
There is no simple answer to your question. Nevertheless, I will try to explain what different flavours of E1 links can be encountered.
First, all E1 links have a particular electrical specification (see ITU-T G.703, for example). The line coding is specified (HDB3) and the bit-rate is specified (2.048 Mbps, nominal); the nominal clock frequency underlying an E1 is 2.048 MHz (and must be within 50 ppm). When an E1 is carried as a tributary (for example within an E3 or an SDH signal), the electrical specifications are moot but the clock and format are important.
Second, an E1 bit-stream could be framed or unframed. When it is framed, there is the notion of 32 time-slots every 125 microseconds where each time-slot corresponds to an octet (byte). One of the time-slots contains a frame-synch pattern to identify the 32 time-slots in the bit-stream. Thus each time-slot can be viewed as a 64 kbps channel. When un-framed, it is a raw 2.048 Mbps bit-stream and the terminating equipments utilize this stream in an appropriate, possibly proprietary, manner.
Third, when an E1 is used to carry voice in the conventional (time-division-multiplex, or TDM) manner, it is common to assign 30 time-slots for voice and one time-slot for signalling information. In CAS (channel associated signalling), time-slot-16 carries the signalling bits. In the case of a PRI (Primary Rate Interface), 30 time-slots carry 30 "bearer channels" by assigning one time-slot per bearer channel (each bearer is 64 kbps or "DS0") and one time-slot carries the (packetized) signalling information in what is called the "D Channel" (hence an E1-PRI is often called "30B+D"). In this scenario the E1 terminates in the Central Office Switch(i.e. the Voice Switch deployed by the Network Operator) or a Cross-connect device (DCS-1/0). The timing for the E1 in this case is provided by the CO equipment and is usually stratum-1-traceable; the Customer-Premise-Equipment (CPE) must "loop-time", deriving all its clocks from the incoming E1. A common cause of failure or substandard functionality is when the CPE does not loop-time correctly.
Fourth, when an E1 is used for data (voice would be in the form of VoIP, i.e., packetized) it could be framed or un-framed. The terminating equipments determine the format which could be HDLC or ATM (or a proprietary layer-2 format). When un-framed, the entire 2.048 Mbps bit-stream is available for "data"; when framed, up to 31 time-slots can be assigned to "data" for a net ("usable") bit-rate of 31x64 kbps.
|
