Jump to content

U interface

Edit links
From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by Conquerist (talk | contribs) at 15:45, 6 January 2014 (→Time Compression Multiplex (TCM): multiplex -> duplex). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

U interface: For basic-rate access in an Integrated Services Digital Network (ISDN) environment, a user-to-network interface reference point that is characterized by the use of a 2-wire-loop transmission system that

  • conveys information between the 4-wire user-to-network interface, i.e., the S/T reference point in the network termination type 1 (NT1), and the line termination (LT) in the local exchange,
  • is located in the servicing central office, and
  • is not as distance sensitive as a service using a T interface.[1]

Logical interface

Like all all other ISDN basic rate interfaces, the U interface carries two B (bearer) channels at 64 kbit/s and one D (data) channel at 16 kbit/s for a combined bitrate of 144 kbit/s (2B+D).

Duplex transmission

While in a four-wire interface such as the ISDN S and T-interfaces one wire pair is available for each direction of transmission, a two-wire interface needs to implement both directions on a single wire pair. To that end, ITU-T recommendation G.961 specifies two duplex transmission technologies for the ISDN U interface, either of which shall be used: Echo cancellation (ECH) and Time Compression Multiplex (TCM).[2]

Echo cancellation (ECH)

When a transmitter applies a signal to the wire-pair, parts of the signal will be reflected as a result of imperfect balance of the hybrid and because of impedance discontinuities on the line.[2] These reflections return to the transmitter as an echo and are indistinguishable from a signal transmitted at the far end. In the echo cancellation (ECH) scheme, the transmitter locally simulates the echo it expects to receive, and subtracts it from the received signal.[2]

Time Compression Multiplex (TCM)

The Time Compression Multiplex (TCM) duplex method, also referred to as "burst mode", solves the echo problem indirectly.[2] The line is operated at a rate at least twice the signal rate and both ends of the line take turns transmitting, in a time-division duplex fashion.[2]

Line Systems

ITU-T G.961 specifies four line-systems for the ISDN U interface: MMS43, 2B1Q, TCM, and SU32.[2] All line systems except TCM use echo cancellation for duplex operation.[2]

MMMS43

Appendix I of G.961 specifies a line system using echo cancellation and a Modified Monitoring State Code mapping 4 bits into 3 ternary symbols (MMS43), which is also referred to as 4B3T (four binary, three ternary).[2] The ternary alphabet consists of positive voltage (+), zero voltage (0), and negative voltage (-).[2] A 1 ms frame carrying 144 bits of 2B+D data is mapped to 108 ternary symbols.[2] To this frame, an 11-symbol preamble and a symbol from the CL channel are added, yielding a frame size of 120 ternary symbols and a symbol rate of 120 kBaud.[2] The CL channel is used to request activation or deactivation of a loopback in either the NT1 or a line regenerator.[2] The redundancy introduced by mapping 24=16 into 33=27 symbols is used to generate a zero DC-bias signal.[2] Each (binary) information word is either mapped to a DC-component free (ternary) code word, or can be mapped to one of two (ternary) code words, one with positive and the other with negative DC-component.[2] In the latter case, the transmitter chooses whether to send the code-word with negative or positive DC-bias based on the accumulated DC-offset.[2] Prior to transmission, the symbols from the 2B+D channels are scrambled with different codes for the two transmission directions, in order reduce correlation between transmitted and received signal.[2]

2B1Q

Appendix II of G.961 specifies a line system using echo cancellation and a line code mapping of two binary into one quaternary symbol (2B1Q).[2] A 1.5 ms frame carrying 216 scrambled bits of 2B+D data is mapped to 108 quaternary symbols.[2] To this frame, a 9-symbol preamble and 3 symbols from the CL channel are added, yielding a frame size of 120 quaternary symbols and a symbol rate of 80 kBaud.[2] The CL channel is used for communication between LT and NT1, a 12-bit cyclic redundancy check (CRC), and various other physical layer functions.[2] The CRC covers one 12 ms multiframe (8×1.5 ms frames).[2]

TCM / AMI

Appendix III of G.961 specifies a line system using the Time Compression Multiplex (TCM) duplex method and an alternate mark inversion (AMI) line code.[2] The AMI line code maps one input bit to one ternary symbol.[2] Like with MMS43, the ternary symbol can either be a positive (+), zero (0), or negative (-) voltage.[2] A 1 bit is represented by a zero voltage, while a 0 bit is alternatingly represented by a positive and a negative voltage, resulting in a DC-bias free signal.[2] In a 2.5 ms interval, each side can send a 1.178 ms frame representing 360 bits of 2B+D data.[2] To the 2B+D data, an 8-bit preamble, 8 bits from the CL channel, as well as a parity bit are added, yielding a frame size of 377 bits and a baud rate of 320 kBaud.[2] The CL channel is used for operations and maintenance, and a 12-bit CRC covering 4 frames.[2]

SU32

Appendix IV of G.961 specifies a line system using echo cancellation and a substitutional 3B2T (SU32) line code, which maps three bits into 2 ternary symbols.[2] As with MMS43 and AMI, the ternary symbol can either be a positive (+), zero (0), or negative (-) voltage.[2] The mapping from 23=8 to 32=9 symbols leaves one unused symbol.[2] When two subsequent input (binary) information words are identical, the (ternary) code word is substituted by the unused code word.[2] A 0.75 ms frame carrying 108 bits of 2B+D data is mapped to 72 ternary symbols.[2] To this frame, a 6-symbol preamble, one CRC symbol, and 2 symbols from the CL channel are added, yielding a frame size of 81 ternary symbols and a symbol rate of 108 kBaud.[2] The CL channel is used for supervisory and maintenance functions between the LT and NT1.[2] The 15-bit CRC covers 16 frames.[2]

References

  1. ^ Public Domain This article incorporates public domain material from Federal Standard 1037C. General Services Administration. Archived from the original on 2022-01-22.
  2. ^ a b c d e f g h i j k l m n o p q r s t u v w x y z aa ab ac ad ae af ag ah ai aj "G.961: Digital transmission system on metallic local lines for ISDN basic rate access". ITU-T. March 1993. Retrieved 2014-01-06.

See also

Further reading

  • Lechleider, J. (September 1989), Line codes for digital subscriber lines (ISDN basic access), IEEE
Retrieved from "https://en.wikipedia.org/w/index.php?title=U_interface&oldid=589454826"