| 词条 | U interface |
| 释义 |
The U interface or U reference point is a Basic Rate Interface (BRI) in the local loop of an Integrated Services Digital Network (ISDN). It is characterized by the use of a 2-wire transmission system that connects the network termination type 1 (NT1) on the customer's premises and the line termination (LT) in the carrier's local exchange. It is not as distance sensitive as a service using an S interface or T interface.[1] In America, the NT1 is customer premises equipment (CPE) which is purchased and maintained by the user, which makes the U interface a User–network interface (UNI).[2] The American variant is specified by the American National Standards Institute (ANSI) in T1.601.[3][2] In Europe, the NT1 belongs to the network operator, so the user doesn't have direct access to the U interface.[2] The European variant is specified by the European Telecommunications Standards Institute (ETSI) in recommendation ETR 080.[4][2] The ITU-T has issued recommendations G.960 and G.961 with world-wide scope, encompassing both the European and American variants of the U interface.[2] Logical interfaceLike 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 transmissionWhile 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).[5] 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.[5] 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.[5] Time Compression Multiplex (TCM){{Distinguish|Trellis coded modulation}}The Time Compression Multiplex (TCM) duplex method, also referred to as "burst mode", solves the echo problem indirectly.[5] 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.[5] Line SystemsITU-T G.961 specifies four line systems for the ISDN U interface: MMS43, 2B1Q, TCM, and SU32.[5] All line systems except TCM use echo cancellation for duplex operation.[5] The American standard ANSI T1.601 specifies the 2B1Q line system, the European ETSI TR 080 recommendation specifies 2B1Q and MMS43.[2] MMMS43 (4B3T){{Main|4B3T}}The Modified Monitoring State Code mapping 4 bits into 3 ternary symbols (MMS43), which is also referred to as 4B3T (four binary, three ternary) is a line system used in Europe and elsewhere in the world. 4B3T is a "block code" that uses Return-to-Zero states on the line. 4B3T converts each group of 4 data bits into 3 "ternary" line signal states (3 symbols). Echo cancellation techniques allow full-duplex operation on the line. MMS43 is defined in Appendix I of G.961,[5] Annex B of ETR 080,[4] and other national standards, like Germany's 1TR220. 4B3T can be transmitted reliably at up to {{val|4.2|u=km}} over {{val|0.4|u=mm}} cable or up to {{val|8.2|u=km}} over {{val|0.6|u=mm}} cable. An internal termination impedance of {{gaps|150|ohms}} is presented to the line at each end of the U-interface. A 1 ms frame carrying 144 bits of 2B+D data is mapped to 108 ternary symbols.[5] These symbols are scrambled, with different scrambling codes for the two transmission directions, in order reduce correlation between transmitted and received signal.[5] 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 kilobaud.[5] The CL channel is used to request activation or deactivation of a loopback in either the NT1 or a line regenerator.[5] In 4B3T coding, there are three states presented to line: a positive pulse (+), a negative pulse (-), or a zero-state (no pulse: 0). An analogy here is that operation is similar to B8ZS or HDB3 in T1/E1 systems, except that there is an actual gain in the information rate by coding 24=16 possible binary states to one of 33=27 ternary states. This added redundancy is used to generate a zero DC-bias signal.[5] One requirement for line transmission is that there should be no DC build-up on the line, so the accumulated DC build-up is monitored and the codewords are chosen accordingly. Of the 16 binary information words, some are always mapped to a DC-component free (ternary) code word, while others can be mapped to either one of two code words, one with a positive and the other with a negative DC-component.[5] In the latter case, the transmitter chooses whether to send the code-word with negative or positive DC-component based on the accumulated DC-offset.[5] 2B1Q{{Main|2B1Q}}2B1Q coding is the standard used in North America, Italy, and Switzerland. 2B1Q means that two bits are combined to form a single Quaternary line state (symbol). 2B1Q combines two bits at a time to be represented by one of four signal levels on the line. Echo cancellation techniques allow full-duplex operation on the line. 2B1Q coding is defined in Appendix II of G.961,[5] ANSI T1.601,[3] and Annex A of ETR 080.[4] It can operate at distances up to about 18,000 feet ({{val|5.5|u=km}}) with loss up to {{gaps|42|dB}}. An internal termination impedance of 135 ohms is presented to the line at each end of the U-interface. A 1.5 ms frame carrying 216 scrambled bits of 2B+D data is mapped to 108 quaternary symbols.[5] 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 kilobaud.[5] The CL channel is used for communication between LT and NT1, a 12-bit cyclic redundancy check (CRC), and various other physical layer functions.[5] The CRC covers one 12 ms multiframe (8×1.5 ms frames).[5] TCM / AMIThe TCM / AMI ISDN line system, also referred to as TCM-ISDN, is used by Nippon Telegraph and Telephone in its "INS-Net 64" service.[6] Appendix III of G.961 specifies a line system based on the Time Compression Multiplex (TCM) duplex method and an alternate mark inversion (AMI) line code.[5] The AMI line code maps one input bit to one ternary symbol.[5] Like with MMS43, the ternary symbol can either be a positive (+), zero (0), or negative (-) voltage.[5] A 0 bit is represented by a zero voltage, while a 1 bit is alternatingly represented by a positive and a negative voltage, resulting in a DC-bias free signal.[5] In a 2.5 ms interval, each side can send a 1.178 ms frame representing 360 bits of 2B+D data.[5] 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 kilobaud.[5] The CL channel is used for operations and maintenance, as well transmitting a 12-bit CRC covering 4 frames.[5] SU32Appendix IV of G.961 specifies a line system based on echo cancellation and a substitutional 3B2T (SU32) line code, which maps three bits into 2 ternary symbols.[5] As with MMS43 and AMI, the ternary symbol can either be a positive (+), zero (0), or negative (-) voltage.[5] The mapping from 23=8 to 32=9 symbols leaves one unused symbol.[5] When two subsequent input (binary) information words are identical, the (ternary) code word is substituted by the unused code word.[5] A 0.75 ms frame carrying 108 bits of 2B+D data is mapped to 72 ternary symbols.[5] 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 kilobaud.[5] The CL channel is used for supervisory and maintenance functions between the LT and NT1.[5] The 15-bit CRC covers 16 frames.[5] References1. ^{{FS1037C}} 2. ^1 2 3 4 5 {{cite book |url=https://books.google.com/books?id=Vxw6I2eC2RkC |chapter=Section 4.5: U interface standards |chapterurl=https://books.google.com/books?id=Vxw6I2eC2RkC&pg=PA122 |last=Burd |first=Nick |date=1997 |publisher=Chapman & Hall |location=London |title=ISDN Subscriber Loop|isbn=9780412497308 }} 3. ^1 {{cite web |url=http://www.etsi.org/deliver/etsi_etr/001_099/080/02_60/etr_080e02p.pdf |title=ETR 080: Digital transmission system on metallic local lines |publisher=ETSI |date=November 1996 |accessdate=2014-01-17}} 4. ^1 2 {{cite web |url=http://ftp.tiaonline.org/TR-30/TR-30.3/Public/xDSL%20Miscellaneous/T1_601_i3Draft.pdf |title=T1E1.4/98-004R2 DRAFT ANSI T1.601-1998: Integrated Services Digital Network (ISDN) Basic Access Interface for Use on Metallic Loops for Application on the Network Side of the NT (Layer 1 Specification) |date=1998 |accessdate=2014-01-17 |publisher=ANSI}} 5. ^1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 {{cite web|url=http://www.itu.int/rec/T-REC-G.961|title=G.961: Digital transmission system on metallic local lines for ISDN basic rate access |publisher=ITU-T |accessdate=2014-01-06 |date=March 1993}} 6. ^{{cite journal |url=http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=149619 |title=Implementation in Japan (ISDN) |journal=IEEE Communications Magazine |volume=30 |issue=8 |pages=54–57 |doi=10.1109/35.149619 |date=August 1992 |subscription=yes |publisher=IEEE |last=Inoue |first=O.}} See also
Further reading
2 : ITU-T recommendations|Integrated Services Digital Network |
| 随便看 |
|
开放百科全书收录14589846条英语、德语、日语等多语种百科知识，基本涵盖了大多数领域的百科知识，是一部内容自由、开放的电子版国际百科全书。