1、Report No. 52 A Technical ReportonNovember 1996 The Value andInterpretation ofDigital Pulse Masksand Eye PatternsPrepared byT1X1.4 Working GroupMetallic HierarchicalInterfacesCommittee T1 is sponsored by the Alliance for Telecommunications Industry SolutionsAccredited by American National Standards
2、InstituteCopyright 1996 by Alliance for Telecommunications IndustrySolutions All rights reserved.No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher.iiA Technical Report onThe Value and Inter
3、pretationof Digital Pulse Masks and Eye PatternsAbstract: Since the late 1950s, the eye pattern, as viewed at the input of a detector, has provena practical and effective means of analyzing the performance of digital systems and equipment.A later development has been the use of a picture of an indiv
4、idual pulse or an eye pattern,unaffected by intersymbol interference or jitter, and viewed at an interface point. Templateshave been prescribed for both and are widely used in todays specifications and standards toensure “origination” of the correct signal at a network interface point rather than fo
5、rmeasurements at the input of a detector for determination of “receiving performance”. Thepurpose of this technical report is to assist technicians and engineers in the application andinterpretation of such signal templates when used for test and maintenance purposes.T1X1.4/95-016R3Prepared by T1X1.
6、4Metallic Hierarchical InterfacesWorking GroupiiiTABLE OF CONTENTSPARAGRAPH PAGE1.0 SCOPE AND PURPOSE 12.0 REFERENCED STANDARDS AND PUBLICATIONS 13.0 ACRONYMS AND DEFINITIONS 24.0 BACKGROUND 44.1 Introduction of the DSX Cross Connect 54.2 Network Interconnections 54.3 Interface Standards 55.0 PULSE
7、REPRESENTATIONS AND DETECTION 65.1 AMI vs CMI 65.2 Why an Interface Specification? 66.0 PULSE MASK AND EYE PATTERN TEMPLATE MEASUREMENTS 76.1 Evolution of Test Equipment 76.2 Measurement Techniques 76.2.1 Isolated Pulses 76.2.2 Eye Patterns 97.0 INTERPRETATION OF MEASUREMENTS 107.1 System Performanc
8、e vs Template Conformance 107.1.1 Line Equipment 117.1.2 Crossconnect Points 117.2 Discrepancies at various parts of the Mask 127.2.1 Baseline 127.2.2 Leading Edge 127.2.3 Top of Pulse 137.2.4 Trailing Edge 137.2.5 Overshoot 137.2.6 Undershoot 137.2.7 Pulse Width 147.3 Simulation of DS1 Eye Pattern
9、viewed at Detector Input 147.4 Examples of STS3 Eye Patterns 14ivTABLE OF CONTENTS (cont)8.0 RECOMMENDATIONS FOR PULSE EVALUATIONS 158.1 In-Service vs Out-of-Service 158.2 Timing Signal Generators 168.3 Bit Error Performance 16Figure Page1 Test Setup For Pulse Template Evaluation 172 AMI vs CMI Deci
10、sion Points 173 Network Transmission Test Access Points 184 DS1 Pulse Mask Discrepancies 195 DS1 Eye Patterns 226 STS-3 Eye Patterns 237 Insertion Loss and Phase Characteristics 271TECHNICAL REPORT ONTHE VALUE AND INTERPRETATIONOF DIGITAL PULSE MASKS AND EYE PATTERNS1.0 SCOPE AND PURPOSEStandardizat
11、ion of electrical interfaces for the digital hierarchy, at rates ranging fromDS1 through STS3 and above, has been determined a necessary function to assureinteroperability of both equipment and networks.Information in the digital hierarchy is transported via cable pairs or coax links by eitherof two
12、 means:(1) By signals represented by the presence or absence of pulses occurring within specific time slots or,(2) By signals consisting of continuous pulses encoded to represent binary ones and zeros.Pulse mask and eye pattern templates have been established by ANSI and ITU-T(CCITT) to provide info
13、rmation on the signals allowable at interface points. Inconjunction with other specifications such as signal power, bit rate, and line code, thesepulse masks and eye patterns, are used to ensure compatible operation of equal-levelDS-n and STS-n signals at interface points provided by the DSX-n and S
14、TSX-n CrossConnects. In principle, however, none of the existing waveform specifications - isolatedpulse templates for DS1 thru STS1 or eye patterns for STS11, STS3, and above -provide a direct measure of transmission performance at the system level.This technical report provides an introduction to
15、the basis for the use of pulses masksand eye patterns to determine signal characteristics at specific electrical interfacepoints in the digital network. Included are definitions and discussions of desired signalparameters, a review of various measurement techniques used to verify that thedesired par
16、ameters are met, and an interpretation of the observed results whenmeasurements are performed using first a non-inductive resistive load and then whenthe signals are terminated by their normal (reactive) sink.2.0 REFERENCED STANDARDS AND PUBLICATIONS1) ANSI T1.102-1993“Digital Hierarchy - Electrical
17、 Interfaces”1STS1 is described by both pulse mask and eye pattern specifications.22) ANSI T1.403-1995“Network-to-Customer Installation - DS1 Metallic Interface”3) ANSI T1.404-1994“Network-to-Customer Installation - DS3 Metallic Interface Specification”4) ANSI T1.408-1990ISDN Primary Rate - Customer
18、Installation Metallic Interface,Layer 1 Specification”5) ATIS Committee T1 Telecommunications - Technical Report # 4 - 1990“A Technical Report on the Mathematical Modeling of DS1, DS1A, DS1C, DS3, and DS4NA Waveshapes”6) IEEE Standard 1007 - 1991“Methods and Equipment for Measuring the Transmission
19、Characteristicsof PCM Telecommunications Circuits and Systems”7) CCITT (ITU-T) Recommendation G.703-1991“Physical/Electrical Characteristics of Hierarchical Digital Interfaces”3.0 ACRONYMS AND DEFINITIONSAMI - Alternate Mark Inversion.A three-level, return-to-zero (RZ) line code that employs a terna
20、ry signal to convey datain the form of binary digits. Successive binary ones are represented by equalamplitude signal elements that are normally of alternating positive and negativepolarity. Binary zeros are represented by signal elements that have zero amplitude.North American implementations use s
21、ignal elements representing binary ones that arenon-zero for only half the unit interval (50% duty cycle). The terms bipolar and pseudo-ternary are also used to describe these implementations.BER - Bit Error RatioThe ratio of the number of bits received in error to the total number of bits transmitt
22、ed ina given time interval.CMI - Coded Mark Inversion.A two-level, non-return-to-zero (NRZ) line code in which each binary zero is coded suchthat both amplitude levels appear during the unit interval corresponding to the binaryzero. A low level is transmitted for the first half of the unit interval
23、followed by a highlevel for the second half. A binary one is coded such that only one of the two levelsappears during the unit interval corresponding to the binary one. The choice of high orlow level for a binary one is made to ensure that the level alternates for successivebinary ones. For electric
24、al implementations of this code, the low level is designated asA1; the high level as A2. Levels A1 and A2 are of equal and opposite polarity with A23being more positive than A1. The CMI code provides that a minimum of one datatransition occurs per bit period - an aid in the clock recovery process.DE
25、CISION POINTThe point in the signal detection process where, with AMI, a determination is made asto the presence or absence of a signal; with CMI, a determination is made as to therelative polarity of the signal sample with respect to the previous signal sample.EYE PATTERNA convenient pictorial tech
26、nique for determining the practical effect of degradationsintroduced into the pulse stream. An eye diagram is formed by superimposing thewaveform of all possible pulse sequences. In an m-level system, where m denotes thenumber of possible signal levels, there will be m-1 eyes.ISOLATED PULSEA pulse t
27、hat: (a) is not affected by intersymbol interference (ISI) from other pulses,and (b) is not corrupted by noise.LBO - Line BuildoutA technique for extending the electrical length of a transmission line in order that theelectrical characteristics of the physical portion and electrical length provided
28、by the linebuildout, when combined, add up to a predetermined amount of insertion loss andphase rotation. (Some of the more modern line repeaters, etc., may include an ALBOAutomatic Line Buildout function which automatically provides an appropriate amountof line buildout prior to the detection proce
29、ss).NE - Network ElementNetwork equipment such as line repeaters, add-drop multiplexes, etc.NEXT - Near End CrosstalkCrosstalk that occurs at the near end of a cableNRZ - Non-Return-to-ZeroA signal where successive ones or zeros follow one another without resting at zero2.PRBS - Pseudo-Random Binary
30、 SequenceA binary sequence that approximates a random signal. The PRBS pattern is 2n-1 bitsin length and generates every combination of n-bit wordsRZ - Return-to-ZeroA signal where there must be a return to zero level between successive ones2.SINKThe termination point of a digital signal - also refe
31、renced as load.SOURCEThe origination point of a digital signal.2The voltage values of ones and zeros are arbitrary and may be positive, negative or both.44.0 BACKGROUNDSince the late 1950s, pictorial representation of digital pulses, in the form of eyediagrams, has been utilized by engineers and tec
32、hnicians to determine systemperformance. At that time, the eye diagram, as viewed at the input of the detector,proved both a practical and effective means for assessing the relative performance ofdigital systems made up of random ones and zeros. A later development was tocontrol pulse shapes at inte
33、rface points by use of a pictorial display of an individualisolated pulse having minimal Intersymbol Interference (ISI) and suffering negligibleeffects from timing jitter. The isolated pulse was superimposed on a pulse masktemplate and inspected at an interface point (e.g., DSX-n) to assure transmis
34、sion of aproper pulse - a technique which is widely used in current interface specifications andstandards. Today, lacking access to a detector input, the original use of an eye patternis no longer a practical technique for assessing system performance.Concurrent with the use of CMI for DS4NA and SON
35、ET rates, eye pattern templateswere developed for inspection of CMI signals at interface points. Unlike AMI signals,the “random” nature of the CMI encoded electrical signals3has made it moreconvenient to characterize such waveforms by the use of eye patterns rather thanpulse waveforms4. Use of eye p
36、atterns also permits qualification of the CMI signal byrevealing signal deterioration due to ISI, pulse shape asymmetry, noise, and jitter.Whereas AMI uses a pulse to represent a one and the absence of a pulse to representa zero, CMI utilizes signal pulses to represent both ones and zeros. The pulse
37、srepresenting ones are twice the width (half the frequency) of the pulses representingzeros. Inspection of the waveform of an isolated pulse generally requires that the CMIsystem be taken out-of-service and keyed with either all ones or all zeros; the use of aneye pattern eliminates the need to take
38、 the system out of service4. In North America(ANSI), CMI signals are inspected by means of an eye pattern which can be viewedeither in-service or out-of-service and keyed by either traffic or a PRBS. At the STS1SONET rate, which utilizes the AMI rather than the CMI line code, both pulse masksand eye
39、 patterns have been specified.It is important to note that, by itself, compliance with either a pulse mask or eye patterntemplate at an interface does not ensure that a system will perform at or below a givenBER threshold. Such performance depends upon the characteristics of theinterconnecting facil
40、ity beyond the interface point and the particular receiver andequalization implemented prior to detection.4.1 Introduction of the DSX Cross ConnectThe early deployments of T1 carrier and D-type channel banks did not recognize theneed for an office Cross Connect frame to facilitate centralized circui
41、t rearrangement,maintenance, and growth. In the early 1970s, a Digital Signal Cross Connect (DSX)3Since STS1 and STS3 signals are scrambled, neither an all ones nor all zeros pattern is available atmany SONET NEs.4ITU-T Recommendations for SDH provide for the observation of CMI pulse waveforms at th
42、esignal source and reserve the use of the eye pattern for observations elsewhere in the network.See also footnote 5.5frame was introduced for DS1 signals; however, existing office repeater designs limitedthe distance this equipment could be located from the frame. Later development ofextended range
43、equipment made the DSX a fully-functional interface point withsymmetric cable restrictions on either side of the frame. These DSX frames have sinceproven to be useful for test and measurement access of digital circuits and facilities.In the same general time period, the higher levels of the digital
44、hierarchy (DS2 andDS3) were being developed. As these incorporated the DSX concept from the outset,the digital signals at these interface points could be characterized by a series ofparameters reflecting the range of connecting cable lengths, environment, etc.Originally, these signal specifications
45、(including pulse mask templates) were directedtoward equipment designers to ensure that compatible signals were produced by thevariety of digital equipment that would feed higher level multiplexes. The signal accessprovided by the DSX has since assumed the role of test/measurement access pointsfor s
46、ystems maintenance. Specifications for SONET (STSX-n) electrical interfaces arecurrently under study.4.2 Network InterconnectionsBeginning in the mid 1970s and continuing to the present, there have been legal andregulatory requirements to increase network access permitting various types ofconnection
47、s to the digital networks, thereby opening the market to outside suppliers ofnetwork equipment. This necessitated the preparation of a family of CompatibilityBulletins to address both hardware and interface technical requirements. The Interfacespecifications were largely taken from the earlier signa
48、l characterizations intended fordesigners and retained all of the specifications including the isolated pulse masktemplates.4.3 Interface StandardsPost divestiture, one of the first North American standards developed under theauspices of Committee T1 was ANSI T1.102 which standardized interfacespeci
49、fications for the digital hierarchy. This standard drew heavily on the earlierA T zeros are represented by the absence of a pulse. Signal rates above STS1,including DS4NA, utilize the coding scheme known as CMI.As discussed in section 4, the CMI signal does not lend itself to the use of a pulse masktemplate for in-service measurements and has been provided, instead, an eye patterntemplate as a portion of the inter
