C:\WINWORD\CCITTREC.DOT_______________ INTERNATIONAL TELECOMMUNICATION UNION CCITT G.957 THE INTERNATIONAL TELEGRAPH AND TELEPHONE CONSULTATIVE COMMITTEE DIGITAL NETWORKS, DIGITAL SECTIONS AND DIGITAL LINE SYSTEMS OPTICAL INTERFACES FOR EQUIPMENTS AND SYS- TEMS RELATING TO THE SYNCHRONOUS DIGITAL HIERARCHY Recommendation G.957 Geneva, 1990 Printed in Switzerland FOREWORD The CCITT (the International Telegraph and Telephone Consultative Committee) is a permanent organ of the International Telecommuni- cation Union (ITU). CCITT is responsible for studying technical, operating and tariff questions and issuing Recommendations on them with a view to standardizing telecommunications on a worldwide basis. The Plenary Assembly of CCITT which meets every four years, establishes the topics for study and approves Recommendations pre- pared by its Study Groups. The approval of Recommendations by the members of CCITT between Plenary Assemblies is covered by the procedure laid down in Resolution No. 2 (Melbourne, 1988). Recommendation G.957 was prepared by Study Group XV and was approved under the Resolution No. 2 procedure on the 14th of December 1990. ___________________ CCITT NOTE In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication Administration and a recognized private operating agency. ã ITU 1990 All rights reserved. No part of this publication may be reproduced or uti- lized in any form or by any means, electronic or mechanical, including pho- tocopying and microfilm, without permission in writing from the ITU. Recommendation G.957 Recommendation G.957 OPTICAL INTERFACES FOR EQUIPMENTS AND SYSTEMS RELAT- ING TO THE SYNCHRONOUS DIGITAL HIERARCHY 1 Introduction This Recommendation covers optical interface parameter specifica- tions for equipments and systems supporting the Synchronous Digital Hier- archy (SDH) defined in RecommendationsG.707, G.708 and G.709 and operating on single-mode optical fibres conforming to RecommendationsG.652, G.653 and G.654. The purpose of this Recommendation is to provide specifications for the optical interfaces of SDH equipment, described in RecommendationsG.782 and G.783, and line systems described in RecommendationG.958, to achieve the possibility of transverse (multiven- dor) compatibility on elementary cable sections, i.e.the possibility of mix- ing various manufacturers' equipments within a single optical section. However, the specifications in this Recommendation are also intended to be in accordance with RecommendationsG.955 and G.956 which provide the possibility to achieve longitudinal compatibility for equipment of compara- ble hierarchical level and application. The present Recommendation is based on the use of one fibre per direction. Any other optical arrangements may require different specifica- tions and are for further study. 1.1 Abbreviations BER Bit error ratio EX Extinction ratio LED Light-emitting diode MLM Multi-longitudinal mode NA Not applicable NRZ Non-return to zero ORL Optical return loss r.m.s. Root-mean-square SDH Synchronous digital hierarchy SLM Single-longitudinal mode STM Synchronous-transport module UI Unit interval WDM Wavelength-division multiplexing 2 Classification of optical interfaces It is expected that optical fibres will be used in SDH-based systems for both inter-office transport between stations and in intra-office applica- tions for connecting equipment within a single station. By appropriate com- binations of transmitters and receivers, power budgets for optical fibre line systems can be achieved which are optimized in terms of attenuation/disper- sion and cost with respect to the various applications. However, to simplify the development of transverse compatible systems, it is desirable to limit the number of application categories and corresponding sets of optical interface specifications for standardization. As shown in Table 1/G.957, this Recommendation recognizes three broad application categories: – intra-office corresponding to interconnect distances less than approximately 2km; – short-haul inter-office corresponding to interconnect distances of approximately 15km; – long-haul inter-office corresponding to interconnect distances of approximately 40km in the 1310 nm window and approximately 60km in the 1550nm window. Within each category, it is possible to consider use of either nominal 1310 nm sources on optical fibre complying with RecommendationG.652 or nominal 1550 nm sources on optical fibre complying with RecommendationsG.652, G.653 or G.654. This Recommendation covers both possibilities for the two inter-office applications and considers only nominal 1310 nm sources on G.652 fibre for the intra-office application. Since the overall system characteristics and specific values for the optical parameters generally depend on system bit rate, it is convenient to classify the SDH optical interfaces based on applications considered in this Recom- mendation using the set of application codes shown in Table1/G.957. This application code is constructed in the following way: Application – STM level – suffix number with the application designations being I (Intra-office), S (Short-haul), or L (Long-haul), and the suffix number being one of the following: – (blank) or 1 indicating nominal 1310 nm wavelength sources on G.652 fibre; – 2 indicating nominal 1550 nm wavelength sources on G.652 fibre for short-haul applications and either G.652 or G.654 fibre for long-haul applications; – 3 indicating nominal 1550 nm wavelength sources on G.653 fibre. The distances chosen for the application codes in Table 1/G.957 are based on parameter values that are achievable with present technology and which are thought to suit network requirements. The intra-office and short- haul inter-office application codes have been proposed as low-cost equip- ment implementations. The long-haul application codes have been proposed to provide maximum length repeater spans consistent with limits set by present technology and the objective of transverse compatibility. The dis- tances proposed may allow for the upgrading of present systems by exploit- ing the 1550nm region. The distances in Table1/G.957 represent approximate maximum repeater span distances. Specific distance limits consistent with the attenuation limits given in Tables2/G.957 to4/G.957, but including allowances for extra connectors or margins, can be derived through consideration of maximum fibre attenuation and dispersion values for each application in Tables2/G.957 to4/G.957. 3 Parameter definitions For the purpose of this Recommendation, optical fibre line system interfaces can be represented as shown in Figure1/G.957. More specific ref- erence configurations which relate the specifications in this Recommenda- tion to actual optical line systems based on the Synchronous Digital Hierarchy are contained in RecommendationG.958. In Figure1/G.957, point S is a reference point on the optical fibre just after the transmitter opti- cal connector (CTX) and pointR is a reference point on the optical fibre just before the receiver optical connector (CRX). Additional connectors at a dis- tribution frame (if used) are considered to be part of the fibre link and to be located between pointsSandR. In this Recommendation, optical parame- ters are specified for the transmitter at pointS, for the receiver at point R, and for the optical path between pointsS andR. FIGURE 1/G.957 = 7cm All parameter values specified are worst-case values, assumed to be met over the range of standard operating conditions (i.e. temperature and humidity ranges), and they include aging effects. These conditions and effects are for further study. The parameters are specified relative to an opti- cal section design objective of a bit error ratio (BER) not worse than 1´10- 10 for the extreme case of optical path attenuation and dispersion conditions in each application of Table1/G.957. The optical line coding used for all system interfaces is binary non- return to zero (NRZ), scrambled according to RecommendationG.709. 3.1 System operating wavelength range To provide flexibility in implementing transversely compatible sys- tems and future usage of wavelength-division multiplexing (WDM), it is desirable to allow as wide a range as possible for the system operating wavelengths. The choice of operating wavelength range for each of the applications of Table1/G.957 depends on several factors including fibre type, source characteristics, system attenuation range, and dispersion of the optical path. The following general considerations affect the specification of operating wavelength ranges in this Recommendation. More detailed description of the system aspects used to develop the operating wavelength range requirements in this Recommendation is contained in AnnexA. The wavelength regions permitting system operation are partially determined by either the cutoff wavelength values of the fibre or of the fibre cable. For G.652 and G.653 fibres, these values have been chosen to allow single-mode operation of the fibre cable at 1270nm and above, with values as low as1260 permitted by some Administrations. For G.654 fibre cables, the cutoff wavelength values have been proposed for single-mode operation at 1525nm (provisional) and above. The allowable wavelength regions are further defined by the fibre attenuation. Although the intrinsic scattering attenuation generally decreases with increasing wavelength, OH-ion absorption can manifest itself around 1385nm, and to a smaller extent around 1245nm. These absorption peaks and the cutoff wavelength therefore define a wavelength region centered around 1310nm. Dispersion-unshifted fibres complying with RecommendationG.652 are optimized for use in this wavelength region. At longer wavelengths bending attenuation occurs towards 1600nm or beyond, and infra-red absorption occurs beyond 1600nm. These attenua- tions and the 1385nm water peak therefore define a second operating wave- length region around 1550 nm. Recommendation G.654 for loss-optimized fibre is limited to this region only. However, both G.652 and dispersion- shifted G.653 fibres may be used in this region. Apart from cutoff wavelength and attenuation that determine the broad operating wavelength regions, the allowable wavelength ranges are determined by the interaction of the fibre dispersion with the spectral char- acteristics of the transmitter. Parts of this range may lie inside or outside the wavelength range determined by attenuation. The overlap of the two ranges is the permissible wavelength range for system operation. Note–Fibre and cable experts have still to confirm that the wave- length ranges indicated in Tables 2/G.957 to4/G.957 can be achieved with existing RecommendationsG.652, G.653 and G.654. 3.2 Transmitter 3.2.1 Nominal source type Depending on attenuation/dispersion characteristics and hierarchical level of each application in Table1/G.957, feasible transmitter devices include light-emitting diodes (LEDs), multi-longitudinal mode (MLM) lasers and single-longitudinal (SLM) lasers. For each of the applications, this Recommendation indicates a nominal source type. It is understood that the indication of a nominal source type in this Recommendation is not a requirement and that SLM devices can be substituted for any application showing LED or MLM as the nominal source type and MLM devices can be substituted for any application showing LED as the nominal source type without any degradation in system performance. 3.2.2 Spectral width For LEDs and MLM lasers, spectral width is specified by the maxi- mum root-mean-square (RMS) width under standard operating conditions. The RMS width or value is understood to mean the standard deviation (s) of the spectral distribution. The measurement method for RMS widths should take into account modes 20dB to 30dB down from the peak mode and is for further study. For SLM lasers, the maximum spectral width is specified by the max- imum full width of the central wavelength peak, measured 20dB down from the maximum amplitude of the central wavelength under standard operating conditions. Additionally, for control of mode partition noise in SLM systems, a minimum value for the laser side-mode suppression ratio is specified. There is currently no agreed reliable method for estimating the disper- sion penalties arising from laser chirp and finite side-mode suppression ratio for SLM lasers. Because of this, SLM laser linewidths and maximum fibre dispersion values for the L-4.2, S-16.2, L-16.2, and L-16.3 applications are under study. A possible need to specify dynamic laser characteristics more accu- rately is being recognized, particularly for long-haul systems. This includes associated measurement methods. One possible method is a fibre transmis- sion test. Its configuration consists of a transmitter under test, test fibres with maximum dispersion specified for the maximum system length, and a reference receiver. The dynamic characteristics of the transmitter can then be evaluated using a bit error rate measurement. This and alternate methods for characterizing laser dynamic performance are for further study. 3.2.3 Mean launched power The mean launched power at reference point S is the average power of a pseudo-random data sequence coupled into the fibre by the transmitter. It is given as a range to allow for some cost optimization and to cover allow- ances for operation under the standard operating conditions, transmitter con- nector degradations, measurement tolerances, and aging effects. These values allow the calculation of values for the sensitivity and overload point for the receiver at reference pointR. The possibility of obtaining cost-effective system designs for long- haul applications by using uncooled lasers with maximum mean launched powers exceeding those of Tables2/G.957 to 4/G.957, necessitating exter- nal, removable optical attenuators in low-loss sections, is for further study. In the case of fault conditions in the transmit equipment, the launched power and maximum possible exposure time of personnel should be limited for optical fibre/laser safety considerations according to[1]. 3.2.4 Extinction ratio The convention adopted for optical logic level is: – emission of light for a logical “1”, – no emission for a logical “0”. The extinction ratio (EX) is defined as: EX = 10 log10 (A/B) where A is the average optical power level for a logical “1” and B is the average optical power level for a logical “0”. Measurement methods for the extinction ratio are under study. 3.2.5 Eye pattern mask In this Recommendation, general transmitter pulse shape characteris- tics including rise time, fall time, pulse overshoot, pulse undershoot, and ringing, all of which should be controlled to prevent excessive degradation of the receiver sensitivity are specified in the form of a mask of the transmit- ter eye diagram at pointS. For the purpose of an assessment of the transmit signal, it is important to consider not only the eye opening, but also the overshoot and undershoot limitations. The parameters specifying the mask of the transmitter eye diagram are shown in Figure2/G.957. AppendixI considers measurement set-ups for determining the eye diagram of the opti- cal transmit signal. 3.3 Optical path To ensure system performance for each of the applications considered in Table 1/G.957, it is necessary to specify attenuation and dispersion char- acteristics of the optical path between reference pointsS andR. 3.3.1 Attenuation In this Recommendation, attenuation for each application is specified as a range, characteristic of the broad application distances indicated in Table1/G.957. However, to provide flexibility in implementing transverse compatible systems, this Recommendation recognizes some overlap between attenuation ranges between the intra-office applications and the short-haul inter-office applications and between the short-haul inter-office applications and the long-haul inter-office applications. Attenuation specifi- cations are assumed to be worst-case values including losses due to splices, connectors, optical attenuators (if used) or other passive optical devices, and any additional cable margin to cover allowances for: 1) future modifications to the cable configuration (additional splices, increased cable lengths, etc.); 2) fibre cable performance variations due to environmental factors; and 3) degradation of any connector, optical attenuators (if used) or other passive optical devices between points S and R, when provided. 3.3.2 Dispersion Systems considered limited by dispersion have maximum values of dispersion (ps/nm) specified in Tables2/G.957 to 4/G.957. These values are consistent with the maximum optical path penalties specified (i.e. 2dB for L-16.2, 1dB for all other applications). They take into account the specified transmitter type, and the fibre dispersion coefficient over the operating wavelength range. Systems considered limited by attenuation do not have maximum dis- persion values specified and are indicated in Tables2/G.957 to 4/G.957 with the entry NA (not applicable). 3.3.3 Reflections Reflections are caused by refractive index discontinuities along the optical path. If not controlled, they can degrade system performance through their disturbing effect on the operation of the laser or through multi- ple reflections which lead to interferometric noise at the receiver. In this Recommendation, reflections from the optical path are controlled by speci- fying the: – mimimum optical return loss (ORL) of the cable plant at point S, including any connectors, and – maximum discrete reflectance between points S and R. The possible effects of reflections on single fibre operation using directional couplers have not been considered in this Recommendation and are for further study. Measurement methods for reflections are described in Appendix II. For the purpose of reflectance and return loss measurements, points S and R are assumed to coincide with the endface of each connector plug (see Figure1/G.957). It is recognized that this does not include the actual reflec- tion performance of the respective connectors in the operational system. These reflections are assumed to have the nominal value of reflection for the specific type of connectors used. The maximum number of connectors or other discrete reflection points which may be included in the optical path (e.g. for distribution frames, or WDM components) must be such as to allow the specified overall optical return loss to be achieved. If this cannot be done using connectors meeting the maximum discrete reflections cited in Tables2/G.957 to 4/ G.957, then connectors having better reflection performance must be employed. Alternatively, the number of connectors must be reduced. It also may be necessary to limit the number of connectors or to use connectors having improved reflectance performance in order to avoid unacceptable impairments due to multiple reflections. Such effects may be particularly significant in STM-16 and STM-4 long-haul systems. In Tables 2/G.957 to 4/G.957 the value -27 dB maximum discrete reflectance between points S and R is intended to minimize the effects of multiple reflections (e.g interferometric noise). In Tables3/G.957 to 4/ G.957, the value -27dB for maximum receiver reflectance will ensure acceptable penalties due to multiple reflections for all likely system config- urations involving multiple connectors, etc. Systems employing fewer or higher performance connectors produce fewer multiple reflections and con- sequently are able to tolerate receivers exhibiting higher reflectance. As an extreme example, if only two connectors exist in the system, a 14dB receiver return loss is acceptable. For systems in which reflection effects are not considered to limit sys- tem performance, no values are specified for the associated reflection parameters and this is indicated in Tables2/G.957 to 4/G.957 by the entry NA (not applicable). However, when using this Recommendation for a par- ticular application, it should be noted that if upgradability to other applica- tions having more stringent requirements is contemplated, then these more stringent requirements should be used. The possible need to develop a specification for transmitter signal-to- noise ratio under conditions of worst-case optical return loss for the applica- tions in Tables2/G.957 to 4/G.957 is for further study. 3.4 Receiver Proper operation of the system requires specification of minimum receiver sensitivity and minimum overload power level. These are taken to be consistent with the mean launched power range and attenuation range specified for each application. 3.4.1 Receiver sensitivity Receiver sensitivity is defined as the minimum acceptable value of average received power at point R to achieve a 1´10-10 BER. It takes into account power penalties caused by use of a transmitter under standard oper- ating conditions with worst-case values of extinction ratio, pulse rise and fall times, optical return loss at point S, receiver connector degradations and measurement tolerances. The receiver sensitivity does not include power penalties associated with dispersion, jitter, or reflections from the optical path; these effects are specified separately in the allocation of maximum optical path penalty. Aging effects are not specified separately since they are typically a matter between a network provider and an equipment manufac- turer. Typical margins between a beginning-of-life, nominal temperature receiver and its end-of-life, worst-case counterpart are desired to be in the2 to 4dB range. An example of a measurement method for determining aging effects on receiver sensitivity is given in AppendixIII. 3.4.2 Receiver overload Receiver overload is the maximum acceptable value of the received average power at point R for a 1´10-10 BER. It should be noted that the use of an optical attenuator in front of the receiver may be required to avoid overloading the receiver. 3.4.3 Receiver reflectance Reflections from the receiver back to the cable plant are specified by the maximum permissible reflectance of the receiver measured at reference pointR. 3.4.4 Optical path power penalty The receiver is required to tolerate an optical path penalty not exceed- ing 1dB (2 dB for L-16.2) to account for total degradations due to reflec- tions, intersymbol interference, mode partition noise, and laser chirp. 4 Optical parameter values for SDH applications Optical parameter values for the applications of Table1/G.957 are given in Table2/G.957 for STM-1, Table3/G.957 for STM-4, and Table4/ G.957 for STM-16. Parameters defining the mask of the transmitter eye dia- gram at reference point S for each of the three hierarchical levels are given in Figure2/G.957. These tables do not preclude the use of systems which satisfy the requirements of more than one application for any given bit rate. include 957-t02e à l'italienne include 957-t03e à l'italienne include 957-t04e à l'italienne 5 Optical engineering approach The selection of applications and set of optical parameters covered by this Recommendation are chosen to reflect a balance between economic and technical considerations to provide the possibility for transverse compatible systems using the synchronous digital hierarchy. This section describes the use of the parameters in Tables2/G.957 to 4/G.957 to obtain a common sys- tem design approach for engineering SDH optical links. 5.1 Design assumptions To meet the greatest number of application possibilities with the smallest number of optical interface component specifications, three inter- face categories are assumed for each level of the SDH hierarchy. These are distinguished by different attenuation/dispersion regimes rather than by explicit distance constraints to provide greater flexibility in network design while acknowledging technology and cost constraints for the various appli- cations. Worst-case parameter values are specified in this Recommendation to provide simple design guidelines for network planners and explicit compo- nent specifications for manufacturers. As a result, neither unallocated sys- tem margins nor equipment margins are specified and it is assumed that transmitters, receivers, and cable plant individually meet the specifications under the standard operating conditions. It is recognized that, in some cases, this may lead to more conservative system designs than could be obtained through joint engineering of the optical link, the use of statistical design approaches, or in applications and environments more constrained than those permitted under the standard operating conditions. 5.2 Worst-case design approach For a worst-case design approach, the optical parameters of Tables2/ G.957 to 4/G.957 are related as shown in Figure 3/G.957. In loss-limited applications, a system integrator may determine the appropriate application code and corresponding set of optical parameters by first fixing the total optical path attenuation, which should include all sources of optical power loss and any cable design margin specified by the system integrator. For those situations in which the system attenuation falls within the attenuation overlap region of two applications, then either set of optical parameters would apply. The most economical designs will generally correspond to the application code having the narrower attenuation range. For each installa- tion, it should be verified that the total optical path penalty, which includes combined dispersion and reflection degradations, does not exceed the value given in §3.4.4 and Tables2/G.957 to 4/G.957 since a higher value may lead to rapidly deteriorating system performance. For dispersion-limited systems, the system integrator may select an appropriate application code and corresponding set of optical parameters by determining the total dispersion (ps/nm) expected for the elementary cable section to be designed. The most economical design generally corresponds to the selection of the application having the smallest maximum dispersion value exceeding the dispersion value determined for the system design. Again, the total optical path power penalty should be verified as described above. FIGURE 2/G.957 = 11cm FIGURE 3/G.957 = 10cm 5.3 Statistical design approach The statistical approach is based on designing an enhanced elemen- tary cable section, possibly exceeding the section length obtained by a worst-case design. By admitting a certain probability that the attenuation or dispersion between points S and R is larger than specified system values or that a transverse compatible design may not be obtained, cost savings may be achieved in long-haul high bit-rate optical systems through the reduction of the number of repeaters. When using the statistical approach, the sub-system parameters are expressed in terms of the statistical distributions, which are assumed to be available from the manufacturers. Such distributions can be handled either numerically (e.g. by Monte Carlo methods) or analytically (e.g. Gaussian averages and standard deviations). Examples of parameters which can be considered statistical in nature are the following: – cable attenuation; – cable zero-dispersion wavelength and zero-dispersion slope; – splice and connector loss; – transmitter spectral characteristics (central wavelength, spectral width, etc.); – available system gain between points S and R (e.g. optical power available at point S and receiver sensitivity at point R. These parameters may need to be considered separately for transverse compatibility considerations). According to design practices, each of the above parameters can be considered either statistical or worst-case. In a semi-statistical approach, those parameters assumed deterministic may be given a zero-width distribu- tion around the worst-case value. Details are given in RecommendationsG.955 and G.956. 5.4 Upgradability considerations Two possibilities arise with regard to system upgradability: 1) It may be desired to upgrade from existing plesiochronous systems to SDH systems (e.g. from a 139264kbit/s system complying with RecommendationG.956 specifications to an STM-1 system based on this Recommendation); 2) It may be desired to upgrade from one SDH hierarchical level to another (e.g. from STM-1 to STM-4). It is not always feasible to satisfy both possibilities simultaneously for long-haul applications, and opinions differ on the best approach to be taken for system upgrade. For example, to maintain compatibility with 139264kbit/s and 4´139264kbit/s systems complying with RecommendationG.956, maximum attenuation values for STM-1 and STM-4 long-haul applications in this Recommendation are taken to be 28dB and 24dB, respectively. The difference in maximum attenuation for these two levels reflects the current wide-scale availability of STM-4 receivers meeting the sensitivity requirements of the lower attenuation value compared to the current relatively high cost of STM-4 receivers meeting the sensitivity requirements of the higher attenuation value. Two examples for accomplishing upgradability are described in Appendix IV. Also, Recommenda- tionG.958, §4.3, addresses the issue of joint engineering to meet not only the upgradability requirements, but any instances where the interface specifications of RecommendationG.957 are not sufficient to meet the requirements of the specific application. ANNEX A (to Recommendation G.957) System operating wavelength considerations This annex provides further information on the choice of range of operating wavelengths specified in Tables2/G.957 to 4/G.957. A.1. Operating wavelength ranges determined by fibre attenuation The general form of attenuation coefficient for installed fibre cable used in this Recommendation is shown in FigureA-1/G.957. Included here are losses due to installation splices, repair splices, and the operating tem- perature range. RecommendationG.652 states that attenuation values in the range 0.3-0.4dB/km in the 1310nm region and 0.15-0.25dB/km in the 1550nm region have been obtained. The variation of attenuation coefficient with wavelength and with temperature, and the losses due to splices are for fur- ther study. A.2 Operating wavelength ranges determined by fibre dispersion For G.652 fibres, the zero-dispersion wavelength lies between 1300 nm and 1322 nm, so the fibre is dispersion-optimized in the 1310nm region. These wavelengths and corresponding requirements on the zero-dis- persion slope result in the maximum permitted absolute value of the disper- sion coefficient (as determined by fibres having the minimum or maximum zero-dispersion wavelengths) shown in FigureA-2a)/G.957. However, the G.652 fibres can be used also in the 1550 nm region, for which the maxi- mum dispersion coefficient is comparatively large as shown in FigureA- 2b)/G.957. For G.653 fibre, the permitted range of the zero-dispersion wave- length lies between 1500nm and 1600nm, so the fibre is dispersion-opti- mized in the 1550nm region. The analytical expressions for the dispersion coefficient result in the maximum permitted values shown in FigureA-3/ G.957. The G.653 fibres can be used also in the 1310nm region, for which the maximum dispersion coefficient is comparatively large. However, this possible application is currently not considered in RecommendationG.957. For G.654 fibres in the 1550 nm region, the dispersion coefficient is similar but slightly larger than that for G.652 fibres. This is still under study and has not been taken into account in Tables2/G.957 to 4/G.957. For G.652 fibres in the 1310 nm region and for G.653 fibres in the 1550nm region, the dispersion-limited wavelength range is chosen such that the absolute values of the dispersion coefficient at the limiting wave- lengths are approximately equal. As can be seen from the shapes of FigureA-2a)/G.957 and FigureA-3/G.957, absolute dispersion values are therefore smaller within the operating wavelength range. For G.654 fibres, and also for G.652 fibres in the 1550 nm region, FigureA-2b)/G.957 shows that dispersion limits the upper operating wave- length while attenuation limits the lower operating wavelength. The interaction between the transmitter and the fibre is accounted for by a parameter epsilon. It is defined as the product of 10-6 times the bit rate (in Mbit/s) times the path dispersion (in ps/nm) times the RMS spectral width (innm). For a 1 dB power penalty due to dispersion, epsilon has a maximum value. For intersymbol interference alone, the value 0.306 is applied to LEDs and SLM lasers. The 20dB width for SLM lasers is taken as 6.07times the RMS width. (For L-16.2 only, it is necessary to increase epsilon to 0.491, corresponding to a 2dB power penalty.) For intersymbol interference plus mode partition noise, the maximum value 0.115 is applied to MLM lasers. (For I-1 and I-4, the large spectral widths may not often occur, but they are retained here for possible cost savings.) For wavelength chirp, no known value is applied to SLM lasers. For a particular spectral width, the optical path dispersion is fixed for a particular application code. With the appropriate path distance from Table1/G.957, the maximum allowed dispersion coefficient follows. The spectral dependence of the dispersion coefficient then determines the disper- sion-limited wavelength range. (The use of the dispersion coefficient beyond the wavelength ranges stated in RecommendationsG.652, G.653 or G.654 is for further study.) FIGURE A-1/G.957 = 11.5cm Ranges A and B are suitable for long-haul (L-N.x) applications, and ranges C and D are suitable for short-haul (S-N.x) and intra-office (I-N) applica- tions. FIGURE A-2/G.957 = 10cm FIGURE A-3/G.957 = 11 APPENDIX I (to Recommendation G.957) Measurement of the mask of the eye diagram of the optical transmit signal I.1 Measurement set-up In order to ensure the suitability of the optical transmit signal for the performance of the receiver, a measurement set-up according to FigureI-1/ G.957 is recommended for the eye diagram of the transmit optical signal. An optical attenuator may be used for level adaptation at the reference point OI. An electrical amplifier may be used for level adaptation at the reference point EO. Values for the mask of the eye diagram in Figure2/G.957 include measuring errors such as sampling oscilloscope noise and manufacturing deviations of the low-pass filter. I.2 Transfer function of the optical reference receiver The nominal transfer function of the optical reference receiver is char- acterized by a fourth-order Bessel-Thomson response according to: with The reference frequency is fr = 0.75 f0. The nominal attenuation at this frequency is 3 dB. The corresponding attenuation and group delay dis- tortion at various frequencies are given in TableI-1/G.957. FigureI-2/G.957 shows a simplified circuit diagram for the low-pass filter used for measuring the mask of the eye diagram of the optical transmit signal. Note–This filter is not intended to represent the noise filter used within an optical receiver. In order to allow for tolerances of the optical reference receiver components including the low-pass filter, the actual attenuation should not deviate from the nominal attenuation by more than the values specified in TableI-2/ G.957. The flatness of the group delay should be checked in the frequency band below the reference frequency. The tolerable deviation is for further study. FIGURE I-1/G.957 = 8cm FIGURE I-2/G.957 = 6cm APPENDIX II (to Recommendation G.957) Methods for measuring reflections Two methods are in general use. The optical continuous-wave reflec- tometer (OCWR) utilizes a continuous or modulated stable light source with a high sensitivity time-averaging optical power meter. It is suitable for measuring the optical return loss of the cable plant at point S or the reflectance of the receiver at point R. The optical time- domain reflectometer (OTDR) utilizes a pulsed source having a low duty cycle along with a sensitive time-resolving optical receiver. It is suitable for measuring discrete reflectances between S and R or the receiver reflectance at R. Both instruments utilize 2 ´ 1 optical couplers, and both are available commercially. Instructions contained with the instrument may super- sede those given below. Moreover, test procedures are under develop- ment. For calibration purposes, a jumper with a known end reflector may be used. The value of reflectance may be near zero (as obtained with careful index matching and/or a tight bend in the fibre), or about -14.5 dB (as with a good cleave), or some other known reflectance R0 (as with an imperfect cleave or an applied thin film coating). The connection between the jumper and the instrument must have a low reflectance. II.1 Optical continuous-wave reflectometer The coupler nomenclature is shown in Figure II-1/G.957, and the fol- lowing calibration measurement needs to be performed only once. Power Ps is measured by connecting the optical source directly to the power meter. The source is then connected to output port 3 of the coupler, while the power meter measures P32 at the input port2. The source is now connected to input port1, while the meter measures power P13 at port3. Finally, the non-reflecting jumper is connected to port3, while power P0 is measured at port2. To measure the reflectance of the detector, the connector at point R is connected to port 3; to measure the ORL of the cable plant, the connector at pointS is connected to port3. In either case, power PR is measured by the meter at port2. The reflectance of the detector is: The ORL of the cable plant is: ORL = -R. II.2 Optical time-domain reflectometer Here the coupler is usually internal to the instrument. A variable opti- cal attenuator, and a pigtail of length beyond the dead-zone of the instru- ment are both supplied, if they are not already internal to the instrument. The following calibration measurement needs to be performed only once. A jumper with known reflectance R0 is attached, giving an OTDR trace sche- matically shown in FigureII-2/G.957. The optical attenuator is adjusted until the reflection peak falls just below the instrumental saturation level, and the peak height H0 is noted. The calibration factor: is calculated. (If the temporal duration D of the pulse is measured, the back- scatter coefficient of the fibre is B=F-10log10 D. If D is in ns, B is about -80dB.) To measure the maximum discrete reflectance between S and R, the OTDR is connected to point S or R. The peak height H is noted for a particular reflectance. The resulting value is: FIGURE II-1/G.957 = 6cm FIGURE II-2/G.957 = 8.5cm APPENDIX III (to Recommendation G.957) Possible method for evaluating aging margin contribution in receiver sensitivity specifications This appendix presents a possible method for determining the contri- bution due to aging effects in the specification of receiver sensitivity used in RecommendationG.957. III.1 Receiver sensitivity and eye opening Figure III-1/G.957 shows eye opening at the receiver as a function of optical received power. The eye opening value, E, is the value which is determined by the system designer for operation at a BER of 10-10. The received power P2 corresponds to the power required for maximum eye opening at the receiver. For stable system operation, the optical received power is typically set to a level higher than P1 such that, at the end of sys- tem life, the specified eye opening, E, is still satisfied. Thus, P1 is the end- of-life receiver sensitivity and P0 is the beginning-of-life receiver sensitiv- ity. M is the margin between P1 and P0 to account for the effects of receiver aging. The amount of eye margin depends on receiver characteristics and the values, for example, may be E1-E and E2-E for different receivers (e.g.typeI or typeII). An appropriate eye margin cannot be obtained if the received power is P0. With respect to the effects of aging on receiver performance, it may be assumed that the eye opening as a function of received optical power is shifted parallel to the initial characteristics as shown in FigureIII-2/G.957. For the purposes of simulating aging effects, it may also be assumed that the shifted curve can be obtained by adding a certain amount of intersymbol interference noise to the signal corresponding to the initial value of eye mar- gin. The test method proposed for evaluation of the eye opening by this technique is the S/X test. III.2 S/X test method To simulate intersymbol interference noise, the S/X test is performed by using an NRZ signal modulated at a low frequency compared to the sys- tem operating bit rate. This interfering signal is combined optically with a normal optical signal and injected into the receiver under test. In the S/X test, the normal optical signal power is usually set to P1. The amount of the optical power of the interference noise, X, can be deter- mined by a relationship between eye opening and S/X ratio whose charac- teristics are shown in FigureIII-3/G.957. From FigureIII-3/G.957, the S/X ratio can be determined as (S/X)E by the relationship between E1 and E. The aging margin M and (S/X)E are given by: The test configuration is shown in Figure III-4/G.957. FIGURE III-1/G.957 = 9cm FIGURE III-2/G.957 = 9cm FIGURE III-3/G.957 = 9.5cm FIGURE III-4/G.957 = 21.5cm APPENDIX IV (to Recommendation G.957) Upgradability examples Two examples for accomplishing upgradability are described below: Example 1 To realize low-cost designs optimized for a particular hierarchical level by using current, widely available optical components, the following maximum attenuation ranges may be adopted for the long-haul applications: – STM-1 28 dB – STM-4 24 dB – STM-16 20 dB. For upgrading from one hierarchical level to a higher one when it is desired to maintain regenerator spacings for the original and upgraded sys- tem, the following options are available: i) The original system design may be based on the smallest attenuation (i.e. highest hierarchical level) expected for the upgraded long- haul system. ii) If the original system operates in the 1310 nm region on G.652 fibre, then the upgraded system may be chosen to operate in the 1550nm region to obtain lower cable attenuation, although with increased dispersion penalty. iii) Relatively high-loss components (e.g. connectors) may be replaced with lower-loss components for the upgraded system. iv) Statistical design approaches may be employed to provide enhanced cable sections for the upgraded system. Example 2 Another approach to upgradability is to employ the concept of a set of grades in higher order STM-N systems for the long-haul inter-office inter- faces. TableIV-1/G.957 and FigureIV-1/G.957 show the grade classifica- tion based on maximum attenuation. Parameter values for the various grades are for further study. These grades might be applied by users when consider- ing network planning and cost performance, etc. Moreover, higher grade system design should allow incorporation of future technology advances and changing service requirements. FIGURE IV-1/G.957 = 10cm Reference [1] IEC 825 Radiation safety of laser products, equipment classification, requirements and user's guide, 1984.