1、TIA/EIA TELECOMMUNICATIONS SYSTEMS BULLETIN ITM-20 Enhanced Bandwidth Performance over Laser-Based, Multimode Fiber Local Area Networks TSB62-20 FEBRUARY 2001 TELECOMMUNICATIONS INDUSTRY ASSOCIATION The Telecommunications Industry Association represenis the communications sector of Eisctrtlnis Irrau
2、stries Alliance Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-NOTICE TIMEIA Engineering Standards and Publications are designed to serve the public interest through eliminatin
3、g misunderstandings between manufacturers and purchasers, facilitating interchangeability and improvement of products, and assisting the purchaser in selecting and obtaining with minimum delay the proper product for his particular need. Existence of such Standards and Publications shall not in any r
4、espect preclude any member or nonmember of TIMEIA from manufacturing or selling products not conforming to such Standards and Publications, nor shall the existence of such Standards and Publications preclude their voluntary use by those other than TIMEIA members, whether the standard is to be used e
5、ither domestically or internationally. Standards, Publications and Bulletins are adopted by EIA in accordance with the American National Standards Institute (ANSI) patent policy. By such action, TIA/EIA does not assume any liability to any patent owner, nor does it assume any obligation whatever to
6、parties adopting the Standard, Publication, or Bulletin. Technical Bulletins are distinguished from TIMEIA Standards or Interim Standards, in that they contain a compilation of engineering data or information useful to the technical community, and represent approaches to good engineering practices t
7、hat are suggested by the formulating committee. This Bulletin is not intended to preclude or discourage other approaches that similarly represent good engineering practice, or that may be acceptable to, or have been accepted by, appropriate bodies. Parties who wish to bring other approaches to the a
8、ttention of the formulating committee to be considered for inclusion in future revisions of this Bulletin are encouraged to do so. It is the intention of the formulating committee to revise and update this Bulletin from time to time as may be occasioned by changes in technology, industry practice, o
9、r government regulations, or for other appropriate reasons. (From Project No. 4892, formulated under the cognizance of the TIA FO-2.2 Subcommittee on Digital Multimode Systems. ) Published by TELECOMMUNICATIONS INDUSTRY ASSOCIATION 2001 Standards and Technology Department 2500 Wilson Boulevard Arlin
10、gton, VA 22201 PRICE: Please refer to the current Catalog of ELECTRONIC INDUSTRIES ALLIANCE STANDARDS and ENGINEERING PUBLICATIONS or call Global Engineering Documents, USA and Canada (1-800-854-7179) International (303-397-7956) All rights reserved Printed in U.S.A. Copyright Telecommunications Ind
11、ustry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license fro
12、m IHS-,-,-Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 ITM-20 Enhanced Bandwidth Performance over Laser.Based. Multimode Fiber Local Area Networks Contents . Forewo
13、rd III 1 Introduction 1 2 Background and History . 2 2.1 Launch Conditioning 2 2.2 Source Near Field 2 2.2.1 Mode Power Distribution 2 2.2.2 Encircled Flux . 3 2.3 Encircled Flux Round Robins . 4 2.4 Encircled Flux and Link Performance 5 2.5 The Validation Experiment . 6 2.6 Restricted Mode Launch F
14、iber Bandwidth . 8 2.7 Enhanced Bandwidth Criteria . 9 2.8 Risk Assessment 10 3 Recommendation . 11 4 Normative references . 12 Annex A (informative) 13 Annex B (informative) 19 i Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or
15、 networking permitted without license from IHS-,-,-TSB-62-20 This page left blank. ii Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 ITM-20 Enhanced Bandwidth Perform
16、ance over Laser-Based, Multimode Fiber Local Area Networks Foreword From TIA Project No. 4892 formulated under the cognizance of TIA FO-2.2.1, Subcommittee on Modal Dependence of Bandwidth. This bulletin is part of the series of test methods included within TINEIA TSB-62. There are two annexes, whic
17、h are informative. Key words: bandwidth, effective modal bandwidth, encircled flux, launched power distribution, restricted mode launch, restricted mode launch bandwidth iii Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or netwo
18、rking permitted without license from IHS-,-,-TSB-62-20 This page left blank. iv Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 1 Introduction 1.1 Intent This bulletin
19、 describes and gives background information for the laser source and fiber selection criteria required to achieve enhanced bandwidth performance over local area networks (LANs) based on multimode fiber links and laser sources. Such a link is shown in Figure 1. Figure 1 - Basic link addressed by this
20、 bulletin. These criteria come from developments in the TIA Working Group on the Modal Dependence of Bandwidth (FO-2.2.1) (the Working Group) from 1997 to 2000. Without appropriate source and fiber selection criteria, fiber link lengths for LANs must be set very conservatively to minimize risk of li
21、nk failure. The Working Group has developed and tested new test procedures that characterize the laser source near field and the multimode fibers bandwidth performance under restricted launch conditions. Limiting the allowable range of these source and fiber parameters assures that a link will meet
22、enhanced bandwidth performance conditions with an acceptable risk of link failure (below 1%) for maximum link lengths. Shorter links will have lower risks of link failure. 1.2 Scope Although the methodology described here to characterize both sources and fiber may be broadly applicable to different
23、source and fiber types, the specific selection criteria addressed in this bulletin are only applicable to LANs based on short-wavelength (830 - 860 nm) laser sources and graded-index 62.5/125 m optical fibers. Short-wavelength networks based on graded-index 50/125 m fiber alread perform at gigabit-p
24、er-second data rates over link lengths of up to 550 meters , criteria for enhanced network performance is therefore less urgent for these networks. Nevertheless, enhanced network performance for 50/125 m fiber systems (e.g. 1 O gigabit-per-second systems) is currently (as of September 2000) under co
25、nsideration in the Working Group. Y. 1 Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 2 Background and History The move toward high-speed (2 1 gigabit-per-second) LAN
26、s requires the greater modulation capabilities of laser sources like the Vertical Cavity Surface Emitting Laser (VCSEL) or the edge-emitting semiconductor laser diode. Unlike the less coherent light emitting diode (LED), laser sources typically excite only a subset of the fiber modes available. This
27、 change in source property together with the inter- modal dispersion properties of the fiber requires that both source and multimode fiber be characterized and/or modified especially for high-speed LAN applications. 2.1 Launch Conditioning The Working Group began looking into short-wavelength (-850
28、nm) laser source characterization in the spring of 1998. Characterization of the multimode fiber would follow later that same year. This work followed work that the Group performed in association with the I EEE 802.3 Gigabit Ethernet standardization effort. There, it was found that laser launch cond
29、itioning led to more predictable worst-case bandwidth performance. In particular, an offset single-mode fiber launch into multimode fiber employed with 1300 nm laser sources and implemented with a special hybrid patchcord gave a bandwidth performance that predictably met or exceeded the overfilled l
30、aunch (OFL) performance specification of the 62.5 pm core multimode fibeQ. This result allowed the adoption of an acceptably long link length of 550 meters for the Gigabit Ethernet 1000BASE-LX standard. No such launch conditioning was prescribed for the short-wavelength (-850 nm) 1 000BASE-SX Gigabi
31、t Ethernet standard, and the resulting link length for 62.5 m fiber was less than 300 m. The Working Group realized that conditioning the launch of the short-wavelength sources might lead to significant improvement in allowable link length, particularly for 62.5 m fiber. Unlike the offset launch sol
32、ution employed for the 1300 nm sources, the solution for the short-wavelength sources would ideally involve no external patchcord. This was desired to preserve the low-cost potential of the short-wavelength sources. A simple restriction of allowable source launch characteristics might yield the desi
33、red performance enhancement. The Working Group began by looking at the near-field properties of the laser source. 2.2 Source Near Field 2.2.1 Mode Power Distribution Initially, the Working Group attempted to characterize the near field of the source by applying a measurement of the mode power distri
34、bution produced by the laser source in a length of multimode fiber. This measurement is described in EIA/TIA ITM-3. It quickly became apparent, however, that such an approach was not 2 Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproducti
35、on or networking permitted without license from IHS-,-,-TSB-62-20 satisfactory. ITM-3 assumes that all modes within a mode group have the same energy. This equipartion-of-energy assumption does not generally hold for multimode fibers under 1 km in length. Energy redistribution amongst modes within a
36、 mode group occurs over longer fiber lengths, if it occurs at all. Since lasers initially excite a fraction of the modes excited by an LED, this failure of energy redistribution becomes particularly important for laser source characterization. The near-field intensity as a function of radius was oft
37、en observed to be non-symmetrical and non-monotonic. These near-field patterns produced non-physical computed mode power distributions, i.e. negative mode- group powers. 2.2.2 Encircled Flux The Working Group eventually settled on another measure of the laser near field, the Encircled Flux (EF). Enc
38、ircled Flux is obtained from the near-field intensity pattern produced by a laser source in a 10 m length of 62.5 m core, graded-index multimode fiber. It is the near-field pattern at the end of this fiber (as opposed to the near field at the laser source itself) that is more relevant to system perf
39、ormance. This is because the only relevant part of the laser source energy is that which is coupled into the fiber. In general, such a near field will have a non-uniform shape that includes speckle. By integrating over radius and over all azimuthal angles, the Working Group collapsed the two-dimensi
40、onal near-field information into a one-dimensional radial function, giving the fractional power launched by a source within a given fiber radius. The complexity in the raw near-field data is avoided by exploiting the axial symmetry of the fiber, and, unlike the measurement of mode power distribution
41、, no assumption on the equipartition of energy within mode groups is required. Typical near-field patterns for an LED and a VCSEL source at the end of a IO-meter, graded-index, 62.5 m fiber jumper are shown in Figure 2. Note that while the near-field intensity and annular flux for the VCSEL are irre
42、gular, the encircled flux as a function of radius is smooth and monotonically increasing. Since it is the launch power distribution and not the absolute power that is of interest, the normalized encircled flux is plotted and compared. The detailed procedure for obtaining encircled flux is defined in
43、 a test procedure, ElMIA FOTP-203. 3 Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 Radius (Ikm) Figure 2 - Near Fields, Intensity Profiles, Annular Flux, and Encircl
44、ed Flux for an LED and a VCSEL 2.3 Encircled-Flux Round Robins To be a useful parameter, Encircled Flux needs to correlate with link performance and be reproducibly measured from lab to lab. The latter was tested with inter-laboratory round robins. The first round-robin comparison involved 17 laser
45、sources, 9 VCSELS and 8 edge-emitting (CD) lasers and two LEDs. The range of encircled flux encountered in this comparison is shown in Figure 3, where the shaded boxes define the source selection criteria of Table I below. Curves for sources that satisfy the criteria lie between the shaded boxes. 4
46、Copyright Telecommunications Industry Association Provided by IHS under license with EIANot for ResaleNo reproduction or networking permitted without license from IHS-,-,-TSB-62-20 1 0.9 x 0.8 G 3 0.7 al 2 0.6 o S w 0.5 - .- m It) 0.4 m E 0.3 O .- - = 0.2 0.1 O O 5 10 15 20 25 30 35 Radius of Encirc
47、ling Aperture (Pm) Figure 3 - Range of encircled flux for 16 transceivers and 2 LEDs. The solid lines represent the laser sources; the gray lines (two rightmost curves) represent the LEDs. The encircled flux at 15 m radius ranged from 0.5 to 0.97, indicating that a broad range of sources was tested.
48、 The pooled standard deviation for encircled flux at 15 m was 10% across the 17 sources. Subsequent round-robin comparisons showed inter-laboratory agreement of 7.3% and a pooled standard deviation across the five participating labs of 3.6%. This level of inter-laboratory agreement or reproducibilit
49、y in encircled flux results later proved to be sufficient in predicting enhanced link performance. 2.4 Encircled Flux and Link Performance Thirteen of the above EF round-robin sources were used in a bandwidth performance study with 21 randomly selected multimode fibers. The objective was to find practical limits on encircled flux that would yield enhanced bandwidth performa
