Modeling of telecommunication cables for gigabit DSL application

INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE Modeling of Telecommunication Cables for Gigabit DSL Application Marek NEVOSAD, Pavel LAFATA, Petr JARES Department of Telecommunication Engineering, Faculty of Electrical Engineering, Czech Technical University in Prague, Technicka 2, 166 36 Prague, Czech Republic marek.nev[email protected], pav[email protected], [email protected] Abstract. The first part of our paper brings the description and analysis of method used for modeling of metallic cables’ parameters according to the modified KPN model. Moreover we were able to perform measurements and estimations for real metallic cables with various transmission characteristics and constructional arrangement, for which we derived the necessary parameters of a new modeling method. Following part contains the comparison of measured characteristics with their models according to the modified KPN model. Finally, we discussed obtained results. Keywords G-FAST, gigabit digital subscriber line, modeling, telecommunication cables, xDSL. 1. Introduction Today, the existing metallic lines are still being widely used due to the slower development of passive optical networks in many countries. The continuously increasing demands for higher transmission speeds lead to exploit the potentials of metallic lines, which were not considered before. Until recently, the VDSL2 (Very High Speed Digital Subscriber Line) technology, which is standardized for frequencies up to 30 MHz and transfer speeds up to 100 Mbit·s−1, has been considered as the fastest solution in data transmissions over metallic lines [1]. Many experiments tried to increase transmission performance of DSL (Digital Subscriber Line) technology; however, the main limitation is a crosstalk, especially far-end crosstalk (FEXT) [2]. The newly standardized G-fast system, which is currently being developed, will be based on vectored discrete multitone modulation (DMT) to eliminate FEXT and its frequency band will be extended up to 212 MHz. These enhancements will enable reaching gigabit transmission speed. For that purpose, it is necessary to modify existing models and parameters of lines for such high frequencies or design completely new modeling methods. The current version of the G-FAST draft uses a new method of modeling that is not based on calculations of the primary parameters, but instead of it uses the modeling of longitudinal impedance Zsand transverse admittance Ypof a homogenous line. This type of modeling is conceptually based on the method specified in the recommendation ETSI TS 101 270-1, the number of parameters was reduced to 10 and additionally, several modifications were made. Furthermore, the draft also contains specific values of these parameters for four basic types of metallic lines frequently used in access networks. The combined network architecture with the twisted pairs and the optical fibers, called FTTB or FTTC (Fiber to the Building, Fiber to the Curb), should be considered, because the broadband transmission can be used only for short lines. The designation “Fiber Distribution Point” (FTTdp) was introduced for these situations [6]. 2. Theory Instead of using standard primary parameters (R,L,C and G), the models presented within newly developed G-FAST draft are based on estimations of longitudinal impedance Zsand transverse admittance Ypof a homogenous line [4]. An empirical model, which is discussed below, can be used for modeling frequency characteristics up to 300 MHz. The model can be obtained by Eq. (1), Eq. (2), Eq. (3), Eq. (4), Eq. (5), Eq. (6) and Eq. (7) [4] and is conceptually based on the modified equations presented as KPN model [5]. Resistivity is considered as complex value in this model. It is a main difference from other reference models. c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 336 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE Zs(jω) = jω ·Ls∞+Rs0·(1 −qs·qx+v u u u u u t q2 s·q2 x+ 2 ·jω ωs · q2 s+jω ωs ·qy q2 s qx +jω ωs ·qy!,(1) Yp(jω) = jω ·Cp0·(1 −qc)·1 + jω ωd−2·φ/π +jω ·Cp0·q+c, (2) Ls∞=1 ηV F ·c0 ·Z0∞,(3) Cp0=1 ηV F ·c0 ·1 Z0∞ ,(4) qs=1 q2 H·qL ,(5) ωs=q2 H·ωs0=q2 H·4π·Rs0 µ0,(6) ωd= 2π·fd,(7) where c0= 3·108m·s−1and µ0= 4π·10−7H·m−1. This model can be usually simplify by qc= 0 and fd= 1. After this simplification, the transverse admittance Yp is showed in Eq. (8). Yp=j2·π·f·Cp0·(1 + j·f)−2·φ/π.(8) The estimation of characteristic impedance Z0∞of a line is also necessary for appropriate application of presented models, as well as the velocity of propagation ηV F , which can be obtained by using TDR (TimeDomain Reflectometer) method or direct calculations from the phase characteristics. The formulas for the series impedance ZsEq. (11) and parallel admittance YpEq. (12) can be derived from the equations for the secondary parameters Eq. (9), Eq. (10) of symmetrical lines. γ=pZs·Yp,(9) ZC=sZs Yp ,(10) Zs=pγ·Zc,(11) Yp=γ Zc .(12) Parameters φ,qH,qL,qx,qy,Rs0,Z0∞,ηV F for reference model are obtained by measurements and approximation of characteristics. Finally, attenuation factor αis a real part of γas shown in Eq. (13). γ=α+jβ. (13) 2.1. The Method of Measuring the Transmission Parameters The measurement workspace in the Department of Telecommunication Engineering, Faculty of Electrical Engineering, Czech Technical University in Prague had to be upgraded and optimized for measurement of gigabit digital subscriber lines. Figure 1 illustrates the basic schematic of measurements. The measurements were performed by network analyzer Rohde&Schwarz ZVRE (spectral analyzer with vector signal analyzer option) with balun transformers North Hills. The Automated Measuring Workplace (AMW) [3] was also used during our experiments. Fig. 1: The schematic illustration of measurements. The measurements were performed for cable TCEPKPFLE in the frequency band from 5 MHz up to 300 MHz due to the equipment limitations. The rest of the cables, measured by AMW, the frequency band was set from 100 kHz to 250 MHz. c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 337 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE 3. The Results of Performed Measurements and Models The measurements were performed for four typical communication cables: •TCEPKPFLE 75×4×0,4 (one sub-group (8 randomly selected symmetrical pairs)), •SYKFY 4 ×2×0,5, •UTP CAT5e (Unshielded Twisted Pair), •UTP CAT6. 3.1. The Results of Measurements The results of measuring and modeling of attenuation factor and characteristic impedance are given in following figures. 1) TCEPKPFLE 75 ×4×0,4 The cable has star-quad construction with polyethylene insulation and is standard for use in telecommunication. The measuring and modeling of attenuation factor and characteristic impedance are shown in Fig. 2 and Fig. 3. Fig. 2: Attenuation factor αTCEPKPFLE 75 ×4×0,4 and model. 2) SYKFY 4×2×0,5 This type of interior cable with four symmetrical pairs with PVC insulation corresponds to CAT3 (Category 3) for the frequency band up to a few MHz. Its frequency characteristics are presented in Fig. 4 and Fig. 5. Fig. 3: Module of characteristic impedance ZcTCEPKPFLE 75 ×4×0,4. Fig. 4: Attenuation factor αSYKFY 4 ×2×0,5 and model. Fig. 5: Module of characteristic impedance ZcSYKFY 4 ×2× 0,5. 3) UTP CAT5e The CAT5e is a category of cables primary used for local area networks with parameters guaranteed up to c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 338 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE 100 MHz. The measuring and modeling of attenuation factor and characteristic impedance are shown in Fig. 6 and Fig. 7. Fig. 6: Attenuation factor αUTP CAT5e and model. Fig. 7: Module of characteristic impedance ZcUTP CAT5e. 4) UTP CAT6 The CAT6 is a category of cables with the same type of construction as CAT5e, furthermore, in this category the parameters are guaranteed up to 250 MHz. Its frequency characteristics are presented in Fig. 8 and Fig. 9. 3.2. Parameters of the Reference Model The measured values were processed in program Matlab and optimized parameters were estimated. The parameters obtained for reference model (above) are presented in Tab. 1 and Tab. 2. Output parameters of the reference model are presented in Tab. 3. The velocity of propagation ηV F was Fig. 8: Attenuation factor αUTP CAT6 and model. Fig. 9: Module of characteristic impedance ZcUTP CAT6. Tab. 1: q-parameters of the reference model. Cable type qHqLqXqY TCEPKPFLE 75 ×4×0,4 0,902 2,15 0,5 0,723 SYKFY 4 ×2×0,5 0,501 2,75 0,2 1,492 UTP CAT5e 0,880 2,15 1,6 0,820 UTP CAT6 0,835 2,40 1,4 1,800 Tab. 2: Others parameters of the reference model. Cable type φRS0[Ω·m−1]Z0∞[Ω] TCEPKPFLE 1,530e-2 0,2800 131 75 ×4×0,4 SYKFY 4 ×2×0,5 9,992e-3 0,1900 78 UTP CAT5e 1,100e-3 0,1659 97 UTP CAT6 2,200e-3 0,16100 102 Tab. 3: Output parameters of the reference model. Cable type Cp0[F·m−1]LS∞[H·m−1]ηV F TCEPKPFLE 3,877334e-11 6,653892e-07 0,6562575 75 ×4×0,4 SYKFY 5,667929e-11 3,448368e-07 0,7539799 4×2×0,5 UTP CAT5e 4,405455e-11 4,145093e-07 0,7800388 UTP CAT6 4,655176e-11 4,843245e-07 0,7020087 c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 339 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE calculated by using derivation of measured phase characteristic. 4. Conclusion According to our results, the TCEPKPFLE cable can be used up to the frequency of 150 MHz (Fig. 2). The propagation of a signal at higher frequencies is performed by crosstalk and couplings between symmetrical pairs, which cannot be used for reliable transmissions. On the other hand, the frequency characteristics of SYKFY, UTP CAT5e and UTP CAT6 cables have the estimated shapes in the whole frequency band that is why the G-FAST models could be successfully applied. In conclusion, we can say that proposed reference model, which is presented in G-FAST draft of recommendation [4], can be used for modeling of metallic cables up to 300 MHz. The modeled parameters will be used for estimation of gigabit digital subscriber lines data rates in the on-line program “xDSL simulator” [7]. Acknowledgment This work was supported by the Grant of the Technology Agency of the Czech Republic, No. TA02011015, Research and development of a new communication system with multi-channel approach and multi-layer co-operation for industrial applications, and was researched in cooperation with CERTICON. References [1] LAFATA, P., P. JARES and J. VODRAZKA. Increasing the Transmission Capacity of Digital Subscriber Lines. In: 35th International Conference on Telecommunications and Signal Processing. Brno: IEEE, 2012, pp. 292–296. ISBN 978-14673-1118-2. DOI: 10.1109/TSP.2012.6256301. [2] LAFATA, P. and J. VODRAZKA. Modeling of Transmission Functions and Crosstalk in Metallic Cables for Implementation of MIMO Concept. Radioengineering. 2009, vol. 18, no. 4, pp. 491– 496. ISSN 1210-2512. [3] CEPA, L., M. KOZAK and J. VODRAZKA. Innovation of Methods for Measurement and Modelling of Twisted Pair Parameters. Advances in Electrical and Electronic Engineering. 2011, vol. 9, no. 5, pp. 220–224. ISSN 1336-1376. [4] ITU-T SG15. Transport network structures. Germany: International Telecommunication Union, 2012. Available at: http://www.itu.int/md/ T09-SG15-120910-TD-WP1-0759/en. [5] TS 101 270-1 V1.1.1.Transmission and Multiplexing (TM); Access transmission systems on metallic access cables; Very high speed Digital Subscriber Line (VDSL); Part 1: Functional requirements. France: European Telecommunications Standards Institute, 1998. Available at: http://www.etsi.org/deliver/etsi_ts/ 101200_101299/10127001/01.02.01_60/ts_ 10127001v010201p.pdf. [6] VODR´ Aˇ ZKA, J. Potential Use of Gigabit Digital Subscriber Lines in Hybrid Access Networks. In: Proceedings of the Ninth International Conference on Digital Technologies. Zilina: IEEE, 2013, pp. 71–74. ISBN 978-80-554-0682-4. DOI: 10.1109/DT.2013.6566290. [7] Simulator of xDSL lines [online]. 2007. Matlab server. Available at: http://matlab.feld.cvut. cz/view.php?cisloclanku=2005071801. About Authors Marek NEVOSAD was born in Prague, Czech Republic. He received his M.Sc. degree in electrical engineering from Czech Technical University in Prague in 2004. His research interests include telecommunication transmission systems and software development. He also participates in numerous international projects focused on new methods in education as a software developer. Pavel LAFATA was born in Ceske Budejovice, Czech Republic in 1982. He received his M.Sc. degree in 2007 and Ph.D. degree in 2011 at Faculty of Electrical Engineering, Czech Technical University in Prague, specializing in Telecommunication Engineering. Currently he works as an assistant professor and junior research assistant at the Department of Telecommunication Engineering of the CTU in Prague. He is a member of the Transmission Media and Systems scientific group at the Department. His research activities are focused mainly on fixed high-speed access networks, the problems related with disturbance and crosstalk in metallic cables for digital subscriber lines and optical access networks and their topologies. Petr JARES is an assistant professor at the Department of Telecommunication Engineering, Faculty of Electrical Engineering, Czech Technical University in Prague. He received his Ph.D. degree in 2008 at Faculty of Electrical Engineering, Czech Technical University in Prague, specializing in c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 340 INFORMATION AND COMMUNICATION TECHNOLOGIES AND SERVICES VOLUME: 11 |NUMBER: 5 |2013 |SPECIAL ISSUE Telecommunication Engineering. For past few years he has worked on various projects in the transmission systems. His current focus of interest is on data transmission in metallic and optical access networks. c 2013 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 341