Digital Subscriber Line (DSL)

• Dr. John M. Cioffi, Stanford University and ASSIA Inc.

• Dr. Sumanth Jagannathan, Stanford University, Palo Alto, CA

• Wooyul Lee, ASSIA Inc

Digital subscriber line (DSL) service is a broadband digital transmission service offered on ordinary telephone lines. DSL is typically used by telephone companies to offer data, video, and voice services over these existing copper telephone lines. DSL does not displace or disrupt the operation and quality of the existing analog telephone service offered on telephone lines, so that both analog phone and broadband services can be simultaneously offered.

Overview

Figure 1: Near-end crosstalk (NEXT) and far-end crosstalk (FEXT).

DSL has its origins in the early telephone company desires to offer an end-to-end digital network with a service then called "integrated services digital network (ISDN)". ISDN was not successful for many reasons with one being a very low speed. This low speed was caused by “crosstalk” between telephone lines, where crosstalk refers to the electromagnetic coupling between lines. The large noise that results from the sum of crosstalk from all the other telephone lines reduced data rates. Figure 1 illustrates crosstalk, where near-end crosstalk (shown as NEXT) is a larger noise between signals in opposite directions and far-end crosstalk (shown as FEXT) is between signals in the same direction. The large NEXT lowered ISDN speeds. The basic concept of offering digital services evolved when, in 1987, Joseph Lechleider of Bellcore suggested that transmission over a wider separate bandwidth downstream at the expense of a more narrow bandwidth upstream would increase speeds sufficiently that movies could be offered over telephone lines. Thus, NEXT was eliminated in ADSL, unlike ISDN, by separating the upstream and downstream frequency bands. This method was called “Asymmetric” DSL (ADSL) because of the different data rates in the two directions, downstream (movies with high data rates) and upstream (channel-changing controls with low data rates). Unfortunately, phone lines had such a wide variation in length and amount of crosstalk that Lechleider’s early ADSL proposal still was not quite practical for wide-scale use. In the early 1990s, a small company named Amati (a technology “spin-out” of Professor J. Cioffi’s research group at Stanford University) suggested a highly adaptive approach called “Discrete Multitone (DMT)”, which became the basis today for all the modern DSL standards in use and allowed significantly higher data rates than had been believed to be possible. DMT adapted its frequency plan to the line length and crosstalk noise, thereby improving both downstream and upstream data rates. Speeds of a DSL service depend on the length of the line, but can be as low as a few hundred kilobits per second for lines of 4-6 km in length and can be as high as hundreds of Mbps for lines of one km in length or less. Thus, many telephone companies pay to install fiber only to bring their presence closer to the customer, thus dividing fiber’s high costs over hundreds to thousands of users, effectively shortening the phone lines, and thus offering very high-speed DSL services, sometimes called VDSL.

Key technologies and features

Extended bandwidth over analog telephones

Figure 2: Frequency division between analog telephone, upstream DSL, and downstream DSL.

Figure 2 illustrates the shared use of the copper telephone line showing that upstream information from a customer to the phone company passes in a frequency band (US0) just above the old analog phone-service band of 0-4 kHz, while downstream transmissions from the phone company to the customer occupy yet higher frequencies (DS1) on the copper. The further high-frequency bands (US1, DS2, etc.) can also be used for upstream or downstream transmission if allowed by the DSL standard, currently up to a maximum of 30 MHz. ADSL1 uses only US0 and DS1 while VDSL2 uses up to DS2. DSL thus differs from analog modems that reuse the existing analog band and are limited to its 0-4 kHz bandwidth. Unlike analog modems, DSL requires a second hardware transmission and reception device be placed at the side of the telephone company, which is typically housed in a DSLAM (DSL Access Multiplexer) that contains several such devices for several phone lines. A transmission and reception device is also present at the customer’s end, typically a desk-top modem or perhaps one installed at the entry to the customer’s premises. Typically, passive microfilter devices can be placed in series with any analog telephone to prevent higher-frequency "ring-trip" voice-band signals from disrupting the DSL signals. This device also protects the telephone voice-band signals from the interference coming from the DSL signals.

Discrete multitone

Figure 3: A DMT system that matches the transmitted signal to each line’s specific channel by using water-filling and bit-loading.

Discrete Multi-tone (DMT) is a multi-channel transmission technique that divides the available spectrum into smaller sub-channels or tones. For example, in ADSL, the 0 to 1.104 MHz spectrum is divided into 256 small, 4.3125-kHz wide, sub-channels. DMT employs Fast Fourier Transform (FFT) operations at the transmitter and the receiver to simplify multichannel transmission. The basic concept behind using multichannel transmission is illustrated in Figure 3. In both situations (a) and (b) in the figure, each line would experience severe inter-symbol interference if a single wideband signal were transmitted. A heuristic view of DMT is that it partitions the transmit spectra into narrow subchannels, and only those subchannels that pass through the channel are loaded with information (or bits) via the process of water-filling and bit-loading. In implementation, the word ‘narrow’ is intended in comparison to the reciprocal of the time-domain guard interval placed between transmission of consecutive DMT symbols. When the number of tones is large, so that over any fixed bandwidth the tones are correspondingly `narrow,’ the overhead of the guard period becomes negligible. When the number of subchannels is sufficiently large, then an equalizer is not required. Moreover, situation (b) in the figure also shows a common scenario where crosstalk (NEXT or FEXT) from other DSLs or radio-frequency interference ingress, for example AM radio, could significantly impair some of the frequencies in the band of interest. DMT can also counter such impairments using bit-loading during initialization of the DSL modem.

For DSL, DMT differs from the OFDM used in wireless systems in two fundamental ways. The first is the variable bit-loading on each carrier that is used for DMT. This bit-loading in early DSLs is large upon initialization and incremental thereafter so that large changes in channel noise may require a retrain. The Dynamic Spectrum Management (DSM) later also addresses means for allowing more smooth and rapid variation in newer DSL systems. Second is the DMT system is baseband in that it uses an FFT of size two times the number of carriers to ensure the actual transmitted and received signal is one-dimensional and real. (Wireless systems instead use a size N complex-to-complex FFT that produces a quadrature or two-dimensional signal.)

The duplexing method used for ADSL and VDSL DMT is frequency-division as shown in Figure 2 above. This type of duplexing avoids what is called "near-end crosstalk," which is interference between DSL signals traversing opposite directions on the loop. Such crosstalk can be a large from a transmitter in one direction placed closed to a receiver for the opposite direction. This duplexing is achieved with analog filtering in ADSL. However, VDSL progressed to a digital-signal processing method known as digital duplexing or "zippering," introduced by (Sjoberg et al., 1999).

To ensure the independence of the tones in DMT, a guard interval is inserted between successive time-domain packets that contains roughly 8% of the packets time-domain samples repeated at the beginning of the block of samples. This is sometimes also known as a cyclic prefix. The cyclic prefix ensures a simple implementation of one transmit IFFT and one receive FFT. If the guard interval is not sufficiently long for the channel, then some equalization may be used to confine the span of the channel's impulse response to within a time period equal to the length of this guard interval. This technique has many equivalent implementations and sometimes called a "TEQ" or time-domain equalizer.

Selection of DMT

Some readers of this article have asked about the selection of DMT for DSL, so this section is added to explain it. While it was initially controversial, several early manufacturers were misled by supposed theoretical equivalences between single-carrier DSL proposals and DMT. These proposals concluded that the two would perform about the same if implemented correctly. The theoretical equivalence of multi-tone transmission and single-carrier only holds under strict conditions that are almost never met in DSL channels. The lack of equivalent is why DMT outperformed (by more than a factor of 10 in noise immunity, thus an SNR advantage of more than 10 dB) single-carrier methods in two separate DSL competitions held by neutral test laboratories for ADSL in 1993 and again for VDSL in 2003. DMT has a highly efficient implementation that benefits from Fast Fourier Transform (FFT)s that are often used in sophisticated digital-signal-processing systems to reduce the complexity. Even with infinite complexity and infinitely long “equalizer” filters, single carrier methods will not perform as well as DMT on DSL connections – single-carrier methods can always perform as well as DMT on voiceband modem channels below 4 kHz of bandwidth, which lead to the confusion of many in proposing the single-carrier methods for DSL.

Adaptation

Figure 4: Bit-swapping illustration. For a larger figure, refer to [1].

Figure 4 illustrates the key to DMT, in which each subchannel of the DMT system which is continuously adapted in terms of its information content using a process called bit-swapping. After initial loading, changes in the channel gain or noise spectrum are tracked by a DMT-based DSL modem by moving bits between subchannels in order to maintain the target error probability. Gain-swapping is also performed sometimes wherein energy is re-assigned from one subchannel to another after initialization. In some cases, the total power used may also vary after initialization. This process of bit-swapping is independent of any variation in the underlying exact symbol lengths and inter-symbol guard intervals, the latter of which may vary slightly from channel to channel in some of the advanced ITU transmission standards’ details. Typically bit-swapping retains the same data rate but redistributes energy so that best (and lowest necessary power) bit distribution is maintained. Bit swapping and gain swapping allow up to at least a 14 dB range of noise variation (a factor of 60 and large enough to handle many noises). However, some noises may exceed this range and then re-initialization is necessary. DMT’s dynamic adjustment to each and every phone line (through loading), and to the variations in noise (through swapping), enabled the DSL industry. A series of first American, then European, and finally international standards evolved around DMT. The dynamics also allowed a much higher upstream bandwidth to be restored (so while still asymmetric, DSL applications other than movies could be accommodated). The DMT ADSL specification was written at the time that internet web surfing became popular and thus DMT’s agile use of spectrum was an excellent match to internet traffic (and certainly good enough for channel changing as well). Fast internet access demand drove DSL as its first application, while television/video and voice over internet applications followed. Some DSL systems today may not swap correctly (a manufacturer fault since bit-swapping is mandatory in all DMT DSL standards), and the consequent loss in performance can be dramatic.

Forward error correction

Figure 5: Illustration of 3/18 bytes in error that R/N = 1/3 can completely correct.

All DSL systems today use what is called forward-error-correction (FEC). Telephone lines experience intermittent noises that come and go and may change too rapidly for bit-swapping to react. FEC methods add redundant parity bytes to packets (called “codewords”) of information. These parity bytes allow the transmitted data to be recovered even in the presence of channel errors caused by intermittent noise. In addition to the introduction of DMT, the Amati/Stanford group also introduced FEC to DSL. The FEC technology is also crucial to the commercial success of DMT. While the Reed Solomon codes used in ADSL and VDSL were well-known, they were not previously used or suggested in other non-DMT DSL standards. The amount of parity in the ADSL1 standard was highly variable to allow a sufficient amount of parity to recover from the worst noises observed on a line. Addition of parity reduces the bit rate with respect an overall bit-clock speed, but actually causes the number of error-free bits transmitted to increase if used properly. Figure 5 shows such an example where using a parity byte (R) to codeword length (N) ratio of 1/3 can help correct intermittent-noise-induced error rates as high as 17% of the bytes. Later standards of ADSL2+ and VDSL2 reduced FEC options mistakenly, but a new enhanced-VDSL effort will likely restore them. Some ADSL1 manufacturers also did not follow standards and so the parity options were not fully available, often leading to early observed “DSL instabilities”. Recently dynamic management of DSL systems has addressed this oversight via Dynamic Spectrum Management and has assisted the growth of DSL service.

Main DMT-based DSL standards

The DMT-enabled DSL culminated in the widely-used worldwide ITU recommendations:

1. G.992.1 (ADSL1) – speeds to 8 Mbps downstream and 800 kbps upstream (up to 6 km) using 256 tones.
2. G.992.5 (ADSL2+) - speeds to 25 Mbps downstream and 3 Mbps upstream (up to 3 km) using 512 tones.
3. G.993.2 (VDSL2) - speeds to 150 Mbps down and 100 Mbps up (1 to 2 km) using 2048 tones (with an option to use wider tones at 8 KHz for yet higher speeds).

There are currently enhanced VDSL2 standards in definition known as "G.inp" for restoration of the FEC range so that sever impulse noise can be accommodated and "G.vector" for advanced methods that reduce crosstalk. Further, yet higher-speed recommendations are in development and may lead to Gbps speeds eventually on two to four coordinated phone lines to each customer.

Future of DSL

Dynamic spectrum management

Figure 6: Dynamic Spectrum Management Levels (DSM Technical Report, 2007).

Dynamic Spectrum Management (DSM) is a rapidly growing area that enables DSL services to expand. DSM was first standardized in an American National Standards Institute (sub group Alliance for Telecommunications Industries Solutions – ATIS) DSM Technical Report (DSM Technical Report, 2007). That document specifies three levels of increasingly sophisticated management of DSL services. Figure 6 shows the summary of the different DSM levels. Level 1 DSM deals with impulse-noise control and politeness, wherein DSL data rates are improved by resolving channel impairments using techniques that are polite towards other lines sharing the same cable or binder. Methods in this category are often called "tiered rate adaptation (TRA)". Level 1 DSM is sometimes also known under the un-standardized name "dynamic line management (DLM)". Level 2 DSM imposes spectral-politeness controls, largely addressing crosstalk from short lines into long lines (sometimes called the “near/far” problem in electrical engineering) where the short lines’ crosstalk is so large that it overwhelms longer lines signals. Methods in this category are exceptionally polite and often known as "spectrum balancing", "band-preference", or "strong/weak" methods. Level 3 DSM is known as vectoring (Ginis and Cioffi, 2002) and allows for special processing methods in a DSLAM to essentially eliminate all crosstalk between lines, leading to 100s of Mbps speeds in DSL. Modems that support DSM Level 3 management interfaces are being defined in ITU standards project "G.vector".

Figure 7: A Spectrum Management Center (SMC) optimizes the performance of the DSL customers using DSM.

As shown in Figure 7, a Spectrum Management Center (SMC) monitors the channel and noise conditions of the telephone lines sharing the same cable or binder (Song et al., 2002; Kerpez et al., 2003). Based on the available data about the lines and the controls at DSM levels 1, 2, or 3, the SMC optimizes the performance of the DSL lines thus providing a stable service with higher throughput. It is important to note that each service provider in a unbundled DSL operation environment (that is one in which multiple service providers are permitted by local regulations to attach to telephone lines in the same cable of telephone lines) may independently operate an SMC. There is no need for central coordination of all the lines, but instead each service provider may manage their own lines subject to local DSM regulations on such management (for example the North American DSM report (DSM Technical Report, 2007) specifies some informative methods that are also subject to a North American spectrum-management standard on maximum power spectral densities allowed in various situations).

Level 1 DSM focuses largely upon the management of a single line. For example, impulse-noise control is performed by using the Impulse Noise Protection (INP) and delay parameters of each DSL. INP is a parameter in standards that measures the length of noise bursts (in units of 0.25 milliseconds), and delay refers to the delay caused by the combination of transmitter interleaving and receiver de-interleaving in a DSL connection, measured in milliseconds These two controls together essentially allow the parity/packet-length ratio to be controlled. The ratio of INP to delay essentially measures the amount of errors that need to be corrected in a given time frame. Higher INP/delay means more parity bytes per codeword in the applied FEC. Using the statistics of line performance data, the SMC can further stabilize the lines by re-profiling or changing any relevant control parameters. Simple variation of the upstream transmit spectrum according to measured line attenuation (or thus effectively length) could be viewed as a form of level 1 DSM and is sometimes also known as upstream power back-off or UPBO.

Figure 8: Illustration of crosstalk reduction using Level-2 DSM for downstream transmission. For a larger figure, refer to [2].

Level 2 DSM addresses spectrum optimization across multiple DSL lines in a cable/binder (Yu et. al., 2002), (Cioffi et. al., 2004), (Cendrillon et. al., 2006), (Statovci et. al., 2006), (Papandriopoulos et. al., 2006), (Yu and Lui, 2006), (Cendrillon et. al., 2007), (Lee et. al., 2007), (Jagannathan and Cioffi, 2008). Spectrum balancing methods are useful when the crosstalk between lines is significant, but cannot be canceled by more sophisticated Level-3 DSM methods. At this level, the SMC uses minimal control to enable a distributed optimization of each line's transmit spectrum to result in an overall benefit in the network in terms of data-rate improvements or a power reduction. As an example, Figure 8 illustrates the crosstalk reduction possible in a Level-2 DSM system when the SMC directs the line served by the fiber-fed remote-terminal to be polite towards the line served by the central-office. Such a shorter line is designated as a "strong" line by the SMC and instructed to be exceptionally polite, often causing its spectrum to shift to preferred higher frequency bands unused by the longer lines, a practice known as "band preference".

Figure 9: A basic vectored-DSL system.

A Level 3 DSM system (vectored-DSL) (Ginis and Cioffi, 2002), (Cendrillon et. al., 2006), (Cendrillon et. al., 2007) can cancel the upstream or downstream crosstalk by coordinating signals at the central-office or line-terminal, thus increasing the data rates substantially over Level 1 or 2 DSM methods. The term vector is used because the DSL’s individual physical layer voltages can be viewed as a coordinated set or vector of voltages. The group or vector is processed by a common signal processing device for downstream transmission and also for upstream reception as shown in Figure 9. Essentially, the vector/MIMO processor performs pre-processing of the transmitted signal in downstream transmission via pre-coding or linear pre-filtering, and joint processing of the received signals in the upstream via receive filtering and successive cancellation. This group processing allows cancellation or removal of crosstalk. The gain from the vectoring is largest when all the lines in the binder are processed simultaneously, but even partial vectoring or independent cancellation by different operators provide significant improvement over non-vectored systems.

Another major component of advanced DSL management is a database of loop information for DSL provisioning and maintenance (Kerpez et al., 2003). This database is envisioned as having a variety of information on loops, noise, and the histories of deployed DSLs extending far beyond existing loop databases. It would store loop makeups or loop responses, data on deployed DSLs, binder information, measured noise, information on crosstalk between lines, and so on. From this point of view, techniques for loop make-up identification as well as crosstalk identification become important enablers for DSM. One good example of these techniques is SELT (Single-ended loop testing), which enables loop make-up and crosstalk identification without installing any device at the customers’ side, and is standardized as G.selt by ITU. Information gathered from DSL equipment often is collected through an SNMP (Simple Network Management Protocol) line at the SMC and processed to identify line transfers, crosstalk levels and transfers between lines, as well as to calculate and characterize other line noises. Older systems may instead collect such information through element management systems (EMS) over a protocol known as TL1. Such interfaces are standardized in the American DSM Technical Report (DSM Technical Report, 2007) as well as the International Telecommunications Union G.997.1 standard (ITU G. 997.1, 1998), and are sometimes known as maintenance information bases (MIBs).

Such information can be used to characterize DSLs as being significantly affected by the presence of bridged taps (unused branches of the telephone line), bad connections, poor balance with respect to ground and types of noise (like AM radio, crosstalk, impulse). When significant, the DSL operator may elect to use this information to organize maintenance activities to remove the problems. The interested reader is directed to (Zeng et al., 2001), (Galli et al., 2001), (Galli and Waring, 2002), (Salvekar et al., 2002), (Bostoen et al., 2002), (Papandreou and Antonakopoulos, 2005), (Galli and Kerpez, 2006), (Shi et al., 2006), and (Boets et al., 2006) for more information on this topic.

Vectored DSLs may also offer the possibility of using more than 30 MHz of twisted-pair bandwidth. Some pairs (those involving aluminium of some type or other non-copper metals) may have linear bandwidth to only roughly 30 MHz, but others may have a wider usable bandwidth such as evident for instance in category 5 wiring used in Ethernet to 100's of MHz. Some future evolutions may thus exploit even wider bandwidths on the copper twisted pair than 30 MHz.

Gigabit DSL, Common-mode, CuPON

Figure 10: The Basic CuPON architecture

Even better DSL performance can be achieved by using the hidden degrees of freedom in transmission compared to current differential-mode-based vectoring systems. In differential-mode vectoring, the signal is transmitted in the form of the difference between the signals of two lines in a twisted-pair (or two wires of a quad cable), and the number of degree of freedom in a binder is equal to the number of pairs. Instead, in full vectoring, the signal is transmitted through each line with respect to a single ground, which can be one of lines or a binder sheath (Lee et. al., 2007), (Jagannathan et. al., 2008). The signals on all wires can also be sensed (even when all modes are not excited), which is sometimes called "split-pair" sensing. These additional sensed signals can be very useful in the removal of crosstalk. Then, the number of degree of freedom is doubled compared to the differential vectoring. The increased number of transmit dimensions along with a possible multi-pair drop to each customer allows gigabit DSL service (0.5–1 Gb/s data rate per customer) using the CuPON architecture suggested in Figure 10 (Cioffi et al., 2007).

Applications

As the demand for the video service dramatically increases including IPTV, every effort to increase the capacity of access networks has been made recently. DSL methods as described here address that challenge and the latest DSM methods provide a DSL path to bidirectional transmission of 100s of Mbps to each and every customer. DSL thus is a strong alternative to the much higher cost alternative of trenching a fiber to each and every customer that is required in so-called "passive optical networks (PONs)" or more generally fiber-to-the-home (FTTH) systems. The high speeds obtained using ADSL, VDSL, and DSM have enabled high-speed internet services in addition to video and real-time television broadcast delivery. Voice over internet (protocol) or VoIP services are most often also today served over DSL connections, offering the possibility to augment or to replace existing analog phone service altogether. In addition, future vectored-DSL systems as well as CuPON systems would enable a variety of new applications to be supported, for example, next-generation home networks.

References

• Dynamic Spectrum Management Technical Report (2007), ATIS Committee NIPP Pre-published document ATIS-PP-0600007.
• ITU-T Recommendation G. 997.1, Physical Layer Management for Digital Subscriber Line (DSL) Transceivers, Geneva, Oct. 1998.
• T. Bostoen, P. Boets, M. Zekri, L. Van Biesen, T. Pollet, and D. Rabijns (2002) Estimation of the transfer function of a subscriber loop by means of a one-port scattering parameter measurement at the central office. IEEE Journal on Selected Areas in Communications, vol. 20, no. 5, pp. 936–948.
• P. Boets, T. Bostoen, L. Van Biesen, and T. Pollet (2006) Preprocessing of Signals for Single-Ended Subscriber Line Testing. IEEE Transactions on Instrumentation and Measurement, vol. 55, no. 5, pp. 1509 - 1518.
• R. Cendrillon, G. Ginis, E. Van Den Bogaert, M. Moonen (2006) A Near-Optimal Linear Crosstalk Canceler for Upstream VDSL. IEEE Transactions on Signal Processing, vol.54, no.8, pp. 3136-3146.
• R. Cendrillon, G. Ginis, E. Van den Bogaert, M. Moonen (2007) A Near-Optimal Linear Crosstalk Precoder for Downstream VDSL. IEEE Transactions on Communications, vol. 55, no. 5, pp. 860-863.
• R. Cendrillon, W. Yu, M. Moonen, J. Verlinden, T. Bostoen (2006) Optimal multiuser spectrum balancing for digital subscriber lines. IEEE Transactions on Communications, vol. 54, no. 5, pp. 922-933.
• R. Cendrillon, J. Huang, M. Chiang, M. Moonen (2007) Autonomous Spectrum Balancing for Digital Subscriber Lines. IEEE Transactions on Signal Processing, vol. 55, no. 8, pp. 4241-4257.
• J. M. Cioffi, W. Rhee, M. Mohseni, and M. H. Brady (2004) Band Preference in Dynamic Spectrum Management. Eurasip Conference on Signal Processing.
• J. M. Cioffi, S. Jagannathan, M. Mohseni, G. Ginis (2007) CuPON: the Copper alternative to PON 100 Gb/s DSL networks. IEEE Communications Magazine, vol. 45, no. 6, pp. 132-139.
• S. Galli and K. J. Kerpez (2006) Single-ended loop make-up identification — Part I: A method of analyzing TDR measurements. IEEE Transactions on Instrumentation and Measurement, vol. 55, no. 2, pp. 528–537.
• S. Galli, C. Valenti, and K. Kerpez (2001) A Frequency-Domain Approach to Crosstalk Identification in DSL Systems. IEEE Journal on Selected Areas in Communications, vol.19, no.8.
• S. Galli and D. L. Waring (2002) Loop makeup identification via single ended testing: Beyond mere loop qualification. IEEE J. Sel. Areas Commun., vol. 20, no. 5, pp. 923–935.
• G. Ginis and J. Cioffi (2002) Vectored Transmission for Digital Subscriber Line Systems. IEEE Journal on Selected Areas in Communications, vol. 20, no. 5, pp. 1085-1104.
• S. Jagannathan and J. M. Cioffi (2008) Distributed Adaptive Bit-loading for Spectrum Optimization in Multi-user Multicarrier Systems. Elsevier Physical Communication, vol. 1/1, pp. 40-59, doi:10.1016/j.phycom.2008.01.005.
• S. Jagannathan, V. Pourahmad, K. Seong, J. M. Cioffi, M. Ouzzif, and R. Tarafi (Accepted 2008) Common-mode Data Transmission using the Binder Sheath in Digital Subscriber Lines. IEEE Transactions on Communications.
• K. Kerpez, D. Waring, S. Galli, J. Dixon, and P. Madon (2003) Advanced DSL management. IEEE Communication Magazine, vol. 41, pp. 116-123.
• B. Lee, J. M. Cioffi, S. Jagannathan, K. Seong, Y. Kim, M. Mohseni, and M. H. Brady (2007) Binder MIMO channels. IEEE Transactions on Communications, vol. 55, no. 8, pp. 1617-1628.
• B. Lee, J. M. Cioffi, S. Jagannathan, M. Mohseni (2007) Gigabit DSL. IEEE Transactions on Communications, vol. 55, no. 9, pp. 1689-1692.
• W. Lee, Y. Kim, M. Brady, and J. M. Cioffi (Submitted 2007) Distributed Band-Preference Dynamic Spectrum Management for Digital Subscriber Lines.
• N. Papandreou and T. Antonakopoulos (2005) Far-end crosstalk identification method based on channel training sequences IEEE Transactions on Instrumentation and Measurement, vol. 54, no. 6, pp. 2204-2212.
• J. Papandriopoulos, J. S. Evans (2006) Low-Complexity Distributed Algorithms for Spectrum Balancing in Multi-User DSL Networks. IEEE International Conference on Communications.
• A. A. Salvekar, J. Louveaux, C. Aldana, J. L. Fang, E. de Carvalho, and J. M. Cioffi (2002) Profile detection in multiuser digital subscriber line systems. IEEE Journal on Selected Areas in Communications, vol. 20, no. 5, pp. 1116-1125.
• Y. Shi, F. Ding, and T. Chen (2006) Multirate Crosstalk Identification in xDSL Systems. IEEE Transactions on Communications, vol. 54, no. 10, pp. 1878-1886.
• F. Sjoberg, M. Isaksson, R. Nilsson, P. Odling, S. K. Wilson, P. O. Borjesson (1999) Zipper: A duplex Method for VDSL Based on DMT. IEEE Transactions on Communications, vol. 47, no. 8, pp. 1245-1252.
• F. Sjoberg, R. Nilsson, M. Isaksson, P. Odling, and P. O. Borjesson (1999) Asynchronous Zipper. IEEE International Conference on Communications, vol. 1, pp. 231-235.
• K. B. Song, S. T. Chung, G. Ginis, and J. M. Cioffi (2002) Dynamic spectrum management for next-generation DSL systems. IEEE Communications Magazine, vol. 40, no. 10, pp. 101-109.
• D. Statovci, T. Nordström, and R. Nilsson (2006) The normalized-rate iterative algorithm: A practical dynamic spectrum management method for DSL. EURASIP J. Appl. Signal Process., vol. 2006, pp. 1–17.
• W. Yu, G. Ginis, and John Cioffi (2002) Distributed Multiuser Power Control for Digital Subscriber Lines. IEEE Journal on Selected Areas in Communications, vol. 20, no. 5, pp. 1105-1115.
• W. Yu and R. Lui (2006) Dual Methods for Nonconvex Spectrum Optimization of Multicarrier Systems. IEEE Transactions on Communications, vol. 54, no. 7, pp. 1310-1322.
• C. Zeng, C. Aldana, A. Salvekar, and J. M. Cioffi (2001) Crosstalk Identification in xDSL systems. IEEE J. Sel. Areas Commun., vol. 19, no. 8, pp. 1488–1496.

Recommended reading

• P. Golden, H. Dedieu, and K. S. Jacobsen (2006) Fundamentals of DSL Technology. Auerbach Publications.
• T. Starr, J. Cioffi, and P. Silverman, (1999) Understanding Digital Subscriber Line Technology. Prentice Hall.
• T. Starr, M. Sorbara, J. Cioffi, and P. Silverman, (2003) DSL Advances. Prentice Hall.

External links

• Prof. John M. Cioffi’s website
• Sumanth Jagannathan’s website
• Telcordia's DSL Spectral Compatibility Computer (managed by Ken Kerpez)

See also

Sponsored by: Prof. Francesco Vatalaro, University of Rome Tor Vergata, Italy

Reviewed by: Dr. Stefano Galli, Panasonic USA

Reviewed by: Dr. Giovanni Santella, Autorità per le Garanzie nelle Comunicazioni, Italy

Reviewed by: Dr. Rickard Nilsson, FTW, Austria

Accepted on: 2008-07-07 05:41:03 GMT