National Academies Press: OpenBook

Broadband: Bringing Home the Bits (2002)

Chapter: A Broadband Technologies

« Previous: Bibliography
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 245
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 246
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 247
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 248
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 249
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 250
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 251
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 252
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 253
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 254
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 255
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 256
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 257
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 258
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 259
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 260
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 261
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 262
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 263
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 264
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 265
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 266
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 267
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 268
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 269
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 270
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 271
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 272
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 273
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 274
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 275
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 276
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 277
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 278
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 279
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 280
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 281
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 282
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 283
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 284
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 285
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 286
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 287
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 288
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 289
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 290
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 291
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 292
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 293
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 294
Suggested Citation:"A Broadband Technologies." National Research Council. 2002. Broadband: Bringing Home the Bits. Washington, DC: The National Academies Press. doi: 10.17226/10235.
×
Page 295

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

A Broadband Technologies In the course of its work, the Committee on Broadband Last Mile Technology developed highly detailed material related to various broad- band technologies. The committee decided that this level of detail was not appropriate for the main text of its report, but provides the material, which is not intended to be comprehensive, in this appendix for the reader interested in learning more about broadband technologies. HYBRID FIBER COAX TECHNOLOGY1 Coaxial Cable The foundation upon hybrid fiber coax (HFC) broadband communi- cations networks are based is coaxial cable (Figure A.1), a radio frequency (RF) transmission line capable of transporting a large number of carriers (channels). At the head end, or central signal-processing center, each car- rier is modulated with baseband analog or digital information, and all carriers are multiplexed together in the frequency domain (Figure A.2). Spectral separation is accomplished through the use of frequency-selec- tive diplex filters to allow simultaneous transmission of information in opposite directions (Figure A.3), commonly called “reverse” (i.e., from the home to the head end) and “forward” (from the head end to the 1Adapted from James Chiddix. 1999. “The Evolution of the U.S. Telecommunications Infrastructure Over the Next Decade. TTG2: Hybrid-Fiber-Coax Technology” (IEEE work- shop paper). 245

246 APPENDIX A A, Metallic center conductor B, Dielectric (nonconducting) C, Metallic outer conductor D, Plastic jacketed covering (optional) FIGURE A.1 Coaxial cable (cut-away view). Two-Way TV TV Two-Way Radio Ham 2-6 FM 7-13 216 - 470 MHz Rev 79 Analog TV Channels 5-40 MHz 54-550 MHz FIGURE A.2 Typical RF spectrum for analog cable television. Forward carriers RF spectrum Reverse carriers Head-end equipment Diplex filter FIGURE A.3 Forward and reverse spectra and diplex filter.

APPENDIX A 247 home). This physical medium provides for the transport of RF energy within a reasonably secure network with an enormous amount of signal capacity and flexibility. Conceptually, coaxial cable provides cable operators with a private conduit through which RF signals are transported; in addition, the me- dium can support multiple signaling channels without regard to the base- band signals or modulation scheme that may be employed. This medium is generally immune to interfering influences that may exist in free space. From a practical standpoint, coaxial cable supports transmission of sig- nals at frequencies from baseband to more than 1 GHz. Transmission losses (attenuation) within these cables can be signifi- cant; attenuation increases proportionally with frequency, making it nec- essary to use RF amplification to cover the long distances encompassed by the cable plant. Amplifiers may be spaced from a few hundred feet to one or two thousand feet apart. Although the theoretical frequency limit of the cable itself is significantly greater than 1 GHz, and cable systems have been built using upper frequencies in excess of 1 GHz, practical limitations are set by the frequency responses of active and passive com- ponents (e.g., amplifiers and filters) used in the network. Tree-and-Branch Architecture Until recently, coaxial cable systems have followed a “tree-and- branch” topology (Figure A.4), delivering the same RF spectrum of sig- 40+ amplifier Trunk cable cascades were (coax) common RF amplifier HEAD END FIGURE A.4 Tree-and-branch architecture.

248 APPENDIX A Carrier (channel) Carrier-to - Carrier-to-noise intermodulation ratio ratio Noise FIGURE A.5 Carrier-to-noise and carrier-to-intermodulation distortion. nals to every customer within a particular community. This design served the cable industry well, but it did have limitations. The most significant restriction imposed by this topology was the accumulation of noise and distortions (Figure A.5) through the extended cascades of broadband RF amplifiers needed to compensate for transmission losses. This architec- tural facet affected plant reliability and signal quality at the customer’s home. Additionally, for a given design bandwidth, there were practical and theoretical limits to the number of amplifiers that could be cascaded. In order to maintain acceptable performance levels, it was necessary to limit the operational bandwidth of such cable systems to a few hundred megahertz, far below the potential of the cable alone. Another limitation was imposed by this topology: every customer receives the same complement of signals. This is generally acceptable for TV services, but makes the delivery of individually switched or routed services difficult. Fiber-Optic Transmission Technology By the late 1980s, optical lasers were successfully adapted for use in a broadband environment. Optical transmission had been practical for some time through the mechanism of turning the transmitting laser “on” and

APPENDIX A 249 “off” in synchronization with the ones and zeros of a digital signal. A breakthrough came when it was determined that a laser could be left “on” and intensity-modulated with the highly complex analog signal repre- senting the broadband RF spectrum (Figure A.6). Lasers used in this way required characteristics different from their digital counterparts. The most critical were very low internal noise and an extremely linear transfer function. Such devices had been in development for the digital market in an effort to achieve higher data-transmission speeds over optical fibers (in contrast to coaxial cable), but further optimi- zation was required for broadband applications. At the receiving end of an optical link, a relatively simple photo- detector was used to convert the optical signals back into an RF spectrum essentially identical to the one presented at the input (transmitting) end. The cable industry quickly adopted this technology for a portion of its transmission plant, and continues to use it as a way to cost-effectively transform coaxial tree-and-branch systems into something much more powerful—hybrid fiber coax (HFC) architecture (Figure A.7). In essence, this approach transforms large systems into highly concentrated collec- tions of smaller systems. This is a very important characteristic, as dis- cussed below. Current HFC designs are now providing transmission to and from neighborhood clusters of a few hundred homes or fewer (Figure A.8). This arrangement of fiber and coaxial cables allows segmentation of the traditional coax-only transmission plant into many localized areas (called nodes), each of which is capable of providing a unique assortment of information to end users (Figure A.8). The coaxial network that connects to homes from each optical node remains a small version of the original tree-and-branch system (more of a bush than a tree). TV TV Two-way radio UHF spectrum 2-6 FM 7 - 13 216 - 470 MHz 470 - 750 MHz 79 Analog TV Channels Digital Services 54 - 550 MHz 550 - 750 MHz 54 - 750 MHz FIGURE A.6 750-MHz forward spectrum.

250 APPENDIX A Typically 6 or fewer amplifiers in cascade Optical node HEAD END 500 homes passed Optical cable FIGURE A.7 HFC networks allow smaller serving areas. Master head Primary end hub Baseband digital Fiber Coax transport 1,310 nm RF amps Optical node Secondary 1,550 nm 500-2,000 hub High power Home area Taps & passives Optical node FIGURE A.8 HFC networks allow narrowcasting of content to the customer.

APPENDIX A 251 Design Considerations Current HFC designs call for fiber nodes serving about 500 homes on average, but these nodes can be further segmented into arbitrarily small coaxial serving areas. Figure A.9 illustrates one way that the spectrum available within one node may be used. The ability to assign and reassign spectrum to different uses is an important benefit of HFC architecture, because it allows for advances in digital services and technologies while continuing to support existing services. Thus, the architecture can simultaneously support many sepa- rate virtual networks. This makes the investment to upgrade to HFC a sustainable one for most cable companies. At least some cable operators plan to build as many as five separate (virtual) networks on the founda- tion of their upgraded fiber transport plant (Figure A.10). The HFC architecture enables great flexibility to segment the service area. Step-by-step segmentation can match investment with revenues from new, high-bandwidth services; in the extreme case, fiber can be extended to the property lines of homes and businesses (not shown in 2-way TV TV 2-way radio UHF spectrum ham 2-6 FM 7 - 13 216 - 470 MHz 470 - 750 MHz Digital Services Rev. 79 Analog TV Channels (QAM64 - QAM256) QPSK - (NTSC) QAM16 54 - 550 MHz 550 - 750 MHz 5 - 40 MHz 54 - 750 MHz s Digitalce High-speed Voice rvi over IP Voice-on-demand se data SDTV HDTV Digital bank (future use) 550 - 750 MHz FIGURE A.9 Forward and reverse spectra at node.

252 APPENDIX A HUB Mpeg video HFC plant Transport Switch data HFC Fiber Analog video Located in head end FIGURE A.10 Capability to support multiple networks within HFC. Figure A.10), or at least to those with the need for services requiring hundreds of megabits per second of connectivity. Only those nodes that have need of greater data capacity (and the potential for greater revenue) have to be divided; the rest can remain undisturbed. As nodes are divided and fiber is deployed closer to the customer, the total amount of usable bandwidth becomes greater; this makes it possible for every node division to more than double the available data capacity while reducing the number of users who share it.2 Similarly, breaking a 500-home node into four parts, each passing an average of 125 homes, increases the available reverse and forward capacities significantly more than fourfold and provides more than four times the bandwidth per user. Trials within the industry have made use of the spectrum from 900 MHz to 1 GHz (as compared with the traditional use of the 5- to 50-MHz region) for reverse signals. Because of reduced RF interference at these higher frequencies and the resulting higher-modulation efficiencies, it is possible to provide an additional 200 Mbps of transmission capability. Again, this number can be multiplied through segmentation, as outlined above. 2The accompanying reduction in noise over the coaxial portion of the network—in accord with Shannon’s law—means that the usable bandwidth within each subloop also increases significantly.

APPENDIX A 253 It is possible to push these numbers even farther. If very high speed, truly symmetric capacity is required, frequencies above 1 GHz can be used. Some cable plants being constructed today use fiber to feed neigh- borhoods of 60 homes or fewer with a more-than-commensurate increase in the per-user capacity for both switched and routed digital services. In 2001, the latest version of the industry standard, DOCSIS 2.0, em- braced two optional refinements that can substantially increase upstream throughput by using improved modulation in situations where the noise level permits. One is the use of advanced time-division multiple access (TDMA), which allows modulations up to 256 quadrature amplitude modulation (QAM) in upstream bursts (theoretically 8 bits per hertz, real- world about 6.5), compared with the 16 QAM (4 bits per hertz theoreti- cally) of the current version. The other is synchronous code-division mul- tiple access (CDMA), which permits much more robust transmissions in the presence of certain kinds of interference. Providing Services in Year 2010 Information and entertainment services can be classified in two broad categories—common and dedicated. Common services include such pro- gramming as off-air broadcast, PEG (public, educational, and govern- ment) channels, basic networks (such as ESPN and CNN), and subscrip- tion services (such as HBO, Cinemax, and Starz). Dedicated services include any number of specialized programs that are delivered to the end user on an individual basis; video-on-demand (VOD) and high-speed Internet access are examples of this type of service. The cable television (CATV) industry in the United States typically thinks of a channel as being represented by a contiguous 6-MHz portion of the available spectrum—thus, a standard 750-MHz HFC plant has ap- proximately 112 such “channels” within a total usable spectrum of 672 MHz. Table A.1 provides some details regarding a hypothetical 750-MHz HFC plant’s ability to provide almost unlimited service options for cus- tomers, including the following: • Standard analog television. The cable television industry will prob- ably always carry some amount of NTSC signals, perhaps 20 or so RF channels; but it is anticipated that the number of these signals will de- crease as most of them are incorporated into compressed digital formats. • Digital standard definition television (SDTV). This will become the “standard” signal as 256-QAM channels are used to distribute some 200 simultaneous networks (HBO, ESPN, CNN, and so on), including most of the subscription services.

254 APPENDIX A TABLE A.1 Potential Services That a 750-MHz HFC Cable System Could Provide Channels and/or Channels Services Provided Bandwidth Required Remaining Common signals Standard analog television 20 channels (120 MHz) for 92 NTSC signals Digital SDTV 20 channels (120 MHz) for 72 200+ programs of compressed digital video format Digital HDTV 10 channels for 20 programs 62 (60 MHz) Dedicated services Telephony 1 channel (6 MHz) for 300 DSOs 61 (voice channels) IP data—standard service 20 channels (120 MHz)— 41 10-Mbps data rates IP data—very high speed 3 channels (18 MHz)— 38 100-Mbps data rates Video-on-demand 20 channels (120 MHz) for 18 200+ programs of compressed digital video Future 18 channels, services as needed 0 (108 MHz) • Digital HDTV. These networks will be capable of providing ad- equate bandwidth to support as many as 20 RF channels (each 6 MHz wide) as the transition to broadcast HDTV services continues. • Telephony. More than 300 voice channels can be provided within a typical 6-MHz segment of the spectrum, if needed. • IP data services. This class of service includes voice over IP (VoIP), video telephony over IP (VTIP), streaming video, and high-speed data services. Higher-data-rate services (100 Mbps) can be provided, as needed, for work-at-home or commercial uses. • Video-on-demand. Even with all of the services listed above, enough bandwidth remains to handle VOD applications in both SDTV and HDTV formats. Security Considerations Since cable operators have built their plant to provide video and other services, and many of those services are available at lower cost, albeit with lower quality or lacking some other feature, cable operators have had to find ways to secure their services from unauthorized access. The

APPENDIX A 255 typical solution has been to provide a decoder in the customer’s home, then to send commands from a head-end controller to the decoder in order to identify the services for which each subscriber is authorized; this strategy also permits the operator to capitalize on its economies of scale and scope by broadcasting all signals simultaneously over the entire tree- and-branch cable plant. The deployment of HFC networks has complicated the traditional controller-to-decoder scenario. That is, the network architecture and ca- pacity have both changed, enabling the head-end controller to send a discrete broadband signal—custom-tailored to the consumer’s require- ments, preferences, or purchases—to each home. In many parts of the network, the signals for all customers may pass over the bus and to the home of each customer, so the set-top box would be employed to cull out for delivery only those signals that are to be received by a specific con- sumer. When the services are broadcast video programs, there is no in- teractivity and consequently little need for security beyond the remote scrambling of the video signal. However, when interactivity is a signifi- cant portion of the services, the consumer now has access to devices for both receiving and transmitting—particularly with the PC connected to a cable modem connected to the bus network. The potential exists for both intentional and accidental spillage of signals, onto and off of the network. Conclusion The existence of ubiquitous, broadband cable television networks in this country affords an opportunity to see the rapid realization of ex- tremely powerful digital networks. The hybrid fiber coax network offers an excellent high-speed data network solution today and combines that with a high degree of scalability to adapt to new technologies or services that may be introduced in the future. The key to the provision of this capacity is the ability to increase the penetration of optical distribution equipment as the need arises. This path to progress will eventually lead to fiber-to-the-curb (FTTC), and even fiber-to-the-home (FTTH). Leading HFC suppliers drive this technology development and deployment in response to the cable operators’ customer demand and sustaining rev- enue sources. DIGITAL SUBSCRIBER LINE Introduction Digital subscriber line (DSL) service provides high-bit-rate digital ser- vice over ordinary phone lines, allowing from 100 kbps to tens of mega-

256 APPENDIX A bits per second to reach a telephone company customer. DSL service may implement digital telephone service, fast Internet or other data services, and/or digital video and entertainment services. DSL is the phone com- pany’s alternative for broadband access. This section summarizes the ba- sic concept and architectures of DSL service, provides an overview and projection of standards and equipment, and envisions DSL’s future and ultimate broadband-access capabilities of telephone companies. There are 500 million voiceband modems in existence today, most of which are used at speeds to 56 kbps to provide digital connection be- tween various service providers and customers or to transfer data and facsimiles. Voiceband modems are limited in speed because the signals must traverse telephone company switches that allocate only 64 kbps maximum (of which 56 kbps are available) to any voice signal, as shown in Figure A.11. These switches can allow aggregation to higher data rates of several voice channels, but not over a single voice channel through the switch. These digital high-speed data must follow an alternative path through the switch. An additional modem at the telephone company side of the loop differentiates a DSL connection from a voiceband modem connection, as in Figure A.12. DSL’s placement of the extra modem at the telecommunications company (telco) switch enables the much higher speeds of DSL to be switched because the switch can now accept that modem’s digital output into higher-digital-bandwidth routes through the switch. Thus, the DSL signal returns to digital format when it enters the central office, while the voiceband modem signal is effectively embedded in analog throughout the switch network. The bandwidth of the twisted pair alone is potentially very high, much higher than the 64/56 kbps Local loop trunk Local Telco Telco loop CO CO User 1 User 2 modem transmission path FIGURE A.11 Voiceband modem reference model.

APPENDIX A 257 User 1 User 2 modem alt broadband path alt broadband path modem M Local o loop M trunk M Local d split o d Telco Telco o d loop split e m e m CO CO e m modem transmission path modem transmission path FIGURE A.12 DSL modem reference model. allowed for digitized voice paths through the switch. However, high digi- tal speed on the copper loop requires sophistication in the design of the DSL modems that attach to the loop. Telephone companies originally did not appreciate the value of their copper asset and considered replacement with fiber or coaxial systems, but release of DSL standards and availabil- ity of low-cost DSL equipment have provided phone companies with an opportunity to leverage their existing plant. Below are summarized some of the technical challenges of the DSL modem and the use of existing phone lines for high-speed digital service, as well as some of the challenges for network design and support of DSL. Generally, higher DSL data rates occur on shorter phone lines. As phone companies can afford the time and money to install fiber into more of their network, copper phone lines reduce in length. Thus, an incremental migration over the next 50 years to fiber, allowing increasing data rates for customers and greater and greater connectivity and information age service, can occur without need for whole-scale network replacement. Figure A.13 depicts the growth to date in digital transmission speeds on phone lines. Several types of digital transmission on phone lines are shown for comparison. Generally, DSL today often really means ADSL, an asymmetric DSL service that can carry up to 8 Mbps downstream from a telephone company central office to a customer and up to 1.5 Mbps back upstream. Approximately 1 million ADSL lines are now deployed, and the numbers are growing rapidly as early problems and delays with service

258 APPENDIX A VDSL ADSL up to 52 up to Mbps 7 Mbps HDSL 1.5 or 2 Mbps V.34 PCM Basic modem V.22bis modem Bell 103 Bell 202 Rate 56 modem 28.8 kbps modem modem 2,400 b/s ISDN kbps 300 b/s 1,200 b/s 144 kbps 1955 1970 1981 1986 1992 1993 1996 1997 1999 Service introduction date T1 Local Fiber ATM Internet carrier digital optics switch mass switch; market DLC FIGURE A.13 Data rate increase for phone lines. introduction have begun to abate, and telephone company personnel are increasingly trained and fluent in this new service. In a short time, tens of millions of customers will be connected. ADSL service can now be or- dered in nearly one-third of the United States, and telephone companies plan ubiquitous coverage in the near future. VDSL, the latest of the DSLs, can carry up to 60 Mbps on a single phone line and is in early trial and standardization phases. VDSL presumes some use of fiber to shorten phone-line lengths, consistent with eventual migration to fiber by phone companies. These are clearly much faster than voiceband modems. ISDN and high-bit-rate DSL (HDSL), some phone company early DSL alterna- tives, are also shown in Figure A.13 for perspective. ADSL deployment, though, will soon eclipse the number of ISDN and HDSL circuits in ser- vice. As fiber penetrates and very large scale integrated circuit (VLSI) technology allows yet further sophistication in the design of copper-pair modems, eventually 100 Mbps plus symmetric connection to individual customers is possible with DSL, making it by far the broadband access technology of greatest potential individual bandwidth to the customer.

APPENDIX A 259 DSL Standards To establish a DSL connection, two modems—one owned and oper- ated by a telephone company and the other owned by the customer— must interoperate, thus mandating standardization of the interface. As described in Cioffi et al.,3 standards committees have charted the course of DSL technology and the architecture for the associated networks. The International Telecommunication Union (ITU) has headquarters in Geneva, Switzerland, and has a major role in standardization. However, the fundamental DSL standards work has largely been conducted in the T1E1.4 committee of the American National Standards Institute (ANSI),4 the European Telecommunications Standards Institute (ETSI), and the ADSL Forum. The earliest DSL standards, all adopted internationally af- ter minor modification, originated in the American group. These stan- dards groups maintain close cooperation with each other and the ITU. ITU Study Group 15 (SG15) has recently taken the lead in developing an offspring of ADSL called G.lite (also known as Universal ADSL) for con- sumer-oriented use at bit rates of 1.5 Mbps and below. The G.lite standard was released in 1999 as G.992.2 along with an international version of the ADSL standard called G.dmt (G.992.1). The main difference in the two standards is speed, with G.lite at 1.5 Mbps and G.dmt allowing in excess of 8 Mbps. The industry appears to have turned to use of the latter G.dmt modems at the lower speeds of G.lite with imbedded potential for future speed increase as service providers condition and shorten phone lines. (For more on standards bodies and the relationship of the groups for DSLs, see Chapter 16 in Cioffi et al.5) DSL Architectures There are almost 1 billion phone lines worldwide. The telephone lines are twisted pairs of copper wires, with the twisting invented by A.G. Bell himself, in 1887, along with the phones (1876) to which they are attached.6 3J. Cioffi, T. Starr, and P. Silverman. 1998. Digital Subscriber Lines. Prentice-Hall, Upper Saddle River, N.J. 4See American National Standard T1.601-1992, “Integrated Services Digital Network (ISDN) – Basic Access Interface for Use on Metallic Loops for Application on the Network Side of the NT (Layer 1 Specification),” 1992, New York, N.Y., and “VDSL System Require- ments Report,” ANSI Document T1E1.4/98-043R2, June 1998, Huntsville, Ala., Rev 14a. See also ETSI technical specification TS101-270-1 (1998-04), European Telecommunications Stan- dards Institute, Sophia Antipolis, France. 5Cioffi et al., Digital Subscriber Lines, 1998. 6R.B. Bruce. 1973. A.G. Bell, and the Conquest of Solitude. Cornell University Press, Ithaca, N.Y.

260 APPENDIX A The phone lines are varied in a great many respects, but the topology of the loop plant of a phone company usually follows that of Figure A.14. The phone lines are terminated on central office equipment, where the DSL modem can reside. The central office (CO) equipment is con- nected to phone lines at a main distribution frame (MDF) that essentially allows physical connection (“jumpering”) of switch/DSL-modem lines to customer lines—as many as 160,000 of which may enter a single central office. The first segment of the loop plant is typically called the “feeder plant,” where hundreds of phone lines may be bundled in a cable that runs to a smaller distribution point, labeled as SAI in Figure A.14. This feeder segment is the first that phone companies upgrade to fiber, with approximately 10 percent of the United States now so upgraded and smaller percentages in other countries. Such fiber is expensive, but the cost of labor, digging, and so on can be shared over a greater number of customers in the feeder segment, making certain upgrades economical. At the distribution point, splicing and connection to smaller cables con- taining fewer phone lines occurs, and those cables run through the “dis- tribution plant” to pedestals or cabinets within a neighborhood where Customer premises Inside wire NID Central office Drop wire equipment Pedestals SAI Main distributing frame Feeder Distribution 20,000 to 1,500 to 4,000 200 to 800 4 to 12 160,000 Number of lines present at a site 22,000 ft 9,000 ft 3,000 ft 500 ft Wire length to customer (90th percentile) FIGURE A.14 Telephone loop plant topology.

APPENDIX A 261 100 90 Percentage within length 80 70 60 50 40 30 20 10 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 Loop length (km) {1 km = 3.28 ft} FIGURE A.15 Cumulative loop distribution for Italy (solid squares), the United Kingdom (solid circles), and the United States (solid triangles). connection to the actual twisted pair in a specific customer site (home or business) occurs. Phone lines may thus be several miles in length. Eventu- ally, fiber can run to the pedestal as demand for high bandwidth becomes very high, and ultimately to the customer’s premises as economics allow. Figure A.15 illustrates the distribution of lengths of twisted pairs for three countries: Italy, the United Kingdom, and the United States. Clearly the United States has the longest loops, simply because its network was deployed the earliest, when phone company practice was to use longer loops. The United Kingdom is intermediate in terms of loop demograph- ics, while countries that lagged the United States by several decades, such as Italy, have the shortest loop demographics. One can thus expect the achievable data rates to be the lowest for DSLs in the United States. Italy, Germany, and Sweden, for instance, are excellent candidates for higher- speed DSL service because a large fraction of their loops are within a kilometer or two of the central office. However, fiber deployment is oc- curring faster and sooner in the United States in the feeder segment, which will ultimately reverse the relative lengths shown in Figure A.15. For instance, the largest U.S. telephone company, Southwest Bell Corpo- ration, recently announced a $6 billion DSL loop renovation program, known as Project Pronto, that will bring many loops to less than 4 km

262 APPENDIX A within a few years. Project Pronto is an example of reversing the trend in Figure A.15. Deregulation and aggressive DSL and Internet use in the United States have motivated SBC to move quickly, unlike international operators who still have far less competition because unbundling deregu- lation has lagged that of the United States by a few years at least. Central Office Figure A.16 illustrates the network architecture of DSL. A DSL access multiplexer (DSLAM) resides at the telephone company side of the twisted pair. A splitter circuit may precede the DSLAM termination so that analog POTS signals can be passively separated from the DSL signals and conveyed to the voice telephone switches. Splitters are 3-port devices that ensure that DSL signals above 30 kHz and POTS signals below 4 kHz are simultaneously passed over the telephone wire without mutual dis- ruption. The DSLAM houses the modem and processes the customer bit streams into larger rate fiber transported data streams that usually use ATM formatting. Various gateway devices can accept the fiber inputs and Digital Internet service ADSL Analog provider G modem A Mux ADSL split modem T or E Demux ADSL W modem Video A service Y Analog provider (s) DSLAM POTS Digital network Telephone company office FIGURE A.16 Telephone company central office and DSLAM.

APPENDIX A 263 Customer premises microfilter G.pnt Home/business wiring VoDSL ADSL G.pnt interface G.pnt microfilter G.pnt FIGURE A.17 Customer premises (residence or small business) DSL interface. separate the signals into individual applications’ provider networks, such as Internet service providers, video and entertainment providers, or voice service providers. Colocation today involves separate DSLAMs for each service provider. An alternate service provider must be given fair and equal access to the phone lines of the service provider’s customers. Customer Premises Figure A.17 illustrates the customer premises end of a DSL connec- tion. The customer can be a residential user or a business user. While a splitter can be used at the customer premises also, the cost of installation is often perceived as excessive, and so DSL signals typically enter the customer’s premises and terminate on application devices. Existing phones often are augmented by a passive lowpass filter known as a microfilter, which protects the phone from DSL signals and protects DSL from ring-voltage transients that otherwise would be disruptive to DSL service. The DSL modem may be part of a residential or small-business gateway that either connects to another network in the home or redistrib- utes digital signals to all application devices at frequencies above 5 MHz. (ADSL and its latest protégé, very-high-data-rate asymmetric DSL [VADSL or VDSL-lite], exist only below 5 MHz.) The ITU SG15/Q4 group also standardizes this redistribution system, which is known currently as

264 APPENDIX A G.pnt. Customer premises wiring alone can carry huge data rates above 5 MHz, and sophisticated modem design is less necessary because the ac- tual redistributed data rates are well below fundamental limits. However, the DSL signals that traverse the much longer path from central office to customer need a high degree of sophistication to achieve the data rates desired in DSL. DSL Transmission Environment The DSL transmission environment is challenging, and should not be underestimated. This challenge was the first that had to be addressed in developing an opportunity for DSL. In fact, in the early days of DSL, there was little phone company support or interest because it was believed that this challenge would be insurmountable. Fortunately, an initially small (but now very large) group of transmission experts worked together through standards groups to derive practical, high-speed DSL modems. Challenges continue for yet further increases in DSL speed and capability. This section discusses the salient characteristics of telephone lines for digital transmission, with the intent of conveying the difficulty of the transmission problem for DSLs. The journey of a bit over a phone line is analogous to a long, arduous trip with several borders to cross, potentially dangerous trip segments, with various difficulties and costumes imposed upon the traveler, poten- tially disguising that person’s appearance to all but those who know well how to recognize the traveler at the destination. Only the best prepared travelers (bits) can successfully complete the journey, if the receiver also knows well how and what to look for. The shorter the journey, the more successful travelers/bits conveyed to the final destination. Phone lines typically comprise several segments of wire, characterized by gauges (19, 22, 24, or 26 in the United States and equivalent 0.8-, 0.6-, 0.5-, and 0.4- millimeter [mm] diameters in the metric system internationally). The higher the gauge, or more narrow the wire, the more arduous the journey. At the borders between phone line segments, some energy is reflected, meaning that a bit may be harder to recognize as a 1 or 0 by the time it reaches its destination. This energy loss may be equated to an aggressive customs officer confiscating some identifying documents from the trav- eler. Some bit energy may also be diverted to unused open-circuited phone branches (for extension phones or extension phone jacks), further marring the appearance of the bit; these branches are known as bridged- taps. The effect of bridged-taps is analogous to unnecessary dead-end side trips by a traveler to a port that the traveler did not know was closed, but draining their energy with the wasted round trip, making the ex- hausted traveler yet more difficult to recognize. Some standardized phone

APPENDIX A 265 line characteristics and behavior appear in the subsection on loop-transfer characteristics. Phone lines endure a lot of noise, which obscures or disguises the bit. The noise is typically electromagnetically coupled into phone lines. The external sources of energy that contribute to noise can include signals on other phone lines (known as crosstalk), radio, and ham broadcasts, and virtually any type of electrical or mechanical equipment within close proxmity to the phone line. Such noise can be severely disguising, and transmitted bits need to be adequately prepared to avert complete loss of identity if they are to negotiate their journey successfully. The subsection entitled “Sources of Noise in DSL Systems” below, overviews several types of noises. The energy from a bit may also radiate from a phone line, potentially disturbing radios within the vicinity of any portion of the phone line. This is analogous to a boisterous traveler upsetting all the other travelers, thus running the risk of retribution of some sort. The subsection below entitled “Emission Constraints and PSD Masks” describes this problem and the level of concern. Characterization of Twisted-Pair Telephone Lines Chapter 4 of Cioffi et al.7 details the calculation of the frequency- response of phone lines, which are often described by their “insertion loss.” The insertion loss is measured in decibels (10 times the base-10- logarithm) of the ratio of the power injected into a phone line at any given frequency to the power emanating at the end of the phone line at that same frequency. The injected power may be measured without the phone line present, and then measured again after the phone line is “inserted,” whence the name “insertion loss.” Some insertion loss plots versus fre- quency for American standardized 3- and 4-mile phone lines appear in Figure A.18. These loops were chosen by ANSI to represent the top 10 percent of worst-case lines in the United States. Loops 1-4 in the left plot represent simple gauge changes. The usable bandwidth over the lines, where signals are still distinguishable from noise, may extend to about 600 kHz. Note the large range from as high as –20 dB to –100 dB in insertion loss for usable frequencies. This means that the largest signals on the line may be 100 million times more powerful than the smallest signals of interest. By contrast, voiceband modems see a range of only a factor of 100, making DSL transmission a million times more sensitive! 7Cioffi et al., Digital Subscriber Lines, 1998.

266 APPENDIX A ADSL Canonical Loops # 5, 6, 7, and 8 Insertion Loss ADSL Canonical Loops # 1, 2, 3, and 4 Insertion Loss 0 0 -20 -20 -40 -40 Line PSD (dB) -60 -60 -80 -80 -100 -100 -120 -120 -140 -140 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Frequency (Hz) x10 5 Frequency (Hz) x10 5 FIGURE A.18 ANSI loops 1-4 (at the left) and 5-8 (at the right), insertion loss. Loops 5-8 on the right have bridged-taps. Notice the rippling of the inser- tion loss, corresponding to signal energy reflecting from the open-cir- cuited extensions and returning later in time to the main line to add to the current signals there. At some frequencies, the reflected signals are 180 degrees out of phase and destroy the current signals, corresponding to the dips. At other frequencies, energies add, essentially returning the signal closer to its original unreflected signal level. (A good discussion of bridged-tap and other effects appears in an article by J.J. Werner.8) Figure A.19 shows insertion loss characteristics of some shorter stan- dardized loops,9 indicative of what might be used with VDSL over a yet wider bandwidth of 30 MHz. The left plot shows insertion loss for 300-m and 500-m loops of 26-gauge (TP1) and 24-gauge (TP2) wire, respectively. As the length increases, the slope also increases. On the right, a short loop VDSL5 with a bridged-tap has very noticeable rippling, but otherwise is similar in slope to the insertion loss characteristics on the left. As the length is increased to 1 km and then to about 1.5 km for VDSL6 and VDSL7, respectively, the insertion loss decays much more rapidly with frequency and still exhibits significant rippling because of the bridged- taps. The same large dynamic range (now because a greater range of frequencies is used at shorter lengths) is again evident for VDSL. The bridged-tap in VDSL5, 6, and 7 is 10 meters in length, a reason- able and typical number. Thus, the notches are an often-encountered phe- nomena. 8Werner, J.J. 1991. “The HDSL Environment.” IEEE Journal on Selected Areas in Communi- cation 9(6):785-800, August. 9“VDSL System Requirements Report,” ANSI Document T1E1.4/98-043R2, June 1998, Huntsville, Ala., Rev. 14a. See also ETSI technical specification TS101-270-1 (1998-04), Euro- pean Telecommunication Standards Institute, Sophia Antipolis, France.

Insertion loss for "short" VDSL 1, 2, 3 Insertion loss for VDSL 5, 6, 7 0 0 VDSL1 TP1 300m VDSL5 VDSL1 TP2 500m VDSL6 -20 VDSL2 500m -20 VDSL7 VDSL3 500m -40 -40 -60 -60 -80 -80 Magnitude (dB) -100 -100 -120 -120 0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3 x10 7 x10 7 Frequency (Hz) Frequency (Hz) FIGURE A.19 Shorter loops for VDSL systems (TP1 = 26 gauge and TP2 = 24 gauge). 267

268 APPENDIX A Sources of Noise in DSL Systems Noise on a twisted-pair transmission system arises from three mecha- nisms: 1. The thermal noise of the twisted pair itself, 2. The noise generated internally by the receiving modem, and 3. Signals electromagnetically coupled into the phone line. Thermal or actual medium noise on a twisted pair is extremely small, near the Boltzman limit of –174 decibels per millihertz (db/mHz) at room temperature, and essentially can be ignored. The noise generated by ter- minating equipment depends on the design of the receiver electronics. Standards groups often suggest that this noise level should be about –140 dBm/Hz,10 but well-designed modems often generate less, to as low as –160 dBm/Hz. This noise is usually flat in spectrum (i.e., “white”) and determines the ultimate frequency limits of a DSL. The insertion loss of a DSL, as discussed above, and the transmit power-spectral density in dBm/Hz determine the line output power-spectral density (PSD). For instance, an ADSL system transmitting at the maximum PSD of –40 dBm/Hz can tolerate up to about 85 to 90 dB of insertion loss before resulting in a channel output PSD of –125 to –130 dBm/Hz, 10 dB above –140 dBm/Hz (10-dB signal-to-noise ratio) necessary for adequate detec- tion even in well-coded and designed DSLs). Thus if –140 dBm/Hz is the receiver noise floor, then a DSL using one of the lines on the right in Figure A.19 would use bandwidths of up to 600 to 700 kHz. ADSL sys- tems actually allow use of bandwidths up to 1,104 kHz, which would thus occur on lines that are shorter than those displayed in Figure A.19, thus having less insertion loss at 1 MHz. Electromagnetically coupled noise occurs because the twisted pair is often bathed in radiation from a number of electronic sources. The twisted pair has imperfections that cause this radiation to induce noise voltages into the differential signal carried between the two wires of a twisted pair. Figure A.20 shows the twisting of a twisted pair and the opposite spatial polarity of the voltage at adjacent twists. Theoretically, this twisting intro- duced by A.G. Bell himself in his 1887 patent should almost cause cance- lation of induced voltages. This is because impinging radiation would have different polarities in the adjacent segments and thus cancel itself, the reason for the twisting. Of course, the twisting is never perfect, nor is the cancelation, but twisting is better than no twisting. Many phone cus- 10American National Standard T1.413-1995, “ADSL Metallic Interface Specification,” 1995, New York, NY. Please see Issue 2, if available, T1.413-1999.

APPENDIX A 269 + - - + FIGURE A.20 Twisted-pair voltage polarities. tomers are familiar with the “flat pair” they purchase for extending phone lines within their home. This wire is not twisted, and is much more sus- ceptible to noise pickup; nonetheless, fortunately, this flat pair represents only a small segment of the total length of the phone line. Category 3 twisted pair, typically used by phone companies, has a few twists per inch. Category 5 twisted pair is a higher grade, with tighter twisting and about 100 times better rejection of noise. Category 5 twisted pair finds increasing use in new buildings (which almost always still use twisted- pair wiring) and in local area network connection. Flat pair is, of course, the worst for noise pickup. Crosstalk Noise Figure A.21 illustrates crosstalk noise, which is the noise produced by signals on other phone lines. As discussed above, several phone lines share the same cable. Typically, 25 to 50 twisted pairs are wrapped tightly in a binder group. Different twisted pairs within the binder group have different numbers of twists per inch (to prevent radiation patterns from exactly matching and offsetting the twisting pattern on other twisted Far-end H1 ( f , x) H1 ( f , d − x) receiver Near-end receiver Pair 1 NEXT FEXT Pair 2 Χ 21( f ) Crosstalking transmitter H 2( f , x) S 2( f ) x d FIGURE A.21 Crosstalk illustration.

270 APPENDIX A pairs), and there is some level of rejection caused by twisting. Nonethe- less, a signal launched from a near-end transmitter on the right in Figure A.21 will enter the cable and begin to couple into the reverse direction on another twisted pair. This type of opposite-direction crosstalk-noise cou- pling is known as NEXT, or Near-End CROSSTalk. When the insertion loss of the segment of wire between the coupling point in both directions is considered and the noise problem integrated over the total length of wire, basic physics leads to the standardized crosstalk coupling function for DSLs of [ ] .6 PSDNEXT ( f ) = N 49 ⋅ 10 –13 ⋅ f 1.5 ⋅ PSDnear − end , xmit ( f ), (1) which not coincidentally increases with the 1.5 power of frequency match- ing the decrease in balance of the twisted pair as frequency increases. The factor of N represents the number of twisted pairs in the binder expected to carry crosstalking signals. This type of noise often dominates receiver noise when it exists. For instance, the PSDs of several DSL signals appear in Figure A.22, where it is clear that the PSD often exceeds –140 dBm/Hz, especially over the frequency range of operation of the offending DSL. -90 -95 ADSL upstream NEXT -100 -105 HDSL NEXT -110 PSD (dBm/Hz) -115 -120 ISDN NEXT -125 -130 -135 -140 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Frequency (MHz) FIGURE A.22 Some worst-case crosstalk spectra for various DSL types.

APPENDIX A 271 The reader should note that while the crosstalk coupling increases with frequency, the noises plotted in Figure A.22 combine the coupling with the transmit power spectral of each of the signals, and if there is no signal at higher frequencies, then there is no crosstalk. The actual crosstalk into an individual twisted pair from one of its neighbors is not as bad as the model at all frequencies. Clearly the coupling is heavily frequency depen- dent and only at worst case exhibits the behavior of the standards model above in Equation (1). Nonetheless, this model is heavily used for conser- vative DSL design. Also in Figure A.21 is the coupling of signals from one phone line to another in the same direction, which is often called FEXT, for Far-End CROSSTalk. An exercise similar to the one for NEXT finds a standardized FEXT coupling function of [ ] .6 PSDFEXT ( f ) = N 49 ⋅ 9 × 10 −20 ⋅ d ⋅ f 2 ⋅ H ( f , d ) ⋅ PSDxmit ( f ), 2 (2) where d is the length of the line in feet and H is the insertion loss of the twisted pair. The FEXT model is similarly pessimistic and does not in- clude the highly frequency-dependent nature of real crosstalk when only one or at best a few other lines interfere in any given frequency band. The level of FEXT is usually well below NEXT and even below –140 dBm/Hz often, but at higher frequencies above 500 kHz begins to become signifi- cant and indeed dominant at yet higher frequencies because of the depen- dency on the square of the frequency. FEXT is usually of strong concern only in VDSL. Radio Noise As Figure A.23 implies, telephone lines are great radio receivers, es- pecially for AM radio broadcasts. Indeed, AM radio signals are delivered to customers on phone lines in some countries (e.g., Switzerland). AM radio signals exist in the internationally recognized band from 560 kHz to 1,600 kHz. The double-sideband-with-carrier-modulated AM signals are 10 kHz wide and likely to have PSD levels of –100 to –120 dBm/Hz on phone lines, comparable to crosstalk signals and much larger than inter- nal modem noises. It is common, if not the norm, to see 2 or 3 large AM radio signals on a phone line in the metropolitan area of any city. Thus, only a small percentage of the transmit band is disturbed, but in those bands, the disturbance cannot be ignored. The problem is particularly evident on elevated phone lines (telephone poles), but not insignificant even on buried phone lines. AM radio interference is of concern for both ADSL and VDSL but does not overlap the transmission bands for high-bit-rate DSL (HDSL) or ISDN.

272 APPENDIX A FIGURE A.23 Radio interference. Ham radio signals are an even greater problem. While smaller power is transmitted by ham radio operators, the ham antennae are distributed massively through residential environments, often being only 10 to 100m away from a phone line. The level of interference is sometimes as large as –35 dBm/Hz, and typically on the order of –50 to –60 dBm/Hz, well above the levels of any other noise type. Fortunately, ham radio signals are typically only 2.5 kHz wide, and there may likely only be one of them when such interference occurs. Ham radio signals may be transmitted in internationally recognized narrowbands from 1.8 to 2 MHz, 3.5 to 4 MHz, 7 to 7.1 MHz, 10.1 to 10.15 MHz, 14 to 14.35 MHz, and 18.068 to 18.168 MHz that overlap VDSL transmission, but not other DSLs. Impulse Noise Impulse noise is nonstationary crosstalk from temporary electromag- netic events in the vicinity of phone lines. Examples of impulse generators are as diverse as the opening of a refrigerator door (the motor turns on/ off), control voltages to elevators (phone lines in apartment buildings often run through the elevator shafts), and ringing of phones on lines sharing the same binder. Each of these effects is temporary and results in injection of noise into the phone line through the same basic mechanism as RF noise ingress, but typically at much lower frequencies. Differential (metallic) induced voltages are typically a few millivolts (mV) but can be as high as 100 mV, corresponding to levels of –50 to –70 dBm/Hz. Typical impulses last tens to hundreds of microseconds (µs) but can span time intervals as long as 3 µs.

APPENDIX A 273 Emission Constraints and PSD Masks DSLs need not only be concerned with noises generated by other electronic signals, but also with the radiation they create. The concern for such emission exceeds that normally associated with electronic equip- ment, where the FCC (in the United States) mandates certain maximum levels of radiation in various frequency bands. In the case of DSL mo- dems, the telephone line itself, while not inside the modem, does radiate, and so this type of radiation is typically limited by limiting the power spectral density of signals transmitted on phone lines. Unbundling and Standards Solutions The American National Standards Institute’s T1E1.4 group has taken a lead role in discerning problems with crosstalk between various types of DSLs, standardized and nonstandard. The idea is that if all services comply with defined spectrum masks, coexistence of different service providers’ equipment in the same cable of twisted pairs is possible with- out the transmission technique itself having to be standardized. A volun- tary ANSI spectrum management standard was issued in 2001 and of- fered to the FCC for possible use in future rulemaking. Evolution Possibilities for DSL Technology An enabling event for DSL transmission at high speeds occurred on March 10, 1993,11 when ANSI selected the discrete multitone transmission (DMT) technology for ADSL. The technology offered greater adaptivity than previous conventional technologies for transmission. The essential ingredient was an ability of the receiver and transmitter to communicate through a low-speed overhead back-channel that allowed the transmitter’s DSL spectrum and information content to adjust to each and every phone line in an individually optimized manner. The technology outperformed even the best-optimized nonadaptive methods in several independent tests (sometimes showing more than a thousandfold improvement in noise immunity), and was selected. While early modems were expensive, the standards groups successfully bet on VLSI improvements eventually making most of the gains of the DMT technology cost-effective, which has now happened, and DMT is the basis of all ADSL standards, including the recently released ITU standards “G.lite” (G.992.2) and “G.dmt DSL” (G.992.1). The additional benefit of standardization allowed economies of 11Curiously, the 100th anniversary of the invention of the telephone.

274 APPENDIX A competition, as multiple suppliers are assured of interoperable products through the use of and adherence to standards. Good DMT designs gain outstanding performance on telephone lines, and this section enumerates both present performance levels and future performance levels. This particular technology will allow a number of solutions to the unbundling and crosstalk problems mentioned above; these solutions have yet to be implemented but are simulated here to allow an understanding of future research directions in this DSL area. As phone companies increasingly deploy fiber, telephone line lengths become shorter. It is thus of interest to know the possible data rates versus line length. Three such plots are presented in the following subsections. In each, the current state-of-the-art methods are plotted, as well as a num- ber of potential enhancements that researchers have suggested will fur- ther DSL performance. ADSL and Projections Figure A.24 lists the first set of curves for DMT ADSL. The vertical axis plots data rate, while the horizontal axis plots line length in feet. As line length increases, all data rates decrease. The lowest curve shown is the performance of a good design that meets current ADSL standardized performance levels. Most current ADSL modems will not allow more than 10 Mbps maximum speed, which occurs at about 2,000 feet in the lowest curve. At about 3 miles, 1.5-Mbps speed is attainable, while a few hundred kilobits per second are possible beyond that range. ADSL uses only the lower 1.1 MHz of bandwidth on a twisted pair. A first step in improving these curves is to allow the modems to use sophisticated multiuser information-theory-based detection methods to eliminate crosstalk effects between lines. The next two curves eliminate NEXT and NEXT/FEXT, respectively. Note that when crosstalk is re- moved, a huge jump in data rate is attained—by a factor of about 3. Suddenly, 10 Mbps is possible even to a 2-mile range. A 2-mile range is a target of projects such as SBC’s Project Pronto. The so-called bit-cap is related to ADC (analog-to-digital converter) performance levels, and with improvements in such technology beyond today’s state-of-the-art conver- sion devices, additional improvement is possible. Finally, a last curve shows the improvement in performance with some of the most powerful coding methods yet found (i.e., turbo codes). VADSL and VDSL Projections Figure A.25 lists the first set of curves for what is known recently as DMT VADSL. This extension of ADSL allows up to 5 MHz of bandwidth

APPENDIX A 275 70 Current Cancel self-xtalk 60 Cancel all xtalk Increase bit cap Improve coding 50 Total Rate (Mbps) 40 30 20 10 0 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 Length (ft) FIGURE A.24 Current ADSL and projections. to be used on a twisted pair, which allows considerably higher data rates on shorter lines. Data rates jump on lines of a few thousand feet, with 20 Mbps plus possible on 3,000-ft lines (which characterize the second of the fiber termi- nation points—the so-called distribution node) with current methods. Again, with use of more sophisticated methods, another doubling in the data rate is feasible. Figure A.26 is for so-called VDSL, which uses up to 20 MHz of band- width on a twisted pair. Here, data rates on 1,000-ft loops, so-called ped- estal drops, can exceed 250 Mbps with all possible and/or known im- provements included. A last curve, “Ultra DSL,” allows an increase of transmit power to 400 milliwatts, perceived as an analog limit for phone lines with VDSL parameters. Network and Application Interfaces for DSL As discussed in the introduction to the major section “Digital Sub- scriber Line,” DSL technologies are able to achieve their high bandwidths

276 APPENDIX A 60 Current Cancel self-xtalk Cancel all xtalk 50 Increase bit cap Improve coding Total Rate (Mbps) 40 30 20 10 0 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 Length (ft) FIGURE A.25 VADSL and projections for 5 MHz bandwidth use. 300 Current Cancel self-xtalk 250 Cancel all xtalk Increase bit cap Improve coding 200 Ultra-DSL Total Rate (Mbps) 150 100 50 0 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 Length (ft) FIGURE A.26 VDSL performance and projections with 20 MHz of bandwidth.

APPENDIX A 277 over the copper loop because they avoid the existing telephony network of switches and transmission channels optimized for voice traffic. In a sense, the existing non-DSL modem technology “creatively abuses” the physical path that is established from point to point for carrying voice traffic by placing the data on a path designed to carry voice. This has the advantage of allowing for worldwide switched connectivity of data. How- ever, it also limits the bandwidth of the connection to, at most, 64 kbps. This is the rate supported by the digital voice channels within and be- tween the switches of the worldwide transmission network. Providing a DSL-based physical path on a copper loop will allow a user to transmit at high data rates over that loop. However, DSL will get the data only to the end of the copper in the central office. To reach destinations desired by the DSL user, a high-bandwidth data network must be provided from the CO to these remote sites. Telecommunications carriers thus face new problems in constructing and managing a network that is intimately involved in data networking issues. In the current environment, they need only provide the physical- layer connectivity (using either the switched network or a private dedi- cated line for higher-bandwidth connections). They are isolated from the details of their users’ data networks. In the case of the voice network, it is as if the telco provides trains from place to place and does not care what type of car is placed on the trains or what is in the boxcars. In providing DSL services, the telephone companies must provide a network that in- teracts directly with their users’ networks and protocols. The telco now needs not only to operate the tracks and trains but also to provide box- cars, tank cars, and transshipment between trains, ships, and planes. The carrier needs to address the following issues if it is to provide DSL services to its customers: • It must provide a data network connecting the ADSL terminated copper loops to the service providers desired by customers. Examples of service providers include the public Internet, private corporate networks, interactive video services, or highly interactive games. • It must provide cost-effective interfaces between its network and the service providers. • The data protocols that connect the customer to service providers over the ADSL network must be compatible with existing technologies and procedures used by both customers and service providers and must also support the high-bandwidth services provided by ADSL. • The carrier must develop methods to manage this new network. The carrier must be able to add new customers, repair problems, and bill for the services provided.

278 APPENDIX A Although there are many potential implementations that can address these issues, a discussion of two contrasting environments and potential solutions can illustrate the current direction of ADSL end-to-end architec- tures: 1. The large common carrier environment in which a telecommunica- tions provider offers services to masses of customers and supports many different service providers; 2. Use of DSL in very specialized environments such as college cam- puses, military bases, or condominium apartments. The Large Common Carrier A large local exchange telephone company may support more than 2 million telephone lines in a major metropolitan area. Even a 10 percent penetration of ADSL results in 200,000 ADSL customers in the area. For both regulatory and business reasons, the carrier will need to provide common-carrier access between users and any service provider who pays to connect to the access network. A user of the ADSL service may wish to connect to any service provider it chooses. Simultaneous connections from a user to multiple service providers are likely to be required. Other users may connect to only one service provider but require that the connection be physically secure. For example, a remote office might use the ADSL service to connect to a central corporate LAN. Although large-scale ADSL service has yet to be deployed by any carrier, many carriers are converging toward a common architecture. Each central office is served by one or more DSLAMs,12 which terminate the ADSL line in the central office. POTS voice traffic is also carried on each loop. Splitters groom the POTS traffic, carried at frequencies below 4 kHz, before the loop is terminated on a port on the DSLAM. The POTS traffic is placed on a voice switch in the carrier’s legacy voice network. The ADSL connection only provides a physical layer between the customer’s ADSL modem and the modem in the CO within the DSLAM. In this architecture, ATM provides a data link layer end-to-end between the users’ computing environment and the service provider’s network. An ATM virtual circuit is established between the user’s ADSL modem and the interface between the carrier’s ATM network and the service provider’s network. 12DSLAM is an acronym for Digital Subscriber Line Access Multiplexer. The term is now never translated but has become a generic term for CO-based devices that terminate DSL loops.

APPENDIX A 279 The DSLAM in the CO terminates many ADSL lines (typically in the range of 200 to 600 ADSL ports per DSLAM) and concentrates the traffic to and from the customers over DS3 (48-Mbps) or OC3 (155-Mbps) trunks that connect to ATM switches in the carrier’s ATM data network. The service providers are also connected to the carrier’s ATM network via similar high-speed trunks. The use of ATM allows for scalability of the service to support hundreds of thousands of users in a metropolitan area and a wide range of potential future ADSL services. If the service provider and customer communicate using IP, the use of PPP (point-to-point protocol) over ATM, as defined in Kwok et al.,13 will allow both the user and the service provider to operate in an environment that is similar to that provided by dial-up modems today. In that environ- ment, the IP is placed in PPP that is carried over the voice network be- tween the users’ and service providers’ modems. By using PPP over ATM, the carrier can isolate both the service provider and user from the com- plexities of both ADSL and ATM. The Specialized Carrier In contrast to the large public carrier, many specialized deployments of ADSL are possible. Any organization that has access to copper loop can deploy ADSL. For example, the owner of a large rental apartment build- ing may install an ADSL service to provide high-speed Internet access to the tenants. A competitive local exchange carrier (CLEC) can, under the terms of the Telecommunications Act of 1996, lease copper loops from the incumbent phone company (the incumbent local exchange carrier, or ILEC) and serve customers with ADSL. The largest CLECs may end up resembling ILECs in both scale and architecture of their ADSL services. However, in many cases the deploy- ment will be small and will have similar requirements to the small ADSL implementation in the apartment building. Other examples of small ADSL deployments could include college campuses (which typically own their own copper loop plant), hotels, or military bases. The requirements for these “niche” deployments of ADSL include these: small scale, that is, from 100 to 1,000 users total; limited need to support multiple providers; and the service provider and carrier may be identical. In the case of an ADSL-equipped apartment building, the cus- tomers are connected directly to the ISP contracted to provide service to 13Timothy Kwok et al. 1997. “An Interoperable End-to-End Broadband Service Architec- ture over ADSL Systems (Version 3.0),” ADSL Forum Contribution 97-215.

280 APPENDIX A the building. In the case of a CLEC, the CLEC may itself be an ISP. In the case of a campus, the users will be connected directly to the university or company’s data network. The DSLAM containing the ATU-Cs is either in close proximity to or integrated with an IP router. The router is connected directly to the IP network of the ISP or corporate network. It is managed as an integral part of that network. The ADSL connections to the user support IP over HDLC directly on the ADSL physical layer.14 The ADSL user appears as host directly on the service provider’s IP network. WIRELESS Introduction Broadband wireless access is frequently mentioned as an important alternative to wired technologies, namely, to DSL, cable, and fiber. Wire- less has always played an important role in telecommunications networks because of its inherent advantages of modest infrastructure investment (no wires!), rapid service deployment, and end-user mobility support. The strategic significance of wireless communication services has in- creased over the past decade as cellular-telephony subscriber growth con- tinues to outpace all earlier projections. It is now clear that wireless will continue to play an important role in emerging telecommunications ser- vices, including narrowband data and broadband services because of the same intrinsic advantages. For both Internet and broadband services, wireless services have experienced a larger-than-expected gestation pe- riod owing to a combination of factors such as technology cost and perfor- mance problems, spectrum regulation barriers, and weak standards. How- ever, wireless data and broadband Internet services seem poised for technical and market breakthroughs over the next 3 to 5 years, and should thus provide an important alternative for facilitation of broadband ser- vices in the United States and other parts of the world. Service Concept The concept of a wireless broadband access network is shown in Fig- ure A.27. The basic idea is to provide a high-speed wireless link between subscriber devices such as PCs, Internet appliances, PDAs, and new per- 14ADSL Forum. 1997. Framing and Encapsulation Standards for ADSL: Packet Mode. ADSL Forum Technical Recommendation-003, June.

APPENDIX A 281 Growing proportion of user terminals → 50% +? Future Telecom Mobile communications Network devices Mobile PDA/PIA Broadband Wireless Access Network Semimobile Future laptop, etc. Internet Fixed PC/WS FIGURE A.27 Broadband wireless network service concept. sonal multimedia devices (both fixed and mobile). It is expected that ini- tial applications of broadband wireless access will start with fixed devices such as home PCs connecting to an ISP, with a gradual migration toward mobile applications as end-user devices become more and more portable. Thus, the initial impetus for wireless access comes from the need to rap- idly deploy networks capable of supporting high-speed Internet access. With increasing investment in next-generation wired networks, it may be expected that (at least in developed countries), the focus for wireless sys- tems will shift toward semimobile or mobile services, given that an in- creasing percentage of computing platforms will become inherently por- table. It is noted that the wireless access network shown in Figure A.27 may be expected to interface with both the future public telephony net- work and the Internet, which are themselves experiencing some degree of convergence as they migrate toward broadband services. This means that wireless access networks for future broadband systems are likely to be architecturally aligned with protocols used in fixed networks, rather than being designed as custom overlay solutions (as is the case with today’s cellular networks). In the long term, this may be expected to drive conver- gence between fixed and wireless networks further, where the same ser- vice is offered to both wired and wireless devices in a seamless manner.

282 APPENDIX A Technology Overview As mentioned above, adoption of wireless data services has been inhibited by a rather slow improvement in wireless technology cost and performance during the past decade or so. The previous generation of wireless data technologies (including cellular modems, wide-area data, satellite data, and wireless LANs) fell far behind Moore’s law improve- ments experienced by most computing and telecommunications technolo- gies. While wireless does pose important technical challenges, there ap- pears to be no fundamental reason for this discrepancy, which was probably caused by insufficient R&D and/or venture funding needed to drive this area. The situation has been largely corrected during the last 2 or 3 years, during which various new broadband wireless technologies have emerged as competitive options to wired solutions. This is illus- trated in Figure A.28, which shows the evolution of wireless technology performance over the last decade. As shown in the figure, newer commer- cial or precommercial wireless technologies have reached the Mbps+ bit- rate levels necessary for viable broadband services, either fixed or mobile. Figure A.29 shows the typical bit-rate and mobility regimes for vari- ous broadband wireless networks currently under consideration. The fig- ure shows the relative roles of fixed wireless access, high-speed wireless LAN, semimobile broadband PCS, and wideband cellular (IMT-2000). Although these services are generally viewed as distinct (and may be worked on by different technical and business communities), changes in technology are likely to result in new service models that merge one or more of the existing categories. In the United States, broadband wireless services are likely to start out with fixed access to residences and gradu- ally evolve towards portability and mobility. The reverse is likely to hap- pen in Europe and Japan, where so-called 3G mobile services are expected to appear on the market within the next few years, and may later be applied to broadband wireless local loop (WLL) scenarios. The generic architecture of a broadband wireless network consists of the following major components: radio modem (physical layer), radio link protocol (RLP), and fixed infrastructure network with capabilities for supporting wireless/mobile services. The following subsections summa- rize key technology issues related to each of these major subsystems. Physical Layer Broadband wireless networks require physical-layer bit rates that are orders of magnitude higher than those for current digital cellular or WLL systems, i.e., 10 to 100 Mbps versus the current 10 to 100 kbps. The higher bit rates must be achieved without introducing line-of-sight (LOS) con-

UNII Gbps 1000 1000 1000 1000 1000 WLAN, etc. Router DSL ATM Cable 100 100 100 100 100 Modem Local 3G Mobile Access Wireless Local LAN/WAN CPU Memory access access switching speed size CPU 56K modem Sw Ethernet Wireless 10 10 10 10 10 LAN/WAN CDPD kbps kbps Mbps MHz MB Memory 1 1 1 1 1 1990 1995 1999 FIGURE A.28 Evolution of wireless technology performance, 1990-1999. 283

284 APPENDIX A 10 Mbps+ services Application regime Fixed - Moderate Microwave for broadband Mobility 100 Mbps wireless Packet Data + Broadband Voice/Video 10 Mbps High-speed Broadband WLAN PCS UMTS/ 1 Mbps Wireless IMT-2000 LAN LEOS, etc. 100 kbps PCS Cellular 10 kbps Mobility FIGURE A.29 Broadband wireless service scenarios in terms of mobility versus bit rate. straints, thus indicating a need for robust modulation techniques that work well for non-LOS channels with fading. At the same time, spectrum limitations imply the need for significantly higher spectrum usage effi- ciency (bps/Hz/unit area), that is, 5 to 10 bps/Hz/cell versus the current 0.5 to 1 bps/Hz/cell. Clearly, if broadband wireless services are to reach significant penetration, cell sizes will have to be relatively small (~1- to 5- km radius), capable of providing, say, 100 Mbps to 1 Gbps of data through- put per square kilometer for frequency allocations of a few hundred mega- hertz. The above order-of-magnitude improvements can indeed be achieved via a combination of technology improvements, including these: • High-speed (Mbps+) radio modems based on advanced signal-processing techniques, such as equalization, spread spectrum, multicarrier modulation, spa- tial processing, and smart antennas. Examples include equalized QAM, used in several first-generation fixed wireless systems; equalized VSB in the U.S. terrestrial HDTV standard; wideband direct-sequence CDMA, under consideration for the 3G mobile (IMT-2000) standard; and OFDM, pro- posed for various fixed and semimobile scenarios (including ETSI Hiper- lan II and several proprietary WLL systems, such as Clarity Wireless/ Cisco). Spatial processing with multiple antennas, mentioned above, is a new dimension for improving modem performance, and has recently been proposed by several independent groups (Bell Labs, Stanford University, Iospan) as a means for dramatically improving both non-LOS coverage and spectral efficiency. All of these technologies are maturing rapidly,

APPENDIX A 285 and it may be expected that commercial products will soon deliver 10 Mps+ services with cellular reuse and spectral efficiency on the order of 5 bps/Hz/sector. With continuing advances in signal processing, achievable bit rates should increase to 100 Mbps+ with spectral efficiency of 10 bps/Hz over the next 5 years. • Cellular technology capable of scaling to small cells and multiple sectors necessary for effective coverage of areas with higher population density. Scaling of broadband wireless services to small cells is inevitable in areas with higher population density, where throughputs on the order of 100 Mbps to 1 Gbps/km2 must be achieved in order to serve even a modest fraction of the population. Efficient cellular reuse implies the need for modem technology that can operate at relatively low carrier-to-interference (C/I) ratios. This can be achieved by a suitable combination of time/frequency/ space processing. For example, spread spectrum achieves high spatial reuse via time-frequency processing, while multiple antenna spatial pro- cessing OFDM modems do so using frequency-space processing. Wide- band CDMA adopted for IMT-2000 radio access achieves a spatial reuse factor of 1:1 using spread spectrum and interference cancellation tech- niques. However, net throughput per square kilometer is limited by the relatively low ~0.5-bps/Hz efficiency of spread spectrum modulation. Spatial processing techniques mentioned earlier have the potential for achieving spectral efficiencies on the order of 5 to 10 bps/Hz/cell with ~1:3 spatial reuse. Further gains can be achieved for both CDMA and spatial processing OFDM with directional remote antennas and base sta- tion sectorization. • Spectrum regulation and management policies that facilitate rapid de- ployment of broadband services, while promoting efficient use. The pace of wireless network deployment is critically dependent on spectrum regula- tion policies, both international and domestic. Historically, the process of frequency allocation has been rather slow, with the United States and to some extent the European Union taking the lead in introducing both spec- trum auctions and unlicensed bands in order to stimulate efficient eco- nomic usage. While one-time spectrum auctions in the United States have had their intended effect (e.g., PCS and MMDS bands), it may be time to consider introducing more dynamic market mechanisms that allow spec- trum to change hands in time-constants of minutes or hours rather than months or years. For example, it may be possible to establish an online commodity trading system for spectrum that would permit operators with higher economic utility to bid for their peak usage needs without having to go through a lengthy procurement process. Rapid deployment of wireless services would be further facilitated by streamlined approval processes for a wider range of customer equipment,

286 APPENDIX A including those with higher-powered directive antennas. This would probably require further advances in antenna beam and power control, but should be technically feasible in the near term. The broadband WLL business model depends to a large extent on user-installable or self-con- figuring customer premises equipment (CPE), something that would re- quire some relaxation of current rules in MMDS and other fixed access bands. In addition, it may be expected that fixed access will gradually migrate toward semimobile services as cell sizes become smaller, further increasing the need for simple approval policies. Unlicensed spectrum, such as the 5-GHz unlicensed national infor- mation infrastructure (U-NII) band in the United States, is an important facilitator for broadband access. FCC’s allocation of the U-NII spectrum has stimulated considerable commercial activity in the high-speed wire- less LAN area. It is recognized that the same type of technology (perhaps with somewhat higher power levels and larger coverage areas) could be used as a broadband PCS access network for public semimobile services in urban and suburban communities. There is, however, one remaining technical problem—that of spectrum etiquette—a decision on which was deferred by the FCC in its initial U-NII ruling. The problem is that exist- ing unlicensed band etiquettes such as listen-before-talk (LBT) do not work well for stream services with quality-of-service (QoS) requirements. In such cases, the etiquette must be designed for equitable sharing among contending stream users, without reducing all of them to an unacceptable QoS level. The FCC has invited the industry to propose a suitable eti- quette, but a specific scheme has yet to be identified. A possible technical solution is to introduce a common spectrum coordination channel at the edge of each unlicensed band and require users to execute mutually agreed sharing procedures (priority, dynamic auction, and so on) using a standardized etiquette protocol. Radio Link Protocol Broadband wireless access requires a new type of radio link protocol (RLP) capable of reliably transporting both packets and media streams with specified QoS. The broadband RLP itself may be decomposed into a medium access control (MAC) layer for channel sharing among multiple subscribers, and a data link control (DLC) protocol for error recovery. Broadband wireless networks tend to use either a packet CDMA, dy- namic TDMA type, or an extended 802.11 carrier sense, multiple access/ collision avoidance (CSMA/CA) MAC protocol. CDMA is the basis for the emerging IMT-2000 wideband CDMA standard for 3G mobile, and is associated with the choice of spread spectrum modulation believed to be appropriate for vehicular mobile systems. Dynamic TDMA has generally

APPENDIX A 287 been adopted for broadband applications, as well as for some high-speed LANs (such as wireless ATM and the European Telecommunications Stan- dards Institute’s broadband radio access networks) in view of its ability to support a combination of packet data and constant bit-rate streams (voice and video). Extended 802.11 protocols provide streaming extensions for QoS support, and may be suitable for Ethernet-equivalent WLAN sce- narios. DOCSIS MAC protocols used in cable networks have also been modified for WLL applications, but will generally incur a performance penalty owing to large packet sizes. Data-link-layer retransmission for error recovery is an essential fea- ture for broadband wireless service, since higher-layer protocols are criti- cally dependent on low packet error rates on each link of the end-to-end connections. DLC involves fragmentation of data packets into relatively small units, the optimum for which is typically between 40 and 200 bytes, depending on the channel and traffic model. Many current implementa- tions have adopted the ATM cell payload of 48 bytes as the basic unit of fragmentation on the radio link. This has the advantage of simplifying the interface to ATM backhaul networks, which are often used in carrier broadband and DSL networks. Error control on the radio link involves the addition of a wireless link header containing a sequence number used for identifying data units to be retransmitted. Implementation results have shown that significant improvements in end-to-end protocol performance (typically 2 orders of magnitude in packet error rate) can be achieved with fragmentation and retransmission on the radio link. This in turn permits wireless systems to operate in a higher C/I environment, thus increasing overall capacity of cellular networks. Infrastructure Network Broadband wireless access links are being designed as “plug-ins” to existing fixed network architectures based on IP and/or ATM. In order to facilitate ubiquitous deployment, it is important that both fixed WLL and mobile access be easily integrated with broadband DSL and cable net- works currently being deployed. This means that the radio air application programming interface should be harmonized with both IP and ATM to the extent possible, particularly in terms of providing generic parameters for service establishment and QoS control. For fixed wireless access, inter- face functions specific to the radio link are performed by the base station, which puts out standard IP and/or ATM data and control into the infra- structure network. For mobile scenarios, services (such as location management and handoff) specific to mobile users may be provided either with a mobility overlay, used in current cellular systems, or by integrating mobility sup-

288 APPENDIX A port into the core network protocols, such as IP or ATM. The latter method (i.e., support integrated mobility in IP or ATM) is the preferred method for broadband in view of performance and scalability requirements. More- over, as an increasing proportion of user devices becomes portable, the distinction between fixed and mobile user addresses will become more difficult to administer (the integrated approach does not require a priori partitioning of mobile and fixed addresses). Protocol specification work aimed at integrating mobility support into IP and ATM has been done in both the Internet Engineering Task Force (IETF) (mobile IP) and the ATM Forum. While much further work remains (3G.IP and so on), it may be expected that mobility will increasingly be integrated as a standard fea- ture into fixed network infrastructures. Ultimately, this technical direc- tion will further accelerate fixed and wireless network convergence, which has been predicted for some time. MEDIA COMPRESSION Media signals include (digital) data as well as analog information and entertainment signals: speech, audio, image, video, graphics, and other audiovisual signals such as hand gestures and handwriting. These signal classes are universal and are representative of most if not all information that needs to traverse the first mile, in either direction. Complementary Roles of Modems and Compression Systems (Codecs) Modem and access technologies have evolved to expand the trans- mission pipe for conveying digital information. In parallel, compression technology has evolved to compact the amount of digital information that is needed to convey the information in a signal with a specified level of fidelity. Access speeds have generally advanced on a faster track than has compression technology. That said, it is the combination of faster mo- dems and greater levels of compression that has enabled advances and revolutions in digital communication. This section focuses on the impact of media compression as a direct enabler of digital communication over channels and networks with limited capacity. Computing is an overarching enabler of multimedia communications, whether one is implementing coders or decoders (codecs for short) or is implementing modulators or demodulators (modems for short). Moore’s law has direct implications on the rate at which computing technology (memory and arithmetic capability) advances as a function of time. In this view, advances in computing are much more rapid than are advances in access technologies. That said, advances in computing will only help

APPENDIX A 289 speed up advances in access, but these advances are strictly knowledge- or algorithm-limited, as are the advances in compression. The Dimensions of Performance in Media Compression There are four dimensions of performance in a compression system: (1) quality, (2) bit rate, (3) delay, and (4) complexity. “Quality” refers to the quality of the signal after compression, measured in absolute terms or in terms of closeness to the original version of it. The “bit rate” is the data rate after compression. The “delay” is the sum of delays in the encoding and decoding parts of the system: the compression and decompression algorithms. (This does not include delay components resulting from spe- cific implementation details or specific transmission latencies in the com- munication of the encoder bit stream.) Finally, “complexity” refers to the computational effort needed to perform the compression and decompres- sion algorithms, measured for example, in millions of instructions per second (mips) and kilobytes (the read-only and random-access memories used in the codec [coder-decoder]). As processing technology improves, the importance of the complexity parameter tends to diminish, but delay remains as a fundamental performance metric. Delay is particularly im- portant in interactive, or two-way, communications. The Fuzzy Fifth Dimension: Richness of Content In studies of compression efficiency, where one measures quality deg- radation as a function of increasing levels of compression, one assumes that the bandwidth, or frequency content in the signal, is a prespecified characteristic. For example, telephony is always associated with a speech signal of 4-kHz bandwidth, and television with a signal whose effective horizontal and vertical resolutions are on the order of a few hundred pixels in each case (a total number of pixels per frame on the order of 100,000). In Table A.2, the notation of pps refers to pixels per second. It is the product [H × V × F] of horizontal resolution (H pixels per row), vertical resolution (V pixels per column), and temporal resolution (F frames per second). With the evolution of flexible and scalable communications technol- ogy, one often has the option of considering input signals of higher band- width, as long as the compression is strong enough to delimit the output data rate to a specified number. Examples are high-bandwidth audio (such as FM-grade speech with 12- to 15-kHz bandwidth or CD-grade music with 20-kHz bandwidth, or multichannel sound) and high-definition tele- vision (a total number of pixels on the order of 2 million, 60 frames per second).

290 APPENDIX A TABLE A.2 Multimedia Formats Format Sampling Rate Frequency Band Telephone 8 kHz 200-3,400 Hz Teleconference 16 kHz 50-7,000 Hz Compact disk 44.1 kHz 20-20,000 Hz Digital audio tape 48 kHz 20-20,000 Hz CIF Video 3 Mpps [360 × 288 × 30] CCIR Video 12 Mpps [720 × 576 × 30] HDTV 60 Mpps [1,280 × 720 × 60] NOTE: Mpps = megapixels per second. SOURCE: Nikil Jayant, 1993, “High Quality Networking of Audio-Visual Information,” IEEE Communications Magazine 31(9). Scalability in bandwidth is a somewhat fuzzy situation in that users are often not conditioned to the continuum in this parameter between, or beyond, well-established anchors. For example, wideband speech is a fuzzy term that implies any bandwidth in the range between well-defined telephone and CD grades (4 and 20 kHz), and first-generation Internet video often is understood to mean anything that is usable, albeit below TV quality (such as 10,000 to 100,000 pixels per frame). The video situa- tion has the additional dimensions of viewing distance, physical picture size, and fractional-screen displays, which further control user apprecia- tion of picture quality or user perception of picture degradation. The Algorithms of Media Compression The description of compression algorithms is beyond the scope of this appendix. It is also not needed for the purposes of this report. What is important, however, is to note that all compression algorithms are based on only two basic principles: removal of redundancy in the input signal, and the reduction of irrelevancy in it. “Redundancy” is usually character- ized in a statistical fashion, while “irrelevancy” is best linked to a percep- tual criterion. Compression techniques are also usefully classified into three types: (1) lossy, (2) lossless, and (3) perceptually lossless. Math- ematically lossless compression is used in some archival, legal, and medi- cal applications, while perceptual losslessness is a pragmatic criterion for a large class of applications in transmission and storage. Most compres- sion standards tend to address this criterion. Other characteristics to keep in mind are the delay and complexity of the algorithms, and how they are

APPENDIX A 291 distributed between the compression and decompression parts of the sys- tem. For example, interactive and two-way applications look for low- delay compression, servers can typically afford high complexity, and cli- ent systems need to be relatively simple to implement. Implementation platforms can be ASIC (application-specific integrated circuit), DSP (digi- tal signal processor), or NSP (native signal processor, as on a Pentium). As a matter of calibration, a Pentium II (400-MHz) processor can decode MPEG1 video streams in real time, and a pocket PC in 2001 has a proces- sor that works at half the speed (about 200 MHz). Compression Standards Tables A.3 and A.4 provide nonexhaustive lists of compression stan- dards for audiovisual signals. In general, results refer to lossy compres- sion, although these standards include special functions for lossless com- pression. For example, in JPEG image compression, there is a lossy (perceptually lossless) version with typical bit rates of 0.5 to 2 bits per pixel (bpp), while a mathematically lossless version may use a bit rate of 5 to 6 bpp. In Figure A.30, the horizontal axis displays bit rates after compression for classes of applications that are arranged in clusters that represent speech, audio, and image applications. The bit rates range from 1 kbps to 100 Mbps. Interestingly, the geometric mean of this range is 300 kbps, a number typical of conservative ADSL and cable modem rates in the year 2000. The data rates in Figure A.30 are strict lower bounds in the sense that in most applications, the compressed information needs to be supple- mented with ancillary data. Bit Error Protection In a rate k/n error correction code, k information bits are protected for transmission over an error-prone channel by adding (n-k) redundant bits. The fractional overhead is 1-k/n. Sophisticated methods of error protection include these: • Unequal error protection, in which different parts of the compressor output receive different levels of error protection, depending on models of their relative perceptual importance; • Joint source and channel coding, in which, for example, the total bit rate available is shared dynamically between source bits and error protec- tion bits, depending on the model or knowledge of the channel state.

292 APPENDIX A TABLE A.3 Standards for Speech Compression Standard (year) Algorithm Bit Rate Application G.722 (1988) Subband 64, 56, 48 kbps Teleconferencing ADPCM MPEG-1 (1992) Musicam 384, 256, Two-channel audio ASPEC 128 kbps w/video on CD MPEG-2 (1996) PAC 320 kbps Five-channel surround sound for multimedia recording DAB (1996) PAC 160 kbps Two-channel audio for terrestrial broadcast JBIG (1991) Run length 0.05-0.1 bpp Binary coded images coding (half-tone) JPEG (1991) DCT 0.25-8 bpp Still continuous-tone images MPEG-1/2 MD-DCT 1-8 Mbps Addressable video on CD (1991, 1994) P × 64 (1991) MC-DCT 64-1,536 kbps Videoconferencing HDTV (1996) MD-DCT 17 Mbps Advanced TV G.711 (1972) Mu-Law and 56-64 kbps Network transmission A-Law PCM G.721 (1984, 1987) ADPCM 32 kbps Bit-rate multiplexers, undersea cable G.723 (1988) ADPCM 24, 40 kbps Overload on undersea cable, data modem G.726/G.727 ADPCM 16, 24, 32, High overload rate for 40 kbps undersea cable G.728 (1992) LD-CELP 16 kbps Transmission at low delay G.729 (1995) ACELP 8 kbps Second-generation digital cellular G.723 (1995) ACELP 6.3, 5.3 kbps Low-bit-rate videophone GSM (1987) RPE-TLP 13 kbps European digital cellular full-rate IS-54 (1989) VSELP 8 kbps North American digital cellular-TDMA IS-96 (1993) QCELP 8.5, 4, 2, North American digital 0.8 kbps cellular-CDMA GSM-1/2 (1994) VSELP 5.6 kbps European digital cellular half-rate EVRC (1996) RCELP 8.5, 4, 0.8 kbps NA CDMA, 2nd generation IS-136 (1995) CELP 8 kbps NA TDMA, 2nd generation JDC (1989, 1992) VSELP 8, 4 kbps Japanese digital cellular— full and half rates FS-1016 (1975) CELP 4.8 kbps Secure telephony—full rate FS-1015 (1975) LPC-10E 2.4 kbps Secure telephony—half rate FS-1015 (1996) 2.4 kbps Secure telephony—half rate, 2nd generation SOURCE: After R.V. Cox. 1999. “Current Methods of Speech Coding,” in N. Jayant (ed.), 1999, Signal Compression: Coding of Speech, Audio, Image and Video, World Scientific, Singapore.

APPENDIX A 293 TABLE A.4 Standards for Audio, Image, and Video Compression Standard (Year) Algorithm Bit Rate Application G.722 (1988) Subband ADPCM 64, 56, 48 kbps Teleconferencing MPEG-1 (1992) Musicam/ASPEC 384, 256, 128 kbps Two-channel audio w/ video on CD MPEG-2 (1996) PAC 320 kbps Five-channel surround sound for MM recording DAB (1996) PAC 160 kbps Two-channel audio for terrestrial broadcast JBIG (1991) Run length coding 0.05-0.1 bpp Binary coded images (half-tone) JPEG (1991) DCT 0.25-8 bpp Still continuous-tone images MPEG-1/2 (1991, 1994) MC-DCT 1-8 Mbps Addressable video on CD P × 64 (1991) MC-DCT 64-1,536 kbps Videoconferencing HDTV (1996) MC-DCT 17 Mbps Advanced TV NOTES: kbps = kilobits per second; bpp = bits per pixel. SOURCE: After R.V. Cox. 1999. “Current Methods of Speech Coding,” in N. Jayant (ed.), 1999, Signal Compression: Coding of Speech, Audio, Image and Video, World Scientific, Singapore. MUSIC PREVIEW AND HDTV BROADCAST AUDIO DIGITAL TELEVISION CONFERENCE NETWORK MOVIES ON TELEPHONY COMPACT DISK CELLULAR VIDEO CONFERENCE RADIO HIGH-RESOLUTION VOICEMAIL FACSIMILE SECURE IMAGE- SLIDE VOICE PHONE SHOW 1 2 4 8 16 32 64 128 512 1 2 8 32 KILOBITS PER SECOND MEGABITS PER SECOND FIGURE A.30 Data rates in digital representations of signals. Rates are numbers after compression. SOURCE: After R.V. Cox. 1999. “Current Methods of Speech Coding,” in N. Jayant (ed.), 1999, Signal Compression: Coding of Speech, Audio, Im- age and Video, World Scientific, Singapore.

294 APPENDIX A Resiliency to Packet Losses Packet networks are often limited by packet losses rather than bit errors. Packet losses can be addressed by retransmission in delay-insensi- tive applications. In delay-sensitive communications, packet losses can be anticipated by redundancy in the packet generator. In sophisticated algorithms, such as embedded coding and multiple description coding, this redundancy is contained by unequal protection of subpackets, depending on models of perceptual importance of these subpackets, as in unequal bit error protection. Information Hiding, Steganography, Watermarking, and Multimedia Annotations Increasingly, digital communications will include ancillary informa- tion that may convey a variety of information to the end user that is related to authentication (information about sender and intended receiver, for example), and such information is embedded in the main message in an unobtrusive and imperceptible form. These are the techniques of infor- mation hiding, with subclasses called steganography and watermarking. Multimedia annotations also involve additional data, but not necessarily in imperceptible or hidden form. The overall effect of all of the above processes is that the data rate for digital communication is strictly higher than the data rates at the output of the signal compression stage. While there is no rigorous way of mea- suring the resulting overhead in data rate without regard to the applica- tion and the needs of it, it is useful to use the following guideline: Typical overall overheads are in the range of 10 to 100 percent, and the rates on the horizontal axis of the compression chart in Table A.4 need to be increased by factors as high as 2.0, especially in the case of un- friendly access methods such as wireless links that are power- and inter- ference-limited and/or in networks that are operating in situations of overload. In scalable media communications, the inherent excursions in data rate in the compression algorithm can well exceed the factor of 2 referred to above. In these cases, the metrics of importance are the average data rate in the scalable compression algorithm, and, where available and us- able, more detailed descriptions of the data rate histogram. In fact, assess- ments of traffic and channel loading depend directly on these difficult and highly variable characterizations of the information source. The least complex nontrivial measure of overall data rate is the average data rate after compression, multiplied by the overhead mentioned in the guide- line above.

APPENDIX A 295 Media-Specific Examples: Access Implications and Questions • Toll-quality telephony versus Internet or cable modem telephony. What are the quality and delay targets in IP-telephony and cable telephony? What are the consumer expectations? Is there a business case for AM- radio-grade telephony? What is the competitive landscape? • Audio/video streaming at lite-ADSL, cable modem, and wireless speeds. Are user expectations going to be tied to television quality? What is the longevity of partial-screen solutions? What is the competitive landscape? • Uploading of information from a home. What are the primary cases for upload-speed on demand? What are the demands of such applications as telemedicine, teleworking, home publishing? Is there a case for sym- metric uplink and downlink? Definitive answers to these questions do not exist, but as applications mature, it will be possible to understand and quantify them at least im- plicitly and qualitatively. Research and Technology Outlook At this time, compression technologies are mature. Although it is difficult to define the fundamental limits in the game, typical data rates for specified levels of quality are generally known. Increasing compres- sion ratios will become the preoccupation of specialists. Likewise, decod- ers and clients will become pervasive and affordable. New advances in first mile and first meters multimedia communications will depend in- creasingly on advances in access speed and on innovations in network- ing.

Next: B A Brief History of Telecommunications Regulation »
Broadband: Bringing Home the Bits Get This Book
×
Buy Paperback | $48.00 Buy Ebook | $38.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Broadband communication expands our opportunities for entertainment, e-commerce and work at home, health care, education, and even e-government. It can make the Internet more useful to more people. But it all hinges on higher capacity in the "first mile" or "last mile" that connects the user to the larger communications network. That connection is often adequate for large organizations such as universities or corporations, but enhanced connections to homes are needed to reap the full social and economic promise.

Broadband: Bringing Home the Bits provides a contemporary snapshot of technologies, strategies, and policies for improving our communications and information infrastructure. It explores the potential benefits of broadband, existing and projected demand, progress and failures in deployment, competition in the broadband industry, and costs and who pays them. Explanations of broadband's alphabet soup – HFC, DSL, FTTH, and all the rest – are included as well. The report's finding and recommendations address regulation, the roles of communities, needed research, and other aspects, including implications for the Telecommunications Act of 1996.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!