电子工程师手册（英文版四）：ch075

Lee, W.C.Y., Ziemer, R.E., Ovan, M. Mandyam, G.D. “Personal and Office” The Electrical Engineering Handbook Ed. Richard C. Dorf Boca Raton: CRC Press LLC, 2000 75 Personal and Office 75.1 Mobile Radio and Cellular Communications The Difference between Fixed-to-Fixed Radio Communication and Mobile Communication ? Natural Problems in Mobile Radio Communications ? Description of Mobile Radio Systems ? Mobile Data Systems ? Personal Communication Service Systems 75.2 Facsimile Scanning ? Encoding ? Modulation and Transmission ? Demodulation and Decoding ? Recording ? Personal Computer Facsimile ? Group 4 Facsimile 75.3 Wireless Local-Area Networks for the 1990s The Wireless In-Building Vision ? Market Research ? LANMarket Factors ? Cabling Problems ? User Requirements Environment ? Product Requirements: End User Reaction ? Technology Alternatives in Meeting Customer Requirements 75.4 Wireless PCS Cellular Band Systems ? PCS Services ? 3rd Generation Enhancements 75.1 Mobile Radio and Cellular Communications William C. Y. Lee The Difference between Fixed-to-Fixed Radio Communication and Mobile Communication In fixed-to-fixed radio communications, the transmitter power, antenna location, antenna height, and antenna gain can be determined after calculating the link budget. Also, depending on the frequency range of the carrier affected on the atmospheric variation, different “margin” values will be put in the budget calculation for different system applications. The fixed-to-fixed radio links are usually 10 miles or longer and high above the ground. The signal variation over the link is due mostly to atmospheric changes. Satellite communications, microwave links, troposcatter, etc. are fixed-to-fixed radio communications. In mobile radio communications, the param- eters such as transmitter power, antenna location, antenna height, and antenna gain are determined by covering an area or cell. In mobile radio communications, at least one end is in motion. The sizes of cells in urban and suburban areas are less than 10 miles. In mobile radio communications, the design of cell coverage is based on the average power. No margin is applied in calculating the cell coverage. Natural Problems in Mobile Radio Communications In mobile radio communications, there are many problems which never occur in fixed-to-fixed radio commu- nication system: 1. Excessive pathloss: Vehicles are referred to as mobile units. The antenna height of the mobile unit is very close to the ground. Therefore, the average signal strength received at the mobile unit has two components, William C. Y. Lee AirTouch Communications, Inc. Rodger E. Ziemer University of Colorado at Colorado Springs Mil Ovan Motorola, Inc. Giridhar D. Mandyam Nokia Research Center ? 2000 by CRC Press LLC a direct wave and a ground-reflected wave. These two waves act in canceling their average signal strengths and result in excessive pathloss at the receiver. 2. Multipath fading: Due to the human-made environment in which mobile units travel, the instantaneous signal sent from the base station is reflected back and forth from buildings and other ground objects before arriving at the mobile unit and causes signal fading received in the time domain. This signal fading causes an increase in the bit error rate (BER) and in the degradation of voice quality. 3. Human-made noise: The antenna height of mobile units is usually low. Therefore, human-made industrial noise, automotive ignition noise, etc. are very easily received by the mobile unit. This noise will raise the noise floor and impact system performance. 4. Dispersive medium: Due to the human-made environment and the low antenna height of the mobile unit, the signal after bouncing back and forth from the human-made structures produces multiple reflected waves which arrive at the mobile unit at different times. One impulse sent from the base station propagating through the medium becomes multiple reflected impulses received at different times at the mobile unit. This medium is called a dispersive medium. First the dispersive medium does not affect the analog voice channel, but does affect the data channels. Second, the medium becomes effective depending on the transmission symbol rate of the system. The dispersive medium will impact the reception performance when the transmission rate is over 20 kbps. Third, the dispersive medium becomes more effective in urban areas than in suburban areas. Description of Mobile Radio Systems There are two basic systems: trunked systems and cellular systems. Trunked Systems A trunked system is assigned a channel from a number of available channels to a user. The user is never assigned to a fixed channel. 1. Specialized mobile radio (SMR) is a trunked system. The SMR operator is licensed by the FCC to a group of 10 or 20 channels within 14 MHz of the spectrum between 800 and 900 MHz. ? Loading requirement: A minimum of 70 mobile units per channel is required. SMR can offer privacy, speedier channel access, and efficient services. It can serve up to 125–150 mobile units per channel. ? Channel spacing: 25 kHz. ? Channel allocation: The FCC allocates a spectrum of either 500 kHz or 1 MHz to a SMR operator who will serve 10 or 20 paired transmit-receiver voice channels. ? Coverage: Coverage is about 25 miles in radius since SMR uses only one high-power transmitting tower covering a large area. ? Telephone interconnect: The public service telephone network (PSTN) extends mobile telephone service to SMR users. ? Roaming: Mobile units are equipped with software that allows the radio to roam to any SMR system in the network. ? Handoff: No tower-to-tower handoff capability; the channel frequency does not change as the unit moves from one cell to another. 2. ESMR (enhanced SMR): A system used to enhance the SMR system. It was called MIRS (Mobile Integrated Radio Systems). Now it is called IDEN (Integrated Dispatch and Enhanced Network). Features are: ? Uses the SMR band. ? Uses TDMA (time division multiple access) digital technology, the same digital TDMA standard adopted by the cellular industry. ? Applies network of low-power cells. ? Provides cell-to-cell handoffs through a centralized switching facility. ? A spectrum average of 7–8 MHz is used in each market. The spectrum is not contiguous. ? A channel bandwidth of 25 kHz is specified with three time slots per channel. ? Modulation 16 QAM is applied. ? No equalizer is used. ? 2000 by CRC Press LLC Cellular Systems The cellular system [Lee, 1989] is a high-capacity system that uses the frequency reuse concept. The same frequency is used over and over again in different geographical locations. In large cities, the same frequency can be reused over 30 times. Key Elements: There are several key elements in the cellular system. ? Cochannel interference reduction factor q (see Fig. 75.1):Two cells using the same frequency channels are called cochannel cells. The required distance between two cochannel cells in order to receive the accepted voice quality is D s , and the radius of the cell is R. Then the cochannel interference reduction factor q is q = D s /R There are six co-channel cells at the first tier seen from the center cell as shown in Fig. 75.1(a). For an analog cellular system q s = 4.6, and the cell reuse factor K is K = q s ? = 7. The K = 7 means that a cluster of seven cells will reuse again and again in a serving area. The capacity increase in a cellular system can be achieved by reducing both the radius of cell R by one half and the separation D s by one half such that the q s remains constant and the capacity is increasing by four times. The reason is that a cell shown in Figure 75.1(a) can fit in four small cells shown in Fig. 75.1(b). In Fig. 75.1(c), the size of cells is the same as Fig. 75.1(a), but q s = 3 is achieved by using an intelligent microcell system. The capacity of Fig. 75.1(c) is 7 /3 = 2.33 times over that of Fig. 75.1(a). In Fig. 75.1(d), the radius of the cell is reduced by one half, and K is reduced to 3. The capacity of Fig. 75.1(d) is 4 ′ 2.33 = 9.32 times over that of Fig. 75.1(a). The value of q is different in different kinds of cellular systems such as analog, TDMA, and CDMA (code division multiple access). ? Handoff: Handoff is a feature implemented in cellular systems to handoff a frequency of a cell while the mobile unit changes to another frequency of another cell while the vehicle is entering. The handoff is handled by the system and the user does not notice the handoff occurrences. FIGURE 75.1 Four cases of expression of cochannel interference reduction factor. ? 2000 by CRC Press LLC ? Cell splitting: When a cell provides a maximum of 60 radio channels and all are used during busy hours, the cell has to be split into smaller cells in order to provide more radio channels, normally reducing the cell by using a half radius. As a result a cell will be covered by four subcells. Each subcell provides 60 channels. The total area of an original cell will provide 240 radio channels which is four times higher in capacity as compared with the original cell capacity before splitting. Spectrum Allocation in the United States, Europe, and Japan: In the United States there is 50 MHz of spectrum allocated to cellular radio within 800–900 MHz. Based on duopoly, each city has two licensed operators. Each one operates on a 25-MHz band. There are two bands, Band A and Band B. Each band consists of 416 channels. The channel bandwidth is 30 kHz. Among 416 channels, 21 channels are used for setting up and 395 are used for voice channels. ? Analog: The frequency management of both Band A and Band B is shown in Table 75.5. ? Digital: There are two potential systems, TDMA and CDMA shown in Table 75.1. In Europe the spectrum allocation is as shown in Table 75.2 and 75.3. In Japan the spectrum allocation is as shown in Table 75.4. TABLE 75.1 Specifications of TDMA and CDMA Systems TDMA CDMA Bandwidth per channel 30 kHz Bandwidth per channel 1.23 MHz Time slots 3 Speech coder 8 kbps(max.)—a variable rate vocoder Modulation p/4-DQPSK Forward radio channels Pilot (1) sync (1), paging (7), Speech coder 8 kbps—VSELP code traffic channels (55), total 64 (vector sum excited LPC*) channels Channel coding Rate 1/2 convolutional (13 kbps) Reverse radio channels Access (9), traffic channels (55) Total transmit rate 48 kbps per channel Power control Forward, reverse Equalizer Up to 40 ms Diversity Rake receiver * LPC = linear predictive code. TABLE 75.2 Specification of Three European Systems Analog England Scandinavia West Germany System TACS* NMT* C450 Transmission frequency (kHz) Base station 935–960 463–467.5 461.3–465.74 Mobile station 890–915 453–457.5 451.3–455.74 Spacing between transmission 45 10 10 and receiving frequencies (MHz) Spacing between channels (kHz) 25 25 20 Number of channels 1000 180 222 (control channel 21 ′ 2); interleave used Coverage radius (km) 2–20 1.8–40 5–30 Audio signal Type of modulation FM FM FM Frequency deviation (kHz) ±9.5 ±5 ±4 Control signal Type of modulation FSK FSK FSK Frequency deviation (kHz) ±6.4 ±3.5 ±2.5 Data transmission rate (kbps) 8 1.2 5.28 Message protection Principle of majority Receiving steps are Message is sent again decision is employed predetermined when an error according to the is detected content of the message * TACS = total access cellular system; NMT = nordic mobile telephone. ? 2000 by CRC Press LLC TABLE 75.3 GSM European Standard GSM Characteristics ? TDMA: 8 slots/radio carrier ? 124 radio carriers (200 kHz/carrier) 935–960 MHz, 890–915 MHz ? GMSK modulation ? Slow frequency hopping (FH) (217 hops/s) ? Block and convolutional channel coding ? Synchronization (up to 233 ms absolute delay) ? Equalization (16 ms dispersion) ? TDMA structure: one frame (8 slots) 4.615 ms; each slot 0.557 ms ? Radio transmission rate: 270.833 kbps GSM Physical Channels ? RACH: random-access control channel ? BCCH: broadcast common control channel (system parameters, sync.) ? PCH: paging channel ? SDCCH: stand-alone dedicated control channel (for transmit user’s data) ? FACCH: fast associate control channel (for handoff) ? SACCH: slow associate control channel (for signaling) ? TCH: traffic channel Full rate: use full rate speech code Half rate TABLE 75.4 Specification of the Japanese System Analog System NTT Transmission frequency (kHz) Base station 870–885 Mobile station 925–940 Spacing between transmission and receiving frequencies (MHz) 55 Spacing between channels (kHz) 25 Number of channels 600 Coverage radius (km) 5 (urban area) 10 (suburbs) Audio signal Type of modulation FM Frequency deviation (kHz) ±5 Control signal Type of modulation FSK Frequency deviation (kHz) ±4.5 Data transmission rate (kbps) 0.3 Message protection Transmitted signal is checked when it is sent back to the sender by the receiver Digital System PHS* (Japan) Frequency band 1.9 GHz Access method TDMA/TDD (MC)* Traffic channels/RF carrier 1 (or 8 channels at half rate) Modulation p/4-QPSK Voice codec 32 kbit/s ADPCM Output power 10 mW Radio transmission rate 384 kpbs Carrier spacing 300 kHz * PHS = personal handy phone system; TDD = time division duplexing; MC = multi-carrier; ADPCM = adaptive differential pulse code modulation. ? 2000 by CRC Press LLC T Block A Block B ABLE 75.5 New Frequency Management (Full Spectrum) 1A 2A 3A 4A 5A 6A 7A 1B 2B 3B 4B 5B 6B 7B 1C 2C 3C 4C 5C 6C 7C 12345678910112131415161718192021 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712713714715716XXXX9192939495969798991000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 313* 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 1A 2A 3A 4A 5A 6A 7A 1B 2B 3B 4B 5B 6B 7B 1C 2C 3C 4C 5C 6C 7C 334* 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 X X X X XXXX71771871972072172723724725726727728729730731732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 *Boldface numbers indicate 21 control channels for Block A and Block B, respectively. ? 2000 by CRC Press LLC Mobile Data Systems The design aspect of developing a mobile data system is different from that of developing a cellular voice system, although the mobile radio environment is the same. The quality of a voice channel has to be determined based on a subjective test. The quality of a data transmission is based on an objective test. In a data transmission, the bit error rate and the word error rate are the parameters to be used to measure the performance at any given carrier-to-interference ratio (C/I). The burst errors caused by the multipath fading and the intersymbol inter- ference caused by the time delay spread are the major concerns in receiving the mobile data. The burst errors can be reduced by interleaving and coding. The intersymbol interference can be reduced by using equalizers or lowering the symbol rate or applying diversity. The wireless data transmission can be sent via a circuit switched network or a packet switched network. Also, mobile data transmission can be implemented on cellular systems or on a stand-alone system. Personal Communication Service Systems In June 1990, the FCC started to ask the wireless communication industry to study the development of future personal communication service (PCS) systems. (In late 1994, the FCC started to auction off two of the six spectral bands for over 7 billion dollars. In 1996, Band C was auctioned off for over 4 billion dollars.) PCS systems need to have more capacity than cellular systems. The technologies of increasing the capacity not only apply to GSM, but also apply to CDMA (code division multiple access) and the new microcell system. CDMA A San Diego field test held in 1991 showed that a cellular CDMA scheme can provide higher capacity than cellular TDMA (time division multiple access). A cellular CDMA system [Lee, May 1991] does not require a frequency reuse scheme. All the CDMA cells share the same radio channel. Therefore, the capacity of a cellular CDMA system is higher than either cellular FDMA (frequency division multiple access) or cellular TDMA systems. Assume that a spectral bandwidth of 1.2 MHz can be divided into 120 radio channels with a channel bandwidth of 10 kHz. This is an FDMA scheme. A spectral bandwidth of 1.2 MHz can also be divided into 40 radio channels with a radio channel bandwidth of 30 kHz but each radio channel carries three time slots. Therefore, a total of 120 time-slot channels is obtained. This is a TDMA scheme. A spectral bandwidth of 1.2 MHz can also be used as one radio channel but provide 40 code-sequence traffic channels for each sector of a cell. A cell of three sectors will have a total of 120 traffic channels. This is a CDMA scheme. Now we can visualize that as far as channel efficiency is concerned, TDMA, FDMA, and CDMA provide the same number of traffic channels. However, in FDMA or TDMA, frequency reuse has to be applied. Let the frequency reuse factor K = 7 maintain a required C/I 3 18 dB; then the total channels will be divided by 7 as: ARDIS* RAM* Transmission rate 4.8 kbps and 19.2 kbps Transmission rate 8 kbps Transmit power 1 W Transmit power 4 W Channel Packet radio Channel Packet radio Vendors IBM/Motorola Vendor Ericsson Cellular Plan II Cellular Modems Transmission rate 19.2 kbps Transmission rate 38.4 kbps Transmit power 0.6–1.2 W Transmission 3 W Channel Packet cellular Channel Circuit cellular, carry data over cellular voice channels Modem vendor AT&T, PowerTek, Vital *ARDIS = advanced radio data *RAM = mobile data service 120 7 17= channels/cell (in TDMA or FDMA) ? 2000 by CRC Press LLC In CDMA no frequency reuse is required. Therefore, every cell can have the same 120 channels: number of channels/cell (in CDMA). In cellular, because the frequency reuse factor is applied on FDMA and TDMA schemes but not on CDMA, therefore, cellular CDMA has a greater spectrum efficiency than cellular FDMA or TDMA [Lee, May 1991]. New Microcell System The conventional microcell system [Lee, Nov. 1991, 1993] reduces the transmit power and makes a cell less than 1 km in radius. The concept of using cell splitting is to increase capacity. Furthermore, the new microcell system needs to find a way to make a conventional microcell to be intelligent. The conventional microcell does not have the intelligence to know where the mobile or portable units are located within the cell. Therefore, the cell site has to cover the signal strength over the whole cell or whole sector. The more unnecessary signal power transmitted, the more interference will be caused in the system and less capacity will be achieved. In this new intelligent microcell system, each cell is an intelligent cell. In a new microcell, there are three or more zones. The cell will know which zone a particular mobile unit is in. Then a small amount of power will be needed to deliver in that zone. The cochannel interference reduction factor (CIRF) now will be measured from two cochannel zones instead of two cochannel cells. Then the two cochannel cells can be located much closer. In this new microcell system, the frequency reuse factor K becomes K = 3. As compared to the conventional microcell K = 7, the new microcell system has a capacity increase of 2.33 (= 7/3) times. These two techniques can be used in buildings and outside buildings. Defining Terms CDMA: A multiple access scheme by using code sequences as traffic channels in a comom radio channel. Cell splitting: A method of increasing capacity by reducing the size of the cell. Cochannel interference reduction factor (CIRF): A key factor used to design a cellular system to avoid the cochannel interference. FDMA: A multiple access scheme by dividing an allocated spectrum into different radio channels. Frequency reuse factor ( K): A number based on frequency reuse to determine how many channels per cell. GSM (Global System Mobile): European digital cellular standard using TDMA. Handoff: A frequency channel will be changed to a new frequency channel as the vehicle moves from one cell to another cell without the user’s intervention. IDEN (Integrated Dispatch and Enhanced Network): A cellular-like system. Mobile cellular systems: A high-capacity system operating at 800–900 MHz using a frequency reuse scheme for vehicle and portable telephone communications. PHS (Personal Handy Phone System): A TDD system deployed in Japan. SMR (Specialized Mobile Radio): A trunked system for dispatch. TDMA: A multiple access scheme by dividing a radio channel into many time slots where each slot carries a traffic channel. Related Topic 69.2 Radio References W. C. Y. Lee, Mobile Cellular Telecommunication Systems, New York: McGraw Hill, 1989. W. C. Y. Lee, “Overview of cellular CDMA,” IEEE Trans. on Veh. Tech., vol. 40, pp. 290–302, May 1991. W. C. Y. Lee, “Microcell architecture—Smaller cells for greater performance,” IEEE Commun. Magazine, vol. 29, pp. 19–23, Nov. 1991. W. C. Y. Lee, Mobile Communications Design Fundamentals, 2nd ed., New York: Wiley, 1993. ? 2000 by CRC Press LLC Further Information T. S. Rappaport, “The wireless revolution,” IEEE Commun. Magazine, pp. 52–71, Nov. 1991. Gilhousan et al., “On the capacity of cellular CDMA systems,” IEEE Trans. Vehicular Technol., vol. 40, no. 2, pp. 303–311, May 1991. D. J. Goodman, “Trends in cellular and cordless communications, IEEE Commun. Magazine, pp. 31–39, June 1991. Raith, K. and Uddenfeldt, J., “Capacity of digital cellular TDMA systems,” IEEE Trans. Vehicular Technol., vol. 40, no. 2, pp. 323–331, May 1991. 75.2 Facsimile Rodger E. Ziemer Facsimile combines copying with data transmission to produce an image of a subject copy at another location, either nearby or distant. Although the Latin phrase fac simile means to “make similar,” the compressed phrase facsimile has been taken to mean “exact copy of a transmission” since 1815 [Quinn, 1989]. The image of the subject copy is referred to as a facsimile copy, or record copy. Often the abbreviated reference “fax” is used in place of the longer term facsimile. Facsimile was invented by Alexander Bain in 1842; Bain’s system used a synchronized pendulum arrangement to send a facsimile of dot patterns and record them on electrosensitive paper. Over the years, much technological development has taken place to make facsimile a practical and affordable document transmission process. An equally important role in the wide acceptance of facsimile for image transmission has been the adoption of standards by the Consultative Committee on International Telephone and Telegraph (CCITT). The advent of a nationwide dial telephone network in the 1960s provided impetus for the rebirth of facsimile after television put the damper on early facsimile use. Group 2 fax machines which appeared in the mid-1970s were capable of transmitting a page within a couple of minutes. These machines, based on analog transmission methods, were developed by Graphic Sciences and 3M. The Group 3 fax machines, developed in the mid-1970s by the Japanese, are based on digital transmission technology and are capable of transmitting a page in 20 seconds or less. They can automatically switch to an analog mode to communicate with the older Group 1 and 2 fax machines. Group 4 fax units offer the highest resolution at the fastest rates but rely on digital telephone lines which are just now becoming widely available [Quinn, 1989]. Group 3 facsimile will be featured in the remainder of this article. Group 3 facsimile refers to apparatus which is capable of transmitting an 8.5 ′ 11-inch page over telephone-type circuits in one minute or less. Detailed standards for Group 3 equipment may be found in Recommendation T.4 of CCITT, Vol. VII. Facsimile transmission involves the separate processes of scanning, encoding, modulation, transmission, demodulation, decoding, and recording. Each of these will be described in greater detail below. Scanning Before transmission of the facsimile signal, the subject copy must be scanned. This involves the sensing of the diffuse reflectances of light from the elemental areas making up the subject copy. For CCITT Group 3 high- resolution facsimile, these elemental areas are rectangles 1/208 inch wide by 1/196 inch high. The signal corresponding to an elemental area is called a pixel which stands for picture element. For pixels that can assume only one of two possible states (i.e., white on black or vice versa), the term used is a pel. Various arrangements of illuminating sources, light-sensing transducers, and mechanical scanning methods can be employed. For more than six sweeps per second across the subject copy, electronic scanning utilizing a cathode-ray tube or photosensitive arrays or laser sources with polygon mirrors are utilized. A photosensitive array arrangement for scanning a flood-illuminated subject copy is illustrated in Fig. 75.2. This is the most often encountered scanning mechanism for modern facsimile scanners, and the sensors are typically silicon photosensitive devices. Two photosensor arrays in common use are photodiode arrays and charge-coupled device linear image sensors. For digital facsimile, the array is composed of 1728 sensors in a row 1.02 inches long with the optics designed so that an 8.5 inch subject copy can be scanned. ? 2000 by CRC Press LLC Encoding The output of the photosensor array for one scan or row of the subject copy consists of 1728 pels (1s or 0s) since Group 3 facsimile recognizes only black or white. Typically, facsimile subject copy is 85% white. The data from scanning the subject copy is reduced through run-length encoding. In the encoding process, it is assumed that a white pel (0) always occurs first. A white run is the number of 0s until the first 1 is encountered (the run length is 0 if the first pel is a 1); after a white run, a black run must follow with length equal to the number of 1s until the first 0 is encountered. All possible run lengths of white and black are then encoded into a binary code using a modified Huffman encoding technique [Jayant and Noll, 1984]. On the average, fewer binary symbols are needed to encode run lengths of the subject copy than if the binary values of the pels themselves were transmitted. For Group 3 facsimile, compression is optionally extended to the vertical dimension through employment of a READ (relative element address designate) code. Run lengths from 0 to 63 are encoded by terminating codes, and run lengths in equal multiples of 64 from 64 to 1728 are encoded by makeup codes. Thus any run length up to 1728 can be described by a makeup code plus an appropriate terminating code. Additional makeup codewords are available for equipment that accom- modates wider paper while maintaining the same resolution. Tables of modified Huffman run-length termi- nating and makeup codes are given in [Jayant and Noll, 1984]. Modulation and Transmission Transmission of the encoded facsimile signal makes use of modem signaling techniques based on CCITT recommendations V.27 (standard) and V.29 (optional addition). The former utilizes 8-phase modulation at 4800 bits per second (bps), and the latter employs 16-QAM (quadrature amplitude modulation) at 9600 bps with adaptive, linear equalization. A facsimile telephone call consists of five phases, labeled A through E. In phase A, the telephone call is placed, with a training sequence sent consisting of signals to establish carrier detection, AGC, timing synchronization, and adjust equalizer tap settings. Phase B consists of the called station responding with a confirmation to receive (CFR) signal. The response is a 300 bps binary coded frequency- shift keyed signal (1 = 1650 ± 6 Hz and 0 = 1850 ± 6 Hz), except for the equalizer training sequence which is at the fast data rate of the digital modem. In phase C the encoded facsimile image is transmitted. Phase D consists of the end-of-transmission signal consisting of six consecutive end-of-lines (EOLs), with receipt required from the receiver. If no more images are to be sent or received, phase E (going on-hook) is effected at both terminals. Demodulation and Decoding Demodulation consists of the inverse of the modulation process. Standard techniques are used to demodulate the phase-modulated or QAM signals. Also included in the demodulation process is equalization. The decoding process converts the run-length encoded information to a series of 1s and 0s corresponding to the black and white pels of the image. The demodulated and decoded signal is then used to control the recording of the image. FIGURE 75.2 Arrangement for scanning by means of a linear photosensitive array. ? 2000 by CRC Press LLC Recording Recording of the image at the receiver is effected by applying electricity, heat, light, ink jet, or pressure to a recording medium [Stamps, 1982]. Xerography or ink jet techniques can be used to record on plain paper. Other recording means using electricity, heat, or pressure require specially coated papers. Except for the ink jet, recording processes requiring only a one-step process utilize specially coated papers. Marking transducers are used to apply the image to the recording medium. Personal Computer Facsimile Whereas character-oriented text is readily transmitted between personal computers by means of teletex or computer mail, facsimile transmission in conjunction with personal computers extends this capability to images [Hayashi and Motegi, 1989]. Group 4 Facsimile As mentioned previously, Group 4 facsimile apparatus is used mainly on public data networks of the circuit switched, packet switched, or integrated services digital network varieties (ISDN). Group 4 facsimile machines are subdivided into the following three classes [Yasuda, 1985]: ? Class 1 with the minimum requirement that such equipment can send and receive documents containing facsimile-encoded information ? Class 2, which in addition to the Class 1 capabilities, must be able to receive teletex and mixed-mode documents ? Class 3, which in addition to Class 1 and Class 2 capabilities, must be able to generate and send teletex and mixed-mode documents. An additional feature of the specifications for Group 4 facsimile is that the resolution is equal in the horizontal and vertical directions. Standard resolution for Class 1 is 200 pels per 25.4 mm, and that for Classes 2 and 3 is 200 and 300 pels per 25.4 mm. When operating as a mixed-mode terminal, a receiving density of 240 pels per 25.4 mm is required, which is optional for all three classes. Bit rates range from 2.4 to 48 kbits/s with 64 kbits/s for ISDN. Compression techniques applicable to Group 4 facsimile are overviewed in Arps and Truong [1994]. Defining Terms Facsimile: The process of making an exact copy of a document through scanning of the subject copy, electronic transmission of the resultant signals modulated by the subject copy, and making a record copy at a remote location. Mixed-mode documents: Documents containing both character and facsimile information within a page. Such documents can be handled by Group 4 facsimile very efficiently. Group 3 facsimile treats each document as an image to be transmitted pixel by pixel. Mixed-mode documents are subdivided into increasingly smaller parts, such as pages, frames, and blocks. A block is a rectangular area that can contain only one category (character or facsimile information). Pel: A picture element which has been encoded as black or white, with no gray scale in between. Pixel: A picture element of a subject or record copy that is represented in shades of gray. Record copy: The copy of the document made at the receiving end of a facsimile system. Run-length encoding: The assignment of a codeword to each possible run of 0s (white pel sequence) or run of 1s (black pel sequence) in a scan of the subject copy. Scanning: The process of scanning the subject copy in a facsimile transmission from left to right and from top to bottom. Subject copy: The document that is scanned and transmitted in a facsimile system. Teletex: Representation of character information by code words. Such a representation considerably improves the efficiency of the transmission process, but is not suitable for handwritten characters. ? 2000 by CRC Press LLC Related Topics 69.1 Modulation and Demodulation?70.1 Coding References R. B. Arps and T. Truong, “Comparison of international standards for lossless still image compression,” IEEE Proc., vol. 82 (June), pp. 889–899, 1994. K. Hayashi and C. Motegi, “Personal computer image communications using facsimile,” IEEE Journal on Selected Areas in Communications, vol. 7, pp. 276–282, Feb. 1989. N. S. Jayant and P. Noll, Digital Coding of Waveforms, Englewood Cliffs, N.J.: Prentice Hall, 1984, chap. 10. G. V. Quinn, The FAX Handbook, Blue Ridge Summit, Pa.: Tab Books, 1989. Y. Yasuda, Y. Yamazaki, T. Kamae, and K. Kobayashi, “Advances in FAX,” IEEE Proc., vol. 73 (April), pp. 706–730, 1985. Further Information C. Chamzas and D. L. Duttweiler, “Encoding facsimilie images for packet-switched networks,” IEEE Journal on Selected Areas in Communications, vol. 7, pp. 857–864, June 1989. G. Held, Data Compression: Techniques and Applications: Hardware and Software Considerations, 2nd ed., New York: John Wiley & Sons, 1987. K. McConnell, D. Bodson, and R. Schaphorst, FAX: Digital Facsimilie Technology and Applications, Boston: Artech House, 1992. 75.3 Wireless Local-Area Networks Mil Ovan Wireless local-area networks (LANs) represent a new form of communications among personal computers inside buildings. To better understand its applicability, this paper defines the customer challenges in networking personal computers as well as specific product requirements for a wireless LAN. These insights were gained through market studies of over 1000 corporate and government entities surveyed through different marketing research techniques. The Wireless In-Building Vision To date, the evolution of wireless communications has been exemplified by the dramatic growth in cellular communications. Cellular has enabled customers to transcend the constraints of fixed telephony in commu- nicating outside of buildings with portable and now personal communications devices. There has been significant interest and publicity regarding wireless in-building communications lately, both for data and voice. Throughout the 1980s, we have seen the development of a significant range of in-building business communications problems that have been caused by changes in the technological, business, and regulatory environments. Because of these developments, buyers of telecommunications and data communications systems increasingly are having to face significant time, cost, and logistical problems associated with the installation, movement, and management of computing and communications equipment in dynamic office environments. Over the next 20 years, society will witness a significant “wireless evolution” in both personal and professional communications and change the way we conduct our lives at home, on the road, and at work (see Fig. 75.3). New forms of wireless communications will free us from the “bonds” of wire that today restrict our movements or interaction. Market Research Beginning in the middle to late 1980s, a systematic evaluation of the technological and environmental attributes necessary to anticipate and define wireless in-building communications was undertaken. This included a ? 2000 by CRC Press LLC comprehensive marketing needs assessment and research program. The overriding objective was to anticipate and identify customer needs and trends; that is, “What are the specific needs of various customer groups, and what type of product attributes will satisfy their needs?” To determine answers to these and a whole host of other questions, a multiphased marketing research program was conducted. The overall aim of the program was to anticipate and ascertain the customer need, where this need existed currently, what were the market and customer environmental characteristics, and what product characteristics would be needed to provide an optimal wireless solution. The remainder of this paper describes a higher-level overview of the results from these market research phases. This includes an overview of market needs, the problems/difficulties with current cabling methods, and a description of market requirements. LAN Market Factors Personal Computer Explosion The move from mainframe and central information processing of the 1960s and 1970s provided an opportunity for minicomputers to enter the market. The minicomputer provided greater computer and applications access by employees. Throughout the 1980s the move to more intelligent desktop devices like personal computers was just that—personal. Organizations, in an effort to empower the worker, provided all types of applications, software, and hardware to the worker. The decremental costs of technology facilitated the distribution of personal computers. More importantly, projections state that business personal computer growth will continue its aggressive pace (see Fig. 75.4). FIGURE 75.3Evolution of wireless communications. (Courtesy of Motorola, Inc.) FIGURE 75.4Worldwide business personal computer-installed base. (Source: International Data Corporation.) ? 2000 by CRC Press LLC However, the growth of decentralized storage and computing created yet another problem—work groups needed to share information—and much of this information resided in individual hard disks. Furthermore, despite the declining costs of personal computers and associated technology, it was and still is considerably expensive to “fully load” the workforce with all of the applications it needs. The ability to share applications became desirable. It was these two trends which highlighted the need for LANs. Information and Resource Sharing The success of LAN computing was predictable. It started with the basic tenet of sharing resources and/or information. The need to amortize and justify the purchase of expensive resources, such as printers and storage, was an obvious factor which supported LAN growth. The need for knowledge workers to exchange data was and is imperative. Furthermore, the ability to share applications supported the growth of network computing. LAN Growth The success of LANs throughout the 1980s has been phenomenal. However, the projected growth throughout the 1990s is equally as impressive (see Fig. 75.5). This can be attributed not only to new installations of LANs, but also to the physical and logical segmentation of LANs as traffic and throughput degradations are observed. Moves/Adds/Changes and Increasing Mobility The world economies will continue to develop interdependencies and, likewise, global competition. The increas- ing competitive environment will demand greater worker mobility, changing assignments and reassignments, changing work groups, and mission mobility. The demand to have information how we want it, when we want it, and where we want it will be a strategic and competitive weapon. The need to improve efficiency and the growing need for information will accelerate the adoption of wireless communications. Today’s wired network, for all its great strides, is very restrictive. The cost to deploy and redeploy personnel and workgroups is time consuming and expensive. Cabling in today’s environment inhibits the ability to attain efficiency and competitive advantages. The next section will highlight some of the author’s market research findings. Cabling Problems As each phase of market investigation was conducted, several problems with today’s wired networks were uncovered. Whether copper twisted pair, coax, or optical fiber, hard wiring for telecommunications and data communications systems within a building environment is expensive and troublesome to install, maintain, and, especially, change. Beneath today’s increasingly dense office electronic environment lies a tangled, confusing, unmanageable maze of wiring. What appeared to be very significant in the focus group research was how quickly the respondents stated the problems they have with wiring. Among the majority of respondents, the most favorable solution was to FIGURE 75.5Worldwide business PCs and those PCs that are LAN connected. (Source: International Data Corporation.) ? 2000 by CRC Press LLC free themselves of all wiring. Therefore, their first choice solution would be a wireless system, minimizing the time and effort of implementing a move, add, or change. Moves/Adds/Changes: Cost and Frequency A major portion of the cost of LANs is the cost of interconnecting them, which experts acknowledge can sometimes exceed the cost of computer hardware and software. Labor and material costs for wiring are almost always significant and can reach $1000 per node just for copper wire. Coax and optical fiber, not surprisingly, are considerably higher. The news, however, gets even worse when it comes to maintenance. A study by the Frost and Sullivan 1 group quotes that LAN moves, adds, and changes (MACs) is the third largest cost component for LAN installation and hardware maintenance. They state that MACs account annually for almost $2 billion of a $12.2 billion LAN maintenance market, and that $2 billion does not even include the original cost to install cable. Estimates of the cost to rewire range from $200 to $1000 per change. In fact, a survey by KPMG Peat Marwick 2 quoted that the average relocation cost for just rewiring a LAN station averages $300 per node. But those are just the direct costs; the time to effect the wired change is a significant problem as well. Moves, for example, often take weeks or longer to coordinate in addition to the time to actually make the physical wiring change. Most of the research respondents were asked what proportion of their company’s staff was involved in some kind of a move involving wiring or rewiring. The majority of the respondents, almost 80%, had some type of relocation or addition of personnel over the last year surveyed. Their responses ranged from as few as 20% per year up to as much as 200% annually. Furthermore, according to the KPMG Peat Marwick study, the average company moves its employees approximately 50% annually. Telecommunications consultant Richard Kuehn states that data terminals are moved as often as 1.5 to 3 times per year. The combined problems of the actual hard relocation costs, however, are just the beginning. Soft, or hidden, costs further exacerbate the cabling dilemma. Hidden Costs Significant problems arise when these moves or changes are implemented. There is always the disruption of the workers involved in the move or change, not to mention the loss in productivity. The problems, however, become much more involved when dealing with whole departments and more complex user equipment. In fact, surveyed firms responded that when a relocation takes place, over 60% of the time it involves the movement of an entire department. The toll of wait time and down time on productivity varies greatly and is difficult to quantify, but certainly is significant and costly. In today’s increasingly mobile working environment, it is likely to grow. The situation is exacerbated by relocations and additions which require reconstructions, thereby continuing to add to the effective cost of a move, add, or change. Costs to rewire rise enormously with the age and complexity of the building. The majority of high-rise office space in large metropolitan areas presents major problems and expense for tenants trying to install, add, or move network wiring. Buildings more than 30 or 40 year old, with designs and construction that did not consider today’s electronic office, poorly accommodate communications wiring. If asbestos insulation exists in the building, as it does even in many pre–health-safety regulated buildings, rewiring costs can take on huge proportions. The coordination of personnel and the moving of one group out to prepare for the new group moving in is a very costly and labor intensive ordeal. In some cases, wiring had to be installed, or different cabling may have been needed to accommodate new or different types of users’ equipment. Cable Is Not Business Friendly Although office planners, building managers, and network operators are well aware of the problems with wire, the limitations and huge costs of wire have not generated focused attention outside of this community. The general business world seems to accept wire as inevitable. Perhaps that is because there have been no real 1 PC Week Magazine, “Maintenance Costs of LANs Keep Soaring,” Frost & Sullivan, Inc. 2 KPMG Peat Marwick Study, January 1991. ? 2000 by CRC Press LLC alternatives. Yet, as computing and telecommunications power continues to proliferate and becomes more widely distributed to the “knowledge worker,” the problem will increase. Easy, quick, efficient movement of “people assets” within the working environment is also increasingly being recognized as essential to the productivity and competitiveness of a business. Wiring severely inhibits that movement. The research indicated a need for a flexible, compatible, cost-effective, yet high-performance wireless alter- native to extend and complement, if not replace, the capabilities of wire, cable, and fiber for in-building communications networks. More specifically, it is the convenience and flexibility that users need. In fact, the aggregate need for flexibility and convenience was found to be twice that of the perceived benefit for cost savings. When research respondents were asked how they could improve upon their experiences when implementing a move, add, or change, many solutions were offered. These solutions ranged from having more compatibility among different vendors’ equipment, to providing a better way to organize all the different cabling. Structured Distribution Systems A number of firms in the research study had deployed a structured distribution system (SDS), a topology which advocates cabling saturation of a desired environment to accommodate all potential personnel move- ments and reconstructions within that office. SDS requires firms to invest large sums of capital initially on the assumption of not knowing how many telecommunications devices may be employed or where the devices are to be located. Consequently SDS usually plans for worst-case conditions, meaning that some or much of wiring systems capability may never be utilized. However, many firms which have an SDS deployed also expressed those problems which stress their SDS investment. Some of the most frequently mentioned include: ?High equipment addition/relocations exceeding 40% annually ?Expansion and contraction of their workforce ?Changing technology and business support ?Continued investment and vigilance to maintaining the SDS and its intrinsic advantage ?Continued departmental LAN growth requirements In short, the latter group of SDS respondents provided some notable requirements. A wireless system must: ?Extend the capabilities of their SDS system ?Facilitate the inherent advantages of the SDS ?Offer enhanced flexibility to nonserviced SDS portions of their building or occupancy These points indicate that even in SDS environments, there is an opportunity to employ wireless devices. Wiring—the expense, time, and inflexibility of installing, moving, and changing—limits the way companies can productively use networks. To stay productive, these LANs have to move and change with the workforce they support. Therefore, a wireless offering must be a complementary solution for buildings with an SDS, in bringing wireless flexibility and extensibility to today’s networks. User Requirements Environment Office Friendly Several notable conclusions were derived from the marketing research. Secondary market research suggests that over 70% of LAN node installations were estimated to reside in an office environment (as opposed to factories and warehouses). Therefore, as an office-oriented offering, a wireless system would have to be, by definition, office friendly. A traditional office is composed of hard offices with opening and closing doors, furniture and personnel movement, cubicles, conference rooms, and walls of varying thickness and substance. Therefore, a wireless system must continually adapt to different and changing conditions and office layouts. Optimized Service Area The second wireless in-building need expressed by the office market is the manageability and reuse of any potential system. Unlike the signal propagation characteristics of many lower-frequency radio products, LAN ? 2000 by CRC Press LLC administrators desired the ability to control or, more aptly, contain the coverage of a potential wireless system. The reasons were twofold: ?LAN managers wanted the ability to add different services to a new group of users. In fact these new users may very well be physically adjacent to another system, wireless or wired. ?These same managers wanted the flexibility to connect a new or existing user group to either a backbone or to create a stand-alone LAN. LAN Workgroup Sizes Respondents were asked as to where a wireless offering might be first installed. The market research indicated that approximately 70% of the installations would contain less than 30 users (see Fig. 75.6). Furthermore, the average LAN appeared to be in the 12- to-15-node range. This is further corroborated by the KPMG Peat Marwick study which found that the average LAN size is about 15 users per LAN. Furthermore, the system must have the flexibility to manage the service area. That ability, to either incre- mentally add systems whether on the backbone or in a stand-alone configuration, must accommodate scalability within an organization. It is interesting to note that these figures are consistent with good LAN administration practices for purposes of maintaining high throughput and fault isolation. As LANs become larger and traffic more intensive, there is a natural inclination to begin segmenting LANs into more logical and defined user areas/groups. Coverage Area To satisfy the majority of requirements, we determined that approximately 70% of LANs would be deployed in areas of less than 5000 ft 2 (see Fig. 75.7). This must take into consideration the fairly dense environment, made up of cubicles and apportioned hallway space. The market investigations indicated that a wireless offering must accommodate, at the least, 150 ft 2 per user. This is equivalent to 32 users/system in a 5,000 ft 2 area. Product Requirements: End User Reaction Transparency, Compatibility, and Performance To justify the expense of a wireless system to end users, a wireless offering would have to provide reliable performance, as well as be practical and cost effective. Our market research indicated that the ideal system should be: ?Easy both to install and move, preferably by the user ?Able to coexist with both existing wire and cable, as well as with future optical fiber ?Easy to operate, virtually transparent to the user ?Almost universally applicable, suitable to replace any LAN cable or wire, in any office environment ?Secure, absolutely reliable, and cost effective FIGURE 75.6Survey results: forecast users per wireless LAN. (Courtesy of Motorola, Inc.) ? 2000 by CRC Press LLC A wireless system must be totally transparent. If customers are to enjoy the attributes of wireless, the respondents indicated that the wireless implementation must not require the users to change the way they operate or interface with their personal computers. Also, a wireless offering must provide true compatibility. The wireless connection must be compatible with standards-based components such as operating systems and applications, LAN cards and other devices, as well as LAN wire that is already in place. Security and Reliability In addition, the market mandated that a wireless product offering be absolutely secure and reliable. Security was a requirement across several dimensions. To provide sufficient data security a wireless system should first prevent the effective capture of data by a receiver outside of the wireless system and, second, prevent capture of the data by unauthorized wireless hardware within the system. A wireless product must be secure from eavesdropping, either accidental or intentional. Reliability was another important attribute. The users required absolute reliability. That is, users wanted a guaranteed packet delivery from the entry/exit wireline points—and they wanted it at least as error free as their current cabled environment. Cost Effective Finally, most businesses will place any capital or expense under rigorous financial analysis. As such, the acceptance of a new technology/application must pass the payback test for that business. Therefore, demonstrable payback and justification is needed to facilitate an organization’s evaluation of any potential wireless offering. Technology Alternatives in Meeting Customer Requirements Infrared and Spread Spectrum Technologies Lack Performance Developing the ideal system, obviously, is no trivial problem. Several wireless network products, to be sure, are available, but many suffer from limited performance and operational problems inherent to the technologies on which they are based. As such, they are perhaps interim point solutions, primarily for small networks but are not the long-term answer to the wire communications dilemma described earlier. Two basic technologies characterize the wireless LANs currently available. Infrared (IR) systems use a part of the electromagnetic spectrum just below visible light as a transmission medium. IR, being light, travels in a straight, or line-of-sight, path. It is blocked by opaque objects and reflects well only off hard, mirror-like surfaces. This factor stands as a serious obstacle to IR systems for applications other than in open working environments. Radio technologies form the platform for other wireless LAN products. Many of the radio LANs are based on spread spectrum technology. This technology, developed by the U.S. military, uses a combination of several small, narrow bands within a general region of this band as carrier frequencies. However, the commercial products operate with fewer frequencies than are available for military systems; hence interference rejection FIGURE 75.7 Survey results: forecast office area of wireless LAN. (Courtesy of Motorola, Inc.) ? 2000 by CRC Press LLC and performance are lower. (In radio transmission, the wider the bandwidth or available frequencies on which to encode data, the higher the achievable total data rate.) Another important issue in this ultrahigh frequency (UHF) environment is that radio frequency (RF) energy at these frequencies tends to propagate through and around obstacles, reaching beyond the confines of the network it is serving. That property makes this RF band suitable for receiving commercial broadcast signals from distant stations through the walls of buildings to receivers inside. It also makes it suitable for mobile cellular telephones, but it cannot be well contained within the confines of the microcell described earlier, which limits spectrum reuse and overall network capacity requirements absolutely essential in a viable wireless network communications system. Current UHF spread spectrum wireless in-building communications systems, then, suffer from critical bandwidth and spectrum reuse shortcomings that seem likely to prevent them from expanding beyond limited applications. The 18- to 19-GHz Radio Band: An Ideal Choice for Wireless In-Building Microcellular Networks A region of the electromagnetic spectrum above the kilohertz and megahertz radio bands, yet below the extremely high frequencies of the infrared band, offers two very compelling advantages. Specifically, the 18- to 19-GHz portion of this band fulfills the key requirements of spectrum reuse and bandwidth availability that eliminates most other frequencies from consideration. Properties Right for Both Microcell Coverage and Confinement The first major advantage of the 18-GHz band is its excellent propagation characteristics for a microcellular network. Indeed, the behavior and properties of these higher frequencies that are disadvantages for traditional long-range broadcast applications become critical advantages for wireless microcellular network applications. Propagation characteristics of 18-GHz radio waves make them well suited to diffuse thoroughly through a network microcell using only a minimum of transmitted power, yet still stay confined within it so that the same frequencies can be reused by another system within as little as 120 feet or so, or even on the other side of a dense, continuous barrier such as a cement floor. Typical microcells might encompass, for example, a level or floor, or portions of a floor, in a standard office building. As one might expect, 18-GHz radio waves exhibit a blend of the characteristics of UHF frequencies below them and IR light above them. For example, 18-GHz waves act like light and unlike radio in that they are blocked and reflected by large structures such as concrete and steel. Reflecting back and forth would allow them to fill an area defined by concrete floors and walls with only very small amounts of transmitted power, yet not pass beyond. They also refract like light, penetrating tiny holes and cracks such as closed doors to diffuse and spread through the space beyond. What little radio signal that might escape the microcell would be rapidly dissipated. Also, unlike lower frequency radio, 18-GHz radio is of a high enough frequency that not only office equipment but even high-energy factory equipment and processes do not interfere with it. Likewise, with its high-frequency and low-required transmitted power, the 18-GHz signals themselves from such a system would not interfere with other electronic systems or equipment. On the other hand, like radio and unlike light, 18-GHz signals still can pass through less dense materials such as drywall and interior office separators and, combined with their reflectivity, are thus not subject to “line-of-sight” limitations. They can also be modulated to carry information just as traditional radio signals are. Finally, since antenna size and design are largely a function of wavelength, which decreases as frequency increases, the antennas for the 16-mm wavelengths of an 18-GHz radio system would be relatively small and compact. Plenty of Bandwidth in an Otherwise Crowded Spectrum The second major advantage of 18-GHz radio is its available bandwidth. Few other areas of the electromagnetic spectrum are as interference-free, clear, and available as this band, certainly not the VHF and UHF bands, which must accommodate television, FM radio, cellular telephone, baby monitors, and more. The reason for this clear band is largely that these higher frequencies have been difficult to work with. The particular technical properties of 18-GHz frequencies and the expense, size, and complexity of the equipment to use them have prevented them from being an attractive option for many commercial applications. As a result, the military has been the primary developer and user of the 15- to 300-GHz band, and the few commercial ? 2000 by CRC Press LLC uses that have emerged (weather, aircraft and police radar, point-to-point telecom transmission, etc.) use expensive technology pioneered by military-funded research. Developing a comprehensive in-building radio system, however, had never been done until Motorola recently developed the Altair? wireless ethernet network. Such an application required the creation of new, improved performance data handling and signal processing hardware and software, as well as a radio antenna system that could transmit and capture these data speeds on 18-GHz frequencies in an in-building environment. Summary The numerous problems with wiring will become even more acute in the office of the 1990s. This environment will be characterized by: ? The proliferation of decentralized computing resources ? Increased number of telephones and personal computers as an outgrowth of a country’s economic shift toward service industries As the penetration of personal computers nears a one-to-one relationship with phones in the office workplace, the limitations of separate voice and data networks will become even more evident. If these problems are not addressed, an organization’s flexibility in redeploying “people assets” and ultimately competitiveness will be seriously hindered. The time it takes to move/add/change equipment and reconfigure communications wires will be the limiting factor in rapidly reorganizing workgroups and responding to new assignments. Wireless LANs will become an attractive solution in the office of the 1990s, interconnecting personal computers and offering data communications capabilities without the need for elaborate cabling methodologies. The obvious and inherent flexibility offered by wireless LANs is the obvious primary benefit. However, the ability to retrieve that investment, never retrievable until now, clearly presents a significant economical benefit. Defining Terms Backbone: Wiring which runs within and between floors of a building and connects local-area network segments together. Cellular communications: Traditionally an outside-of-building radio telephone system that allows users to communicate from their car or from their portable telephone. Copper twisted pair, coax, and optical fiber: Wired media which connects telephone and computer equipment. Microcell A low-power radio network which transmits its signal over a confined distance. Secondary marketing research: Market research conducted by other organizations. Spectrum reuse: Reusing frequencies over and over again in a confined area, resulting in more efficient utilization and higher radio network capacity. Structured distribution systems (SDS): A topology which advocates cabling saturation of a desired environ- ment to accommodate all potential personnel movements and reconstructions within that office. Wireless local-area networks: A method of connecting personal computers together without extensive cabling, allowing communications among these devices in an area such as a department or floor of a building. Related Topic 72.3 Local-Area Networks References J. D. Gibson, The Communications Handbook, Boca Raton, Fla.: CRC Press, 1997. N. J. Muller, Wireless Data Networking, Boston, Mass.: Artech House, 1995. Further Information Articles on LANS appear in IEEE Communications Magazine and IEEE Network Magazine. ? 2000 by CRC Press LLC 75.4 Wireless PCS Giridhar D. Mandyam Personal Communications Services (PCS) promise to introduce a wide range of variety of digital wireless services; including high-speed data, improved voice services, and messaging (e-mail or paging). These services are also often identified with the part of the spectrum in which they are deployed, that is, the PCS band. In North America, this part of the spectrum lies between 1850 and 1990 MHz, and is divided into six blocks of either 5 or 15 MHz each. The Federal Communications Commission (FCC) of the United States has been auctioning these blocks since 1994. The DCS (Digital Cellular System) band, which also spans the range of 1.8 to 2 GHz, has been set aside for these advanced services in several parts of the world. PCS can be contrasted with services already deployed in the cellular band — that is, the part of the spectrum ranging from 806 to 890 MHz. This band is divided into channels of 30 kHz apiece. In North America, the mature analog technology known as AMPS is widely deployed and provides the largest amount of coverage of all public wireless technologies available today. However, digital wireless does in fact exist in the cellular band. A depiction of spectral allocation can be found in Figure 75.8. In addition, digital PCS technologies can be divided into two categories: 2nd generation and 3rd generation wireless systems. 3rd generation wireless systems, which have yet to be deployed, promise an improvement on 2nd generation (already deployed) systems in voice quality and data services. In particular, high rate packet data is a critical requirement of 3rd generation systems. Cellular Band Systems The first system to appear in the cellular band in the United States was AMPS. This system provided user traffic channels of 30 kHz, as part of a frequency division multiple access (FDMA) scheme. In addition, this system used the concept of frequency division duplexing (FDD) to provide different channels for an individual user to send and receive traffic. This system used analog frequency-modulation technology, and was primarily designed to provide voice service only, although some systems do exist that provide data services through AMPS. The first public service began operation in the Chicago area in 1983. A problem with AMPS is the occupation of an entire 30-kHz channel by a single user. This affected the overall capacity of AMPS systems. Another problem is the lack of privacy in AMPS, which has led to a serious problem of phone cloning. In addition, the performance of analog FM in the mobile channel suffers from the threshold effect, where signal quality degrades rapidly once received signal levels fall below a threshold value. Digital technologies take advantage of coding and modulation to increase signal quality when received signal levels are low. Moreover, digital technologies employ encryption, which addresses to a certain extent the problem of cloning phones. As a result, digital wireless systems were developed for use in the cellular band to address some of the problems with AMPS. A summary of the AMPS radio interface is given in Table 75.6. A digital technology, which emerged in the United States in the late 1980s, was time division multiple access (TDMA). TDMA systems took advantage of time multiplexing different users into the same 30-kHz channel, which was used by AMPS. The IS-54 public wireless standard introduces the concept of three-slot TDMA, in which three users were time multiplexed into a single 30-kHz channel. This has the effect of tripling the effective capacity of an AMPS network; therefore, sometimes TDMA is referred to as D-AMPS, for digital AMPS. The first commercial TDMA system to be launched in the United States was in 1991. The IS-136 standard, which FIGURE 75.8 Frequency allocation. ? 2000 by CRC Press LLC was released in the mid-1990s, introduced enhancements to IS-54, including PCS functionality. The radio interface for IS-54 and IS-136 is summarized in Table 75.7. GSM is the European-originated digital TDMA standard. It also appears in the sub-1-GHz part of the spectrum, with the uplink band in the 890 to 915 MHz range and the downlink band in the 935 to 960 MHz range. However, GSM differs from IS-136 in several respects, including modulation (GSM uses GMSK), data rate (270.83 Kbps), and channel spacing (200 kHz). GSM systems exist in many parts of the world outside of Europe, including North America. Another digital technology, introduced for cellular systems in the early 1990s, was code division multiple access (CDMA). This technology is based on the principles of spread spectrum, in which narrow-band user traffic is transformed into wideband signals resembling white noise over the resultant signal bandwidth. This is accomplished by modulating user traffic with a higher-rate spreading sequence, normally generated by maximum-length shift registers. This technology was used in military communications for its inherent security and resistance to jamming. The IS-95 public wireless standard introduced CDMA, with each user occupying a 1.25-MHz bandwidth. The first commercial CDMA systems were launched in the United States in 1996. The radio interface for IS-95 is summarized in Table 75.8. TABLE 75.6 AMPS System Parameters System Parameter AMPS Multiple access FDMA Channel bandwidth 30 kHz Number of users per channel 1 Reverse (uplink) frequency range 824–849 MHz Forward (downlink) frequency range 869–894 MHz Voice modulation FM Voice peak frequency deviation 12 kHz Control channel modulation Binary FSK Control channel peak frequency deviation 8 kHz Control channel data rate 10 Kbps (Manchester encoded) Control channel error correcting code BCH TABLE 75.7 IS-54/IS-136 System Parameters System Parameter IS-54/IS-136 Multiple access TDMA Channel bandwidth 30 kHz Number of users per channel 3 Reverse (uplink) frequency range 824–849 MHz Forward (downlink) frequency range 869–894 MHz Modulation π/4-DQPSK Forward and reverse data rate 48.6 Kbps Error correcting code Rate ? convolutional code, K = 7 TABLE 75.8 IS-95 System Parameters System Parameter IS-95 Multiple access CDMA Channel bandwidth 1.25 MHz Reverse (uplink) frequency range 824–849 MHz Forward (downlink) frequency range 869–894 MHz Modulation BPSK — Quadrature Spread Forward and reverse data rate 9.6 Kbps Error correcting code Rate ? convolutional code, K = 9 ? 2000 by CRC Press LLC It is of interest to note that in IS-95, the number of users per data channel is not specified in Table 75.8. This is due to the fact that such a quantity is not easy to derive for all conditions, due to a number of factors. The downlink for IS-95 provides two forms of user channelization, based on spreading sequence time offsets and Walsh codes. However, spreading sequence offsets alone provide uplink channelization. Moreover, both the uplink and downlink are interference-limited, due to multiple users occupying the same frequency band simultaneously. PCS Services Both the IS-136 and IS-95 digital wireless standards have evolved to provide enhanced services for PCS. These enhancements include enhanced voice services, data services and short message services (SMS), and paging and e-mail. Enhanced Voice Services IS-136 employed an 8 Kbps speech coder, VSELP, as its initial codec for voice services. It takes as input 64 Kbps voice in pulse-coded modulation (PCM) format, and outputs 8 Kbps. Processing is performed on 20-ms intervals, or frames. The 13 Kbps EFR codec provides an evolution to PCS services, and has been introduced for TDMA systems, including IS-136 and GSM. This codec provides improvements in voice quality, and provides additional error correction under noisy conditions. IS-95 also initially employed an 8 Kbps speech coder developed by Qualcomm Inc., QCELP8. In order to evolve to higher voice quality, the QCELP13 13 Kbps coder was introduced primarily for PCS applications. Both of these codecs employ variable-rate coding, choosing from four different compression rates for each 20-ms frame. This is done to enhance capacity, allowing individual users to occupy more of the channel when their voice activity was high. However, the evolution to 13 Kbps resulted in weakened error-corrective coding. As a result, another variable-rate 8 Kbps codec, EVRC, was introduced. This codec provides comparable voice quality to QCELP13 at a lower rate. EVRC uses only three of the original four coding rates used by QCELP8 for each frame. Data Services Both IS-95 and IS-136 systems presently support a “packet-over-circuit” approach to data services. This employs transmitting packet data originating from an application (usually running over TCP/IP) through a dedicated wireless traffic channel. Although such data transfer is straightforward, it is wasteful in that a user may be transmitting or receiving blank traffic frames while waiting for a packet burst from the application. IS-136 also employs the CDPD method for packet data transfer over an existing AMPS channel. This method is efficient in that the user must “share” the channel with other users, so as to take advantage of the bursty nature of packet data. IS-95 has also incorporated channel aggregation in recent revisions. As mentioned before, IS-95 supports two voice codec rates of 8 Kbps and 13 Kbps. As proposed in the recent IS-95-B wireless standard, a single user can send or receive multiple traffic channels, each of which either support 8 Kbps or 13 Kbps (but not both rates mixed). The maximum number of such channels is eight in either the uplink or downlink, and only one channel is variable rate. SMS Presently, both IS-95 and IS-136 support SMS. SMS is used to provide a host of teleservices, including over- the-air programming and alphanumeric messaging (typically less than 250 characters). SMS can also support low-rate data applications employing a transparent transport layer, such as UDP. Paging and E-mail Both IS-95 and IS-136 support user paging and wireless e-mail services. This is normally accomplished through generic data burst messaging on common control channels. 3rd Generation Enhancements 3rd Generation versions of IS-95 and IS-136 are currently being developed and standardized, with the projected deployment being in the PCS band. The requirements of these systems has in general been provided by the ? 2000 by CRC Press LLC International Telecommunications Union (ITU) as part of its IMT-2000 (International Mobile Telecommuni- cations 2000) project. Enhancements provided to both systems encompass data rates of up to 2 Mbps, suitable performance under a variety of mobile channel conditions (indoor and outdoor), support of simultaneous user services, compatibility with existing 2nd generation systems, and several other considerations. It is of interest to note that GSM and IS-136 are both evolving to a common standard known as EDGE. Both versions encompass a significant modification of their 2nd generation counterparts, including modu- lation, error correction coding, and user bandwidths. However, both versions take advantage of existing 2nd generation voice codecs. In addition, another CDMA standard known as Wideband CDMA (W-CDMA) is under development in Europe and Japan. Although this standard is a CDMA standard, it is not backwards-compatible with IS-95, and is not designed in such a way that W-CDMA equipment can be used interchangeably with 3rd generation IS-95 equipment. The development of 3rd generation systems is ongoing, with equipment deployment forecast for the first part of the new millennium. Much work remains to be done before these standards can solidify sufficiently for equipment development and deployment to be completed. Defining Terms AMPS: Advanced Mobile Phone Services. A public, multiple-access wireless system that uses analog fre- quency-modulation technology. This service primarily appears in the cellular band (806–890 MHz). CDMA: Code Division Multiple Access. A method of multiple access in which individual users are assigned unique code sequences while using a common frequency. CDPD: Cellular Digital Packet Data. A packet data service over the 30 kHz analog channel. Can easily be overlaid over existing AMPS networks. EDGE: Enhanced Data Rates for Global TDMA Evolution. 3rd Generation standard for both GSM and IS-136. EFR: Enhanced Full Rate coding. 13-Kbps linear predictive coder used in IS-136 TDMA systems. EVRC: Enhanced Variable Rate Coder. 8-Kbps variable-rate linear predictive coder used in IS-95 CDMA systems. FDD: Frequency Division Duplexing. The practice of providing multiple frequency bands for an individual user. For example an individual user can receive traffic in one band and send traffic in a different band. FDMA: Frequency Division Multiple Access. A method of multiple access in which individual users are assigned different frequencies. 1st Generation: Term referring to earliest deployed public wireless systems. AMPS is included in this category. GSM: Global System for Mobile. A TDMA-based digital public wireless system first deployed in Europe. IS-95: North American CDMA standard, developed by the Telecommunications Industry Association. IS-54: North American TDMA standard, developed by the Telecommunications Industry Association. Replaced by IS-136. IS-136: North American TDMA standard, developed by the Telecommunications Industry Association. Sup- planted IS-54. PCS: Personal Communications Services. Refers either to advanced digital wireless services, or to the fre- quency band where such services are normally deployed (1850–1990 MHz). QCELP8: Qualcomm Code Excited Linear Predictive coding. 8 Kbps variable-rate linear predictive coder used in IS-95 CDMA systems. QCELP13: Qualcomm Code Excited Linear Predictive coding. 13 Kbps variable-rate linear predictive coder used in IS-95 CDMA systems. 2nd Generation: Term referring to the first deployed digital public wireless systems. This category includes CDMA and TDMA technologies. TDMA: Time Division Multiple Access. A method of multiple access where individual users are assigned time slots while using a common frequency. 3rd Generation: Term referring to digital public wireless systems that offer significant enhancements over 2nd generation systems. These enhancements include high-speed Internet access, realtime video com- munications, high-fidelity voice, multiple simultaneous services per user, broadcast capability, enhanced capacity, and many other desirable features. ? 2000 by CRC Press LLC VSELP: Vector Sum Excited Linear Predictive coding. 8 Kbps linear predictive coder used in IS-54/IS-136 TDMA systems. W-CDMA: Wideband CDMA. 3rd Generation wireless standard currently being developed in Europe and Japan. References Garg, Vijay K., Kenneth Smolik, and Joseph E. Wilkes, Applications of CDMA in Wireless/Personal Communi- cations, Upper Saddle River, NJ: Prentice-Hall, 1997. Harte, Lawrence J., Adrian D. Smith, and Charles A. Jacobs, IS-136 TDMA Technology, Economics, and Services, Boston: Artech, 1998. Rappaport, Theodore S., Wireless Communications: Principles and Practice, Upper Saddle River, NJ: Prentice- Hall, 1996. Further Information For further information on 1st generation wireless, see the AMPS standard: EIA-553 Mobile Station — Land Station Compatibility Specification. Electronics Industry Association. September, 1989. For further information on 2nd generation wireless, see: TIA/EIA/IS-136-A TDMA Cellular/PCS-Radio Interface-Mobile Station-Base Station Compatibility, Telecommu- nications Industry Association, October, 1996, (TDMA standard). TIA/EIA/IS-95-A Mobile Station — Base Station Compatibility Standard for Dual-Mode Wideband Spread Spec- trum Cellular System, Telecommunications Industry Association, May, 1995 (CDMA standard). For further information on 3rd generation wireless, see: Dennett, Steve, The cdma2000 ITU-R RTT Candidate Submission, Telecommunications Industry Associations, July 28, 1998. Meche, Paul, Updated UWC-136 RTT, Telecommunications Industry Association, September 28, 1998. Ojanpera, Tero and Steven D. Gray, An Overview of cdma2000, WCDMA, and EDGE, The Mobile Communi- cations Handbook, Ed. Jerry D. Gibson, Boca Raton, FL: CRC Press, 1999, Ch. 36. ? 2000 by CRC Press LLC