The Fundamentals of Wireless Communications
A. Principles of Wireless Communications
Radio, or the use of radiated electromagnetic waves, is the only practical way of communicating with people or vehicles that move around on land, on the sea, in the air, or in outer space. It is the use of electromagnetic waves that permits the transmission and reception of information over a distance without the use of wires. The distance covered may range from only a few feet in the case of a cordless telephone to millions of miles in the case of a space probe.
In principle, radio communications is a relatively straightforward process. At the transmitting or sending end, the information to be sent (e.g., a voice signal) is imposed on a locally generated radio frequency (RF) signal called a carrier. The process of imposing the information signal on the carrier is called modulation. This carrier signal, along with the information signal imposed on it, is then radiated by an antenna. The frequency of an electromagnetic or radio wave is simply its oscillation rate measured in cycles-per-second or Hertz. The range of radio frequencies useful for practical communications starts at a few thousand Hertz (Hz) and goes up to a few hundred billion Hertz.
At the receiving end, the signal is picked up by another antenna and fed into a receiver where the desired carrier with the imposed information signal is selected from among all of the other signals impinging on the antenna. The information signal (e.g., voice) is then extracted from the carrier in a process referred to as demodulation. Thus modulation of the carrier wave occurs at the transmitter (the emitter of the radiation) and demodulation occurs at the receiver. These same basic steps or processes can be identified in radio systems ranging from the cheapest cordless telephone with a very low power transmitter and simple antenna to a high power transmitter carrying multiple information signals and utilizing complex, directive antennas.
A pure, unmodulated radio carrier conveys no information and occupies only an infinitesimal amount of the spectrum. Modulation of the radio signal inevitably causes a spreading of the radio wave in frequency. Thus a radio signal conveying information occupies a range of frequencies called a channel. In general, the more information that is sent per unit of time, the wider the channel must be.
As the radio wave expands in surface area after leaving the antenna, it grows weaker and weaker. At the receiver, the signal must still be strong enough to overcome any local radio noise or interference; otherwise the transmission will not be successful. In outer space, where there are no intervening hills or mountains, natural foliage, or man-made objects such as buildings with which to contend, the weakening of the signal with distance from the transmitting antenna can be predicted with great precision. In terrestrial radio systems, the environment for transmission is much more complex. It is even worse in mobile systems -- where one or both of the terminals (transmitters and receivers) can move about -- due to an environment that changes dynamically from moment to moment.
One major effect that appears in the terrestrial environment is multipath. Multipath is produced when the radio wave not only travels directly from the transmitting antenna to the receiving antenna, but is also reflected off of other physical objects such as buildings or mountains. At some locations, the signals traveling by different paths may add up to make the signal stronger, while at other locations, just a short distance away, the signals can cancel one another, causing the signal to fade. This effect is referred to as multipath fading. In addition, a large building or mountain between the transmitter and receiver may block the signal entirely, producing another type of fading.
Thus, in a terrestrial mobile environment, the communications engineer must not only take into account the natural weakening of the signal with distance (the so-called free space loss), but also the rapid changes in signal strength caused by multipath fading, the fading caused by shadowing, as well as the additional weakening of the signal produced when customers use portable units inside buildings or vehicles. In doing so, the communications engineer uses complex computer models and field measurements to determine design parameters (e.g., transmitter power and antenna heights) to ensure that service is adequate over the desired coverage area.
As stated above, in addition to extracting the information from the radio wave through demodulation, it is also a principle function of a receiver to accept only the information in the chosen channel and reject other information being sent simultaneously in other (e.g., adjacent) channels. The measure of the receiver's ability to reject interfering signals on other channels is referred to as its selectivity. Hence, two or more radio systems can use the radio spectrum in the same area at the same time as long as (a) they are separated sufficiently in terms of frequency -- i.e., so that their channels do not overlap, and (b) the receivers involved have sufficient selectivity to reject the signals on adjacent channels.
If two radio systems do occupy the same channel, they must either time share the channel in some way or be separated enough in distance to not cause interference to one another at the desired reception points. In other words, the receiver must be close enough to the desired transmitter location, and far enough from the undesired or interfering transmitter, to ensure the strength of the desired signal relative to the strength of the undesired signal is great enough to provide the needed quality. Generally speaking, because of the drop in signal strength with distance, the further the receiver is from the desired transmitter, the further away the undesired transmitter must be to prevent harmful interference.
In a traditional, two-way radio system used by taxicab companies, for example, where the desired radius of coverage around the base station transmitter is, say, 20 miles, the interfering transmitter must be something like 70 miles away. This is often referred to as the frequency reuse distance. As will be described in more detail in later sections, more intense frequency reuse is extremely important in modern Personal Communications Service (PCS) systems as a way of increasing capacity. That is, by keeping the ranges and, hence the reuse distances, short, the same channel can be reused many times for different conversations in the same geographic area. In summary, at a very basic level, the radio spectrum resource can be shared by many simultaneous users by taking into account its frequency, space, and time dimensions. All of these dimensions are exploited heavily in modern wireless systems.
B. Types of Signals and More Details on Modulation
There are two basic types of signals -- analog and digital. An analog signal is a signal that varies continuously between a maximum and minimum value. At a given instant, an analog signal can assume any one of an infinite number of values between the two extremes. Examples of analog signals include the human voice or other measurable values in the physical universe such as the temperature of a boiler. A digital signal, in contrast, does not take on a continuous set of values. Rather, at a given instant of time, it takes one of a limited set of values called a symbol. A sequence of such values or symbols can be used to represent a number or alphabetical characters. Examples of digital signals include the presence or absence of a current pulse on a wire or a light pulse on a fiber optic cable. In this example, the pulses can be interpreted as binary digits or bits, and particular sequences of bits can be uniquely defined to correspond to numbers or alphanumeric characters. A communications system can be either analog or digital (or a combination of the two); that is, the information can be transmitted in either the analog or digital form within the network itself.
As described before, the carrier signal in a radio system is characterized by its frequency measured in Hertz. In addition to its frequency, the carrier is also characterized by the amplitude or strength of the wave and by its phase. Modulation of the carrier wave is accomplished by varying any or all of these characteristics in a known relationship to the information signal. For example, the amplitude of an analog information signal can be used to vary the amplitude of the carrier wave in a process known as amplitude modulation. Or the amplitude of the analog information signal can be used to vary the frequency of the carrier wave in a process known as frequency modulation. At the receiver, these amplitude or frequency variations in the carrier wave are used to extract the information signal in the demodulation process. These are examples of an analog communications system. In ordinary amplitude modulation, the channel width must be twice the highest frequency present in the information signal. In frequency modulation, the channel width is typically several times the highest frequency present in the information signal.
In the land mobile radio field, which is the focus of this report, the predominant modulation technique has been frequency modulation. At first glance, it might appear that frequency modulation makes inefficient use of the radio spectrum resource since the channel width required is much greater than for amplitude modulation. However, the actual situation is much more complex because, as a general proposition, signals that are spread over wider channels are more resistant to noise and interference. Without going into a lot of technical details, suffice it to state here that the wider the signal being transmitted relative to the width of the information, the greater the ability of the system to suppress noise and interference (including multipath). Thus, there is a tradeoff between transmitted bandwidth and noise and interference resistance. This improvement in performance is particularly useful in (a) the hostile radio signal environment described earlier and (b) rejecting interference from distant transmitters.
Digital signals can also be transmitted over radio systems by varying any of the three parameters described -- frequency, amplitude or phase. The earliest form of digital modulation, Morse Code, simply turned the transmitter on and off to form dots and dashes that could be interpreted by human operators as symbols (e.g., letters). The on and off pulses or bits that comprise a modern digital signal could be sent the same way, i.e., by turning the transmitter on to signify a one and off to signify a zero. But because of the ever-present fading on radio paths, the receiver would not be able to reliably determine whether a zero had been sent or the signal was simply in a fade. Thus the more reliable way is to transmit one frequency to signify a one and another frequency to signify a zero. This is referred to as Frequency Shift Keying (FSK). Modern digital systems use combinations of frequency, amplitude and phase modulation to increase the number of bits that can be transmitted in a given channel.
As will be described in more detail later, the trend in wireless systems (just as in wireline networks) is toward digital systems and the use of advanced forms of digital modulation. Digital systems have a number of important advantages including the fact that digital signals are more immune to noise and, unlike analog systems, even when noise has been picked up, any resulting errors in the digital bit stream can be detected and corrected. Moreover, digital signals can be easily manipulated or processed in useful ways using modern computer techniques. While it is easy to envision how digital information signals are sent over digital communications systems, the method of sending analog signals (like voice) over a digital communications system and reproducing them at the other end is not as obvious.
In a digital system, the analog signal is digitized in an analog-to-digital converter to make it compatible with digital transmission. That is, the analog signal is converted into a sequence of bits that accurately describe the analog signal. More specifically, (a) the analog signal is sampled at sufficiently close intervals to accurately reproduce the signal's shape, (b) the amplitudes of the samples (the amplitude of the analog signal at particular instants of time) are quantized or given an approximate value according to the range within which the amplitude falls, and (c) these amplitude values are then encoded as a sequence of bits representing the corresponding binary number. At the receiving end, the analog signal is reconstructed from the sequence of bits that describe the amplitude of the signal at each instant of time. Thus, in a digital cellular system, for example, a voice signal is first converted into a digital signal and is then carried over digital transmission facilities that employ one of the advanced forms of digital modulation of the carrier wave described above.
The technique for converting an analog signal to a digital signal as just described is known as waveform coding. One popular form of waveform coding, called Pulse Code Modulation converts speech into a digital bit stream operating at 64,000 bits per second (bps) or 64 kbps. Another form of waveform coding is known as Adaptive Differential Pulse Code Modulation (ADPCM) and it operates at 32 kbps. It is possible to reduce the number of bits per second it takes to describe a voice signal by taking advantage of the known characteristics of the human voice. These techniques are known as voice coding and the devices employed to implement the techniques are called vocoders. It is beyond the scope of this report to describe the different techniques used for voice coding; suffice it to state that these techniques all use computer processing power to remove redundancy from speech so that (a) fewer bits per second have to be transmitted to convey a voice signal and (b) the available bandwidth can be used more efficiently. Some of these techniques allow voice to be sent at rates as low as 4 kbps (and even lower), compared with the 64 kbps or 32 kbps associated with waveform coding.
At very low rates, the quality deteriorates and the reproduced voice signal takes on a computer generated-like sound. In addition, the compressing and decompressing of the signal takes time, even with fairly powerful processors. At high levels of compression, the resulting delays can be annoying to end users. Thus, there is a tradeoff between the bit rate, the quality of the voice reproduction (including delay), and the amount of computer processing power employed within the transmitter and receiver. The latter not only has implications for cost, but also for battery life as well, since more processing power translates into increased battery drain. This is a particularly important consideration in portable units.
Most operators of commercial wireless telephone systems have a strong incentive to employ digital voice compression because the lower bit rates translate into (a) more conversations in a given amount of spectrum -- i.e., more efficient use of the radio spectrum and greater capacity, and (b) more conversations per piece of radio equipment and/or radio site -- i.e., greater economies of scale. Moreover, the FCC generally encourages the increased spectrum efficiency that results from voice compression. It is important to stress that the use of voice compression can cause problems for non-voice signals such as those emitted by fax machines, computer modems, and, especially important in the context of this report, text telephones/TTYs. This is because, as described above, vocoders depend critically upon the signal having the characteristics of the human voice. While the range of audio frequencies is the same, anyone who has listened to a fax machine, modem, or TTY on a telephone line knows that it does not sound like the human voice. The use of vocoders in wireless telecommunications and the implications of that use for the deaf community will be discussed in more detail in later sections.
C. Licensed Bands Available and Their Technical Characteristics
Domestically, the FCC has set aside certain bands or ranges of radio frequencies for land mobile radio use. These bands include Low Band in the 40 MHz region of the spectrum, High Band in the 150 MHz region, a band near 220 MHz, the UHF band in the 450 MHz region, a band near 800/900 MHz, and, most recently, a band near 1.9 GHz. With the exception of the 220 MHz band, which was recently set aside for land mobile radio use, the FCC has historically opened up bands higher up in the spectrum as the lower bands have become more congested. Thus, following World War II, most wireless mobile activity was centered in Low Band. However, in response to rapid growth in the land mobile radio service, the FCC, over the intervening years, has steadily increased the amount of spectrum available through successive reallocations of the resource in the higher frequency ranges.
Moving higher in frequency to avoid congestion has advantages and disadvantages. Generally speaking, the radio frequency (RF) devices employed within the system get more costly the higher the frequency and, in terms of propagation effects, the higher frequencies are subject to more blocking or shadowing by buildings or hills. However, the higher frequencies tend to penetrate buildings more readily and the antennas involved are physically smaller -- both important attributes for systems that seek to serve small portable units carried on one's person. At some risk of over-generalizing, it can be said that (a) the lower frequency bands are best for economically covering wide areas in suburban and rural areas where frequency reuse is not as important and (b) the higher frequency bands are best for covering urban areas where building penetration and high levels of frequency reuse are desired.
D. Multiple Access and Duplexing Techniques
If spectrum were unlimited and the radio equipment used in the infrastructure were free, everyone could have their own wireless channel within one of the bands set aside by the FCC for land mobile radio use. But spectrum is not unlimited and the backbone equipment is not free. Thus it is imperative that the spectrum and, often, the backbone equipment, be shared among users. In short, users in a given area must contend for a limited number of channels.
There are different ways of dividing up the spectrum and providing users access to it in an organized way. The simplest and most straightforward method is known as frequency division multiple access (FDMA). With FDMA, the available spectrum is divided into non-overlapping slots in the frequency dimension or domain. These frequency slots or channels are then put into a pool and assigned to users on either a manual or automated basis for the duration of their particular call. For example, a 150 kHz block of spectrum could be divided into 6 channels or frequency slots each 25 kHz wide. Such an arrangement would allow six simultaneous conversations to take place, each with their own carrier within their own frequency slot. In the example, this would mean that each user would be continuously accessing one-sixth of the available spectrum during the duration of the conversation. FDMA is perhaps the most familiar way of dividing up spectrum, and it has traditionally been associated with analog systems.
With TDMA, the available spectrum is divided into non-overlapping time slots in the time dimension or domain. These time slots or channels are then put into a pool and assigned to users for the duration of their particular call. To continue the example given above, in a TDMA system the 150 kHz of spectrum would be divided into recurring groups (frames) of six time slots, and each time slot would carry a sequence of bits representing a portion of one of six simultaneous conversations. The six conversations each take turns using the available capacity. In other words, each user would be accessing all of the available spectrum but only for one-sixth of the available time. Rather than each signal having a particular frequency slot as in FDMA, in TDMA each conversation occupies a particular time slot in a sequential fashion. The frames are repeated fast enough that there is no interruption or delay in the conversation as seen by the end user.
Note that, theoretically at least, there is no difference in capacity between FDMA and TDMA as seen by the end user. Namely, you get access to one-sixth of the capacity all of the time or all of the capacity one-sixth of the time to continue the example. Note further that, in the practical world, digital systems are typically a combination of FDMA and TDMA. In other words, the systems are designed so that the capacity is divided into both the frequency and time dimensions whereby a user contends for a particular channel and then a time slot within that channel.
A third access method is known as Code Division Multiple Access (CDMA). CDMA is both a modulation and an access technique that is based upon the spread-spectrum concept. A spread-spectrum system is one in which the bandwidth occupied by the signal is much wider than the bandwidth of the information signal being transmitted. For example, a voice conversation with a bandwidth of just 3 kHz or so would be spread over 1 MHz or more of spectrum.
In spread spectrum systems, multiple conversations (up to some maximum) simultaneously share the available spectrum in both the time and frequency dimensions. Hence, in a CDMA system, the available spectrum is not channelized in frequency or time as in FDMA and TDMA systems, respectively. Instead, the individual conversations are distinguished through coding; that is, at the transmitter, each conversation channel is processed with a unique spreading code that is used to distribute the signal over the available bandwidth. The receiver uses the unique code to accept the energy associated with a particular code. The other signals present are each identified by a different code and simply produce background noise. In this way, many conversations can be carried simultaneously within the same block of spectrum.
Before going on to discuss the types of services in the wireless field, one further technical topic must be addressed, and that is "duplexing." In many, if not most, communication systems, it is desirable to be able to communicate in both directions at the same time. This system characteristic, which is known as full-duplex operation, is desirable because it lets one party in a voice conversation interrupt the other with a question or one device to immediately request a retransmission of a block of information received in error during a data communications session. There are two basic ways of providing for full-duplex operation in a radio system. By far the most common is to assign two different frequency slots per conversation -- one for transmitting and one for receiving. By separating the slots sufficiently in frequency, filters (say in the portable radio) can be used to prevent the transmitted information from interfering with the simultaneously received information. Thus, in many land mobile radio bands, a channel actually consists of two frequency slots -- one for each direction of transmission in a full-duplex conversation. This arrangement is called Frequency Division Duplexing (FDD).
Another much less common means of achieving full-duplex operation in the digital world is through what is called time division duplexing (TDD). In TDD, a single (unpaired) channel is used with each end taking turns transmitting. Each end sends a burst of information (consisting of bits representing a few samples of the voice signal, for example) and then receives a burst from the other end. As in the case of the TDMA access technique, this process is repeated rapidly enough that the end user does not perceive any gaps or delays in what is heard. To the end user it appears as a true full-duplex connection.
E. Types of Services
Although there are many sub-markets and niches, the land mobile market can be divided into four traditional segments serving four different applications. These four segments or applications are (a) one-way paging or messaging, (b) two-way dispatch, (c) two-way mobile/portable telephone, and (d) two-way data or messaging. Understanding these different applications is important from a technological perspective because they all have different requirements. This means, among other things, that a network or system optimized for one application may not be optimal for another. Thus, one can observe in the marketplace standalone systems optimized for dispatch service, for paging, for interconnected mobile telephone service, and for two-way data/messaging. One can also observe systems that attempt to capture economies of scope by offering combinations of these services on a common infrastructure. In the following few paragraphs, each application will be briefly described.
One-way paging or messaging uses a radio signal to merely alert or to instruct the user to do something. The user (an office equipment repair person or a doctor, for example) carries a very small device to receive the one-way messages. Often this device is referred to as a "pager" or a "beeper." There are four types of paging services -- tone-only, tone-voice, numeric, and alphanumeric. In the tone-only system, the receiver simply emits a tone which alerts the user to take some predetermined action such as calling their office or answering service. In the tone-voice system, the tone is followed by a short voice message entered by the person placing the page. In the numeric system, a short numeric message is sent and displayed on a small screen on the receiver. A typical numeric message might be the telephone number that the user is supposed to call. An alphanumeric system is similar except that it allows a more complex text message to be delivered. Paging, like cellular mobile radio systems, has exhibited rapid growth.
Two-way dispatch is another basic land mobile radio service. It involves communications between and among a dispatcher and units (mobiles and/or portables) in the field. It is typically a "command and control" system where a high degree of coordination among the units is required. Such services are used heavily by the public safety community and by businesses like tow truck and taxicab companies that must dispatch units operated away from the principal place of business. There is typically a requirement for the dispatcher to be able to reach multiple units simultaneously in what is referred to as group or fleet calling (i.e., one-to-many communications). The messages are typically of short duration (tens of seconds) and efficiency and other considerations dictate rapid call setup. Push-to-talk and release-to-listen (PTT/RTL) and half-duplex operation are common. In its pure form, dispatch communications does not involve interconnection with the Public Switched Telephone Network (PSTN) and, in many applications, such interconnection is neither needed nor desired.
Two-way mobile telephone is another basic land mobile radio service. It allows the user to place and receive ordinary telephone calls (i.e., one-to-one communications) and, obviously, provision must be made for interconnection with the PSTN. The messages are typically of much longer duration (compared to dispatch calls) and users typically demand full-duplex operation. Because the call itself is of longer duration, call set-up delay is less critical. This service need not be described in detail, since the basic notion is to duplicate the operation of the ordinary telephone network, but with wireless telephones or handsets.
The fourth and final basic land mobile radio service is two-way data messaging. This service is of more recent origin. It facilitates various forms of data communications such as computer aided dispatch, electronic messaging/mail, telemetry, and computer-to-computer communications on a wireless basis. The data traffic on such networks is typically "bursty" in nature, and errors cannot be tolerated in many critical applications. On the other hand, unlike with voice applications, the systems are typically tolerant of transmission delays of up to several seconds. These wireless data communications services are used, for example, by package delivery services to track packages and to schedule pickups and deliveries.
Customers typically have differing requirements for the four services. Some users may only need one-way paging or mobile telephone service, some may need dispatch and mobile telephone, while others may have a need for all four. Systems to provide these services typically started out on a separate, standalone basis and systems are still evolving. At the same time, as mentioned earlier, systems are also evolving that try to offer an integrated set of services on a common infrastructure or platform.
It should be pointed out that these services can be (and are) provided on both a private and third-party (e.g., common carrier or other commercial) basis. For example, a user can purchase and operate a radio system and provide dispatch communications to its own fleet or purchase the service from a Specialized Mobile Radio operator who provides the services on a commercial, for-hire basis. As noted in Section I.D., third parties who offer interconnected services on a commercial, for-hire basis are categorized as Commercial Mobile Radio Service (CMRS) providers.
