27
CHAPTER
Wireless Transmitters and Receivers
IN WIRELESS COMMUNICATIONS, A TRANSMITTER CONVERTS DATA INTO ELECTROMAGNETIC (EM) WAVES intended for recovery by one or more receivers. In this chapter, we’ll learn how to convert data to an EM field, and then learn how we can intercept and decode that field at remote points.
Modulation
When we modulate a wireless signal, we “write” data onto an EM wave. We can carry out this process by varying the amplitude, the frequency, or the phase of the wave. We can also obtain a modulated signal by generating a series of multiple-wave pulses and varying their duration, amplitude, or timing. The heart of a wireless signal comprises a sine wave called the carrier whose frequency can range from a few kilohertz (kHz) to many gigahertz (GHz). If we expect effective data transfer, the carrier frequency must be at least 10 times the highest frequency of the modulating signal.
On/Off Keying
The simplest form of modulation involves on/off keying of the carrier. We can key the oscillator of a radio transmitter to send Morse code, one of the simplest known binary digital modulation modes. The duration of a Morse-code dot equals the duration of one binary digit, more often called a bit. (A binary digit is the smallest or shortest possible unit of data in a system whose only two states are “on” and “off.”) A dash measures three bits in duration. The space between dots and dashes within a character equals one bit; the space between characters in a word equals three bits; the space between words equals seven bits. Some technicians refer to the key-down (full-carrier) condition as mark and the key-up (no-signal) condition as space. Amateur radio operators who enjoy using the Morse code send and receive it at speeds ranging from about 5 words per minute (wpm) to around 60 wpm.
Frequency-Shift Keying
We can send digital data faster and with fewer errors than Morse code allows if we use frequency-shift keying (FSK). In some FSK systems, the carrier frequency shifts between mark and space conditions, usually by a few hundred hertz or less. In other systems, a two-tone audio-frequency (AF) sine wave modulates the carrier, a mode known as audio-frequency-shift keying (AFSK). The two most common codes used with FSK and AFSK are Baudot (pronounced “baw-DOE”) and ASCII (pronounced “ASK-ee”). The acronym ASCII stands for American Standard Code for Information Interchange.
In radioteletype (RTTY), FSK, and AFSK systems, a terminal unit (TU) converts the digital signals into electrical impulses that operate a teleprinter or display characters on a computer screen. The TU also generates the signals necessary to send RTTY when an operator types on a keyboard. A device that sends and receives AFSK is sometimes called a modem, an acronym that stands for modulator/demodulator. A modem is basically the same as a TU. Figure 27-1 is a block diagram of an AFSK transmitter.
27-1 Simplified block diagram of a transmitter that uses audio-frequency-shift keying (AFSK).
The main reason why FSK or AFSK work better than on/off keying is the fact that the space signals are identified as such, rather than existing as mere gaps in the data. A sudden noise burst in an on/off keyed signal can “confuse” a receiver into falsely reading a space as a mark, but when the space is positively represented by its own signal, this type of error happens less often at any given data speed.
Amplitude Modulation
An AF voice signal has frequencies mostly in the range between 300 Hz and 3 kHz. We can modulate some characteristic of an RF carrier with an AF voice waveform, thereby transmitting the voice information over the airwaves. Figure 27-2 shows a simple circuit for obtaining amplitude modulation (AM). We can imagine this circuit as an RF amplifier for the carrier, with the instantaneous gain dependent on the instantaneous audio input amplitude. We can also think of the circuit as a mixer that combines the RF carrier and the audio signals to produce sum and difference signals above and below the carrier frequency.
27-2 An amplitude modulator using an NPN bipolar transistor.
The circuit shown in Fig. 27-2 performs well as long as we don’t let the AF input amplitude get too great. If we inject too much audio, we get distortion (nonlinearity) in the transistor resulting in degraded intelligibility (understandability), reduced circuit efficiency (ratio of DC power input to useful power output), and excessive output signal bandwidth (the difference between the highest and lowest component frequency). We can express the modulation extent as a percentage ranging from 0%, representing an unmodulated carrier, to 100%, representing the maximum possible modulation we can get without distortion. If we increase the modulation beyond 100%, we observe the same problems as we do when we apply excessive AF input to a modulator circuit, such as the one shown in Fig. 27-2. In an AM signal that’s modulated at 100%, we find that ⅓ of the signal power conveys the data, while the carrier wave consumes the other ⅔ of the power.
Figure 27-3 shows a spectral display of an AM voice radio signal. The horizontal scale is calibrated in increments of 1 kHz per division. Each vertical division represents 3 dB of change in signal strength. The maximum (reference) amplitude equals 0 dB relative to 1 mW (abbreviated as 0 dBm). The data exists in sidebands above and below the carrier frequency. These sidebands constitute sum and difference signals produced by mixing in the modulator circuit between the audio and the carrier. The RF energy between −3 kHz and the carrier frequency is called the lower sideband (LSB); the RF energy from the carrier frequency to +3 kHz is called the upper sideband (USB).
27-3 Spectral display of a typical amplitude-modulated (AM) voice communications signal.
The signal bandwidth equals the difference between the maximum and minimum sideband frequencies. In an AM signal, the bandwidth equals twice the highest audio modulating frequency. In the example of Fig. 27-3, all the AF voice energy exists at or below 3 kHz, so the signal bandwidth equals 6 kHz, typical of AM voice communications. In standard AM broadcasting in which music is transmitted along with voices, the AF energy is spread over a wider bandwidth, nominally 10 kHz to 20 kHz. The increased bandwidth provides for better fidelity (sound quality).
Single Sideband
In AM at 100% modulation, the carrier wave consumes ⅔ of the signal power, and the sidebands exist as mirror-image duplicates that, combined, employ only ⅓ of the signal power. These properties make AM inefficient and needlessly redundant.
If we could get rid of the carrier and one of the sidebands, we’d still convey all the information we want while consuming far less power. Alternatively, we could get a stronger signal for a given amount of RF power. We could also reduce the signal bandwidth to a little less than half that of an AM signal modulated with the same data. The resulting spectrum savings would allow us to fit more than twice as many signals into a specific range, or band, of frequencies. During the early twentieth century, communications engineers perfected a way to modify AM signals in this way. They called the resulting mode single sideband (SSB), a term which endures to this day.
When we remove the carrier and one of the sidebands from an AM signal, the remaining energy has a spectral display resembling Fig. 27-4. In this case, we eliminate the USB along with the carrier, leaving only the LSB. We could just as well remove the LSB along with the carrier, leaving only the USB.
27-4 Spectral display of a typical single-sideband (SSB) voice communications signal, in this case lower sideband (LSB).
Balanced Modulator
We can almost completely suppress the carrier in an AM signal using a balanced modulator—an amplitude modulator/amplifier using two transistors with the inputs connected in push-pull and the outputs connected in parallel, as shown in Fig. 27-5. This arrangement “cancels” the carrier wave in the output signals, leaving only LSB and USB energy. The balanced modulator produces a double-sideband suppressed-carrier (DSBSC) signal, often called simply double sideband (DSB). One of the sidebands can be suppressed in a subsequent circuit by a bandpass filter to obtain an SSB signal.
27-5 A balanced modulator using two NPN bipolar transistors. We connect the bases in push-pull and the collectors in parallel.
Basic SSB Transmitter
Figure 27-6 is a block diagram of a simple SSB transmitter. The RF amplifiers that follow any type of amplitude modulator, including a balanced modulator, must all operate in a linear manner to prevent distortion and unnecessary spreading of the signal bandwidth, a condition that some engineers and radio operators call splatter. These amplifiers generally work in class A except for the PA, which operates in class AB or class B. We’ll never see a class-C amplifier as the PA in an SSB transmitter because class-C operation distorts any signal whose amplitude varies over a continuous range.
27-6 Block diagram of a basic SSB transmitter.
Frequency Modulation
In frequency modulation (FM), the instantaneous signal amplitude remains constant; the instantaneous frequency varies instead. We can get FM by applying an audio signal to the varactor diode in a voltage-controlled oscillator (VCO). Figure 27-7 shows an example of this scheme, known as reactance modulation. This circuit employs a Colpitts oscillator, but we could use any other type of oscillator and get similar results. The varying voltage across the varactor causes its capacitance to change in accordance with the audio waveform. The fluctuating capacitance causes variations in the resonant frequency of the inductance-capacitance (LC) tuned circuit, causing small fluctuations in the frequency generated by the oscillator.
27-7 Generation of a frequency-modulated (FM) signal by employing reactance modulation in a Colpitts oscillator. We can modify other oscillator types in a similar way.
Phase Modulation
We can indirectly obtain FM if we modulate the phase of the oscillator signal. When we vary the phase from instant to instant, we inevitably provoke small variations in the frequency as well. Any instantaneous phase change shows up as an instantaneous frequency change (and vice-versa). When we employ phase modulation (PM), we must process the audio signal before we apply it to the modulator, adjusting the frequency response of the audio amplifiers. Otherwise the signal will sound muffled when we listen to it in a receiver designed for ordinary FM.
Deviation for FM and PM
In an FM or PM signal, we can quantify the maximum extent to which the instantaneous carrier frequency differs from the unmodulated-carrier frequency in terms of a parameter called deviation. For most FM and PM voice transmitters, the deviation is standardized at ±5 kHz. We call this mode narrowband FM (NBFM). The bandwidth of an NBFM signal roughly equals that of an AM signal containing the same modulating information. In FM hi-fi music broadcasting, and in some other applications, the deviation exceeds ±5 kHz, giving us a mode called wideband FM (WBFM).
The deviation obtainable with FM is greater, for a given oscillator frequency, than the deviation that we get with PM. However, we can increase the deviation of any FM or PM signal with the help of a frequency multiplier. When the signal passes through a frequency multiplier, the deviation gets multiplied along with the carrier frequency. The deviation in the final output should equal the highest modulating audio frequency if we expect optimum audio fidelity. Therefore, ±5 kHz is more than enough deviation for voice communications. For music, a deviation of ±15 kHz or ±20 kHz is required for good reproduction.
Modulation Index for FM and PM
In any FM or PM signal, the ratio of the frequency deviation to the highest modulating audio frequency is called the modulation index. Ideally, this figure should be somewhere between 1:1 and 2:1. If it’s less than 1:1, the signal sounds muffled or distorted, and efficiency is sacrificed. Increasing the modulation index much beyond 2:1 broadens the bandwidth without providing significant improvement in intelligibility or fidelity.
Power Amplification for FM and PM
A class-C PA can function in an FM or PM transmitter without causing distortion because the signal amplitude remains constant. Nonlinearity (characteristic of class-C operation) has no meaning, let alone any adverse effects, when the signal amplitude never changes! For this reason, we’ll often find class-C PAs in FM and PM transmitters, especially those that have high output power. Remember: Class C offers the best efficiency of any PA mode!
Pulse-Amplitude Modulation
We can modulate a signal by varying some aspect of a constant stream of signal pulses. In pulse-amplitude modulation (PAM), the strength of each individual pulse varies according to the modulating waveform. In this respect, PAM resembles AM. Figure 27-8A shows an amplitude-versus-time graph of a hypothetical PAM signal. The modulating waveform appears as a dashed curve, and the pulses appear as vertical gray bars. Normally, the pulse amplitude increases as the instantaneous modulating-signal level increases (positive PAM). But this situation can be reversed, so higher audio levels cause the pulse amplitude to go down (negative PAM). Then the signal pulses are at their strongest when there is no modulation. The transmitter works harder to produce negative PAM than it does to produce positive PAM.
27-8 Time-domain graphs of various modes of pulse modulation. At A, pulse-amplitude modulation (PAM); at B, pulse-width modulation (PWM), also called pulse-duration modulation (PDM); at C, pulse-interval modulation (PIM); at D, pulse-code modulation (PCM).
Pulse-Width Modulation
We can modulate the output of an RF transmitter output by varying the width (duration) of signal pulses to obtain pulse-width modulation (PWM), also known as pulse duration modulation (PDM) as shown in Fig. 27-8B. Normally, the pulse width increases as the instantaneous modulating-signal level increases (positive PWM). But this situation can be reversed (negative PWM). The transmitter must work harder to accomplish negative PWM. Either way, the peak pulse amplitude remains constant.
Pulse-Interval Modulation
Even if all the pulses have the same amplitude and the same duration, we can obtain pulse modulation by varying how often the pulses occur. In PAM and PWM, we always transmit the pulses at the same time interval, known as the sampling interval. But in pulse interval-modulation (PIM), pulses can occur more or less frequently than they do under conditions of no modulation. Figure 27-8C shows a hypothetical PIM signal. Every pulse has the same amplitude and the same duration, but the time interval between them changes. When there is no modulation, the pulses emerge from the transmitter evenly spaced with respect to time. An increase in the instantaneous data amplitude might cause pulses to be sent more often, as is the case in Fig. 27-8C (positive PIM). Alternatively, an increase in instantaneous data level might slow down the rate at which the pulses emerge (negative PIM).
Pulse-Code Modulation
In digital communications, the modulating data attains only certain defined states, rather than continuously varying. Compared with old-fashioned analog communications (where the state is always continuously variable), digital modes offer improved signal-to-noise (S/N) ratio, narrower signal bandwidth, better accuracy, and superior reliability. In pulse-code modulation (PCM), any of the above-described aspects—amplitude, width, or interval—of a pulse sequence (or pulse train) can be varied. But rather than having infinitely many possible states, the number of states equals some power of 2, such as 22 (four states), 23 (eight states), 24 (16 states), 25 (32 states), 26 (64 states), and so on. As we increase the number of states, the fidelity and data transmission speed improve, but the signal gets more complicated. Figure 27-8D shows an example of eight-level PCM.
Analog-to-Digital Conversion
Pulse-code modulation, such as we see in Fig. 27-8D, serves as a common form of analog-to-digital (A/D) conversion. A voice signal, or any continuously variable signal, can be digitized, or converted into a train of pulses whose amplitudes can achieve only certain defined levels.
In A/D conversion, because the number of states always equals some power of 2, we can represent the signal as a binary-number code. Fidelity improves as the exponent increases. The number of states is called the sampling resolution, or simply the resolution. A resolution of 23 = 8 (as shown in Fig. 27-8D) is good enough for basic voice communications. A resolution of 24 = 16 can provide fairly decent music reproduction.
The efficiency with which we can digitize a signal depends on the frequency at which we carry out the sampling. In general, the sampling rate must be at least twice the highest data frequency. For an audio signal with components as high as 3 kHz, the minimum sampling rate for effective digitization is 6 kHz. For music it’s higher, of course.
Image Transmission
Non-moving images can be sent within the same bandwidth as voice signals. For high-resolution, moving images such as video files, we need more bandwidth than we do for low-resolution, non-moving images such as simple drawings.
Facsimile
Non-moving images (also called still images) are commonly transmitted by facsimile, also called fax. If we send the data slowly enough, we can render as much detail as we want within a 3-kHz-wide band, the standard for voice communications. This flexibility explains why detailed fax images can be sent over a plain old telephone service (POTS) line.
In an electromechanical fax transmitter, a paper document is wrapped around a drum. The drum rotates at a slow, controlled rate. A spot of light scans laterally across the document. The drum moves the document so that a single line is scanned with each pass of the light spot. This process continues, line by line, until the device has scanned the complete frame (image). A photodetector picks up the light rays reflected from the paper. Dark portions of the image reflect less light than bright parts, so the current through the photodetector varies as the light beam passes over various regions. This current modulates a carrier in one of the modes described earlier, such as AM, FM, or SSB. Typically, the data for pure black corresponds to a 1.5-kHz audio sine wave, and the data for pure white corresponds to a 2.3-kHz audio sine wave. Gray shades produce audio sine waves having frequencies between these extremes.
A fax receiver duplicates the transmitter’s scanning rate and pattern, and a display or printer reproduces the image in grayscale (shades of gray ranging from black to white, without color).
Slow-Scan Television
We can think of slow-scan television (SSTV) as a fast, repetitive form of fax. An SSTV signal, like a fax signal, propagates within a band of frequencies as narrow as that of a human voice. And, like fax, SSTV transmission reproduces only still images, not moving ones. However, an SSTV system scans and transmits a full image in much less time than a fax machine does. The time required to send a complete frame (image or scene) is only eight seconds, rather than a minute or more. Therefore, we can send multiple images in a reasonable amount of time, giving our viewers some sense of motion in a scene. This speed bonus comes with a tradeoff: in SSTV, we get lower resolution, meaning less image detail, than we get with fax.
We can program a personal computer so that its monitor will act as an SSTV display. We can also find converters that allow us to look at SSTV signals on a conventional TV set.
An SSTV frame contains 120 lines. The black and white frequencies are the same as those for fax transmission; the darkest parts of the picture are sent at 1.5 kHz and the brightest parts are sent at 2.3 kHz. Synchronization (sync) pulses, which keep the receiving apparatus in step with the transmitter, are sent at 1.2 kHz. A vertical sync pulse tells the receiver that it’s time to begin a new frame; this pulse lasts for 30 milliseconds (ms). A horizontal sync pulse tells the receiver when it’s time to start a new line in a frame; its duration equals 5 ms. These pulses prevent rolling (haphazard vertical image motion) or tearing (lack of horizontal synchronization).
Fast-Scan Television
Old-fashioned analog television is also known as fast-scan TV (FSTV). While broadcasters no longer use this mode, some radio amateurs still do. The frames are transmitted at the rate of 30 per second. There are 525 lines per frame. The quick frame time, and the increased resolution, of FSTV make it necessary to use a much wider frequency band than is the case with fax or SSTV. A typical video FSTV signal takes up 6 MHz of spectrum space, or 2000 times the bandwidth of a fax or SSTV signal. Fast-scan TV is usually sent using AM or wideband FM. With AM, one of the sidebands can be filtered out, leaving only the carrier and the other sideband. Engineers call this mode vestigial sideband (VSB) transmission. It cuts the bandwidth of an FSTV signal down to about 3 MHz.
Because of the large amount of spectrum space needed to send FSTV, this mode isn’t practical for use at frequencies below about 30 MHz. Even at 30 MHz, a vestigial sideband signal consumes 10 percent of the entire RF spectrum below its own frequency! In the “olden days of TV,” all commercial FSTV transmission was carried out above 50 MHz, with the great majority of channels having frequencies far higher than this. Channels 2 through 13 on your TV receiver are sometimes called the very high frequency (VHF) channels; the higher channels are called the ultra high frequency (UHF) channels.
Figure 27-9 portrays a time-domain graph of the waveform of a single line in an FSTV video signal. This graph shows us 1/525 of a complete frame. The highest instantaneous signal amplitude corresponds to the blackest shade, and the lowest amplitude corresponds to the lightest shade. Therefore, the FSTV signal is sent “negatively.” This convention allows retracing (moving from the end of one line to the beginning of the next) to be synchronized between the transmitter and the receiver. A well-defined, strong blanking pulse tells the receiver when to retrace, and it also shuts off the beam while the receiver display is retracing. If you watched much commercial analog TV using rooftop antennas or “rabbit ears,” did you notice that weak TV signals had poor contrast? Weakened blanking pulses result in incomplete retrace blanking. But this little problem is better than having the TV receiver completely lose track of when it should retrace, as might happen if the highest instantaneous signal amplitude went with the lightest image shade.
27-9 Time-domain graph of a single line in a conventional fast-scan television (FSTV) video frame.
Color FSTV works by sending three separate monochromatic signals, corresponding to the primary colors red, blue, and green. The signals are black-and-red, black-and-blue, and black-and-green. The receiver recombines these signals and displays the resulting video as a fine, interwoven matrix of red, blue, and green dots. When viewed from a distance, the dots are too small to be individually discernible. Various combinations of red, blue, and green intensities can yield any color that the human eye can perceive.
High-Definition Television
The term high-definition television (HDTV) refers to any of several methods for getting more detail into a TV picture than could ever be done with FSTV. The HDTV mode also offers superior sound quality, making for a more satisfying home TV and home theater experience.
A standard FSTV picture has 525 lines per frame, but HDTV systems have between 787 and 1125 lines per frame. The image is scanned about 60 times per second. High-definition TV is usually sent in a digital mode; this offers another advantage over conventional FSTV. Digital signals propagate better, are easier to deal with when they are weak, and can be processed in ways that analog signals defy.
Some HDTV systems take advantage of a technique called interlacing that “meshes” two rasters (complete image frames) together. Interlacing effectively doubles the image resolution without doubling the cost of the hardware. However, interlaced images can exhibit annoying jitter in fast-moving or fast-changing scenes.
Digital Satellite TV
Until the early 1990s, a satellite television installation required a dish antenna roughly six to 10 feet (two or three meters) in diameter. A few such systems are still in use. The antennas are expensive, they attract attention (sometimes unwanted), and they’re subject to damage from ice storms, heavy snows, and high winds. Digitization has changed this situation. In any communications system, digital modes allow the use of smaller receiving antennas, smaller transmitting antennas, and/or lower transmitter power levels. Engineers have managed to get the diameter of the receiving dish down to about two feet or ⅔ of a meter.
The Radio Corporation of America (RCA) pioneered digital satellite TV with its so-called Digital Satellite System (DSS). The analog signal was changed into digital pulses at the transmitting station using A/D conversion. The digital signal was amplified and sent up to a satellite. The satellite had a transponder that received the signal, converted it to a different frequency, and retransmitted it back to the earth. A portable dish picked up the downcoming (or downlink) signal. A tuner selected the channel. The digital signal was changed back into analog form, suitable for viewing on a conventional FSTV set, by means of digital-to-analog (D/A) conversion. Although digital satellite TV technology has evolved somewhat since the initial days of the RCA DSS, today’s systems work in essentially the same way as the original ones did.
The Electromagnetic Field
In a radio or television transmitting antenna, electrons constantly move back and forth. Their velocity constantly changes as they speed up in one direction, slow down, reverse direction, speed up again, and so on. Any change of velocity (speed and/or direction) constitutes acceleration. When charged particles accelerate in a certain way, they produce an electromagnetic (EM) field.
How It Happens
When electrons move, they generate a magnetic (M) field. When electrons accelerate, they generate a changing M field. When electrons accelerate back and forth, they generate an alternating M field at the same frequency as that of the electron motion.
An alternating M field gives rise to an alternating electric (E) field, which, in turn, spawns another alternating M field. This process repeats indefinitely in the form of an EM field that propagates (travels) through space at the speed of light. The E and M fields expand alternately outward from the source in spherical wavefronts. At any given point in space, the lines of E flux run perpendicular to the lines of M flux. The waves propagate in a direction perpendicular to both the E and M flux lines.
Frequency versus Wavelength
All EM fields have two important properties: the frequency and the wavelength. When we quantify them, we find that they exhibit an inverse relation: as one increases, the other decreases. We’ve already learned about AC frequency. We can express EM wavelength as the physical distance between any two adjacent points at which either the E field or the M field has identical amplitude and direction.
An EM field can have any conceivable frequency, ranging from centuries per cycle to quadrillions of cycles per second (or hertz). The sun has a magnetic field that oscillates with a 22-year cycle. Radio waves oscillate at thousands, millions, or billions of hertz. Infrared (IR), visible light, ultraviolet (UV), X rays, and gamma rays comprise EM fields that alternate at many trillions (million millions) of hertz. The wavelength of an EM field can likewise vary over the widest imaginable range, from many trillions of miles to a tiny fraction of a millimeter.
Let fMHz represent the frequency of an EM wave in megahertz as it travels through free space. (Technically, free space constitutes a vacuum, but we can consider the air at the earth’s surface equivalent to free space for most “real-world” applications.) Let Lft represent the wavelength of the same wave in feet. Then
Lft = 984/fMHz
If we want to express the wavelength Lm in meters, then
Lm = 300/fMHz
The inverses of these formulas are
fMHz = 984/Lft
and
fMHz = 300/Lm
Velocity Factor
In media other than free space, EM fields propagate at less than the speed of light. As a result, the wavelength grows shorter according to a quantity called the velocity factor, symbolized v. The value of v can range from 0 (representing no movement at all) to 1 (representing the speed of propagation in free space, which equals approximately 186,000 mi/s or 300,000 km/s). We can also express the velocity factor as a percentage v%. In that case, the smallest possible value is 0%, and the largest is 100%. The velocity factor in practical situations rarely falls below 0.50 or 50%, and it usually exceeds 0.60 or 60%.
Velocity factor constitutes a crucial parameter in the design of RF transmission lines and antenna systems, when sections of cable, wire, or metal tubing must be cut to specific lengths measured in wavelengths or fractions of a wavelength. Taking the velocity factor v, expressed as a ratio, into account, we can modify the above-mentioned four formulas as follows:
Lft = 984v/fMHz
Lm = 300v/fMHz
fMHz = 984v/Lft
fMHz = 300v/Lm
The EM and RF Spectra
Physicists, astronomers, and engineers refer to the entire range of EM wavelengths as the electromagnetic (EM) spectrum. Scientists use logarithmic scales to depict the EM spectrum according to the wavelength in meters, as shown in Fig. 27-10A. The radio-frequency (RF) spectrum, which includes radio, television, and microwaves, appears expanded in Fig. 27-10B, where we label the axis according to frequency. The RF spectrum is categorized in bands from very low frequency (VLF) through extremely high frequency (EHF), according to the breakdown in Table 27-1. The exact lower limit of the VLF range is a matter of disagreement in the literature. Here, we define it as 3 kHz.
27-10 At A, the electromagnetic (EM) spectrum from 108 m to 10−12 m. Each vertical division represents two orders of magnitude (a 100-fold increase or decrease in the wavelength). At B, the radio-frequency (RF) portion of the EM spectrum, with each vertical division representing one order of magnitude (a 10-fold increase or decrease in the wavelength).
Table 27-1. Bands in the RF Spectrum.
Wave Propagation
Radio-wave propagation has fascinated scientists ever since Marconi and Tesla discovered, around the year 1900, that EM fields can travel over long distances without any supporting infrastructure. Let’s examine some wave-propagation behaviors that affect wireless communications at radio frequencies.
Polarization
We can define the orientation of E-field lines of flux as the polarization of an EM wave. If the E-field flux lines run parallel to the earth’s surface, we have horizontal polarization. If the E-field flux lines run perpendicular to the surface, we have vertical polarization. Polarization can also have a “slant,” of course.
In some situations, the E-field flux lines rotate as the wave travels through space. In that case we have circular polarization if the E-field intensity remains constant. If the E-field intensity is more intense in some planes than in others, we have elliptical polarization. A circularly or elliptically polarized wave can rotate either clockwise or counterclockwise as we watch the wavefronts come toward us. The rotational direction is called the sense of polarization. Some engineers use the term right-hand instead of clockwise and the term left-hand instead of counterclockwise.
Line-of-Sight Wave
Electromagnetic waves follow straight lines unless something makes them bend. Line-of-sight propagation can often take place when the receiving antenna can’t be seen visually from the transmitting antenna because radio waves penetrate nonconducting opaque objects, such as trees and frame houses, to some extent. The line-of-sight wave consists of two components called the direct wave and the reflected wave, as follows:
1. In the direct wave, the longest wavelengths are least affected by obstructions. At very low, low and medium frequencies, direct waves can diffract around things. As the frequency rises, especially above about 3 MHz, obstructions have a greater and greater blocking effect on direct waves.
2. In the reflected wave, the EM energy reflects from the earth’s surface and from conducting objects like wires and steel beams. The reflected wave always travels farther than the direct wave. The two waves might arrive at the receiving antenna in perfect phase coincidence, but usually they don’t.
If the direct and reflected waves arrive at the receiving antenna with equal strength but 180° out of phase, we observe a dead spot. The same effect occurs if the two waves arrive inverted in phase with respect to each other (that is, in phase opposition). The dead-spot phenomenon is most noticeable at the highest frequencies. At VHF and UHF, an improvement in reception can sometimes result from moving the transmitting or receiving antenna only a few inches or centimeters! In mobile operation, when the transmitter and/or receiver are moving, multiple dead spots produce rapid, repeated interruptions in the received signal, a phenomenon called picket fencing.
Surface Wave
At frequencies below about 10 MHz, the earth’s surface conducts AC quite well, so vertically polarized EM waves can follow the surface for hundreds or thousands of miles, with the earth helping to transmit the signals. As we reduce the frequency and increase the wavelength, we observe decreasing ground loss, and the waves can travel progressively greater distances by means of surface-wave propagation. Horizontally polarized waves don’t travel well in this mode because the conductive surface of the earth “shorts out” horizontal E flux. At frequencies above about 10 MHz (corresponding to wavelengths shorter than roughly 30 m), the earth becomes lossy, and surface-wave propagation rarely occurs for distances greater than a few miles.
Sky-Wave EM Propagation
Ionization in the upper atmosphere, caused by solar radiation, can return EM waves to the earth at certain frequencies. The so-called ionosphere has several dense zones of ionization that occur at fairly constant, predictable altitudes.
The E layer, which lies about 50 mi (roughly 80 km) above the surface, exists mainly during the day, although nighttime ionization is sometimes observed. The E layer can provide medium-range radio communication at certain frequencies.
At higher altitudes, we find the F1 layer and the F2 layer. The F1 layer, normally present only on the daylight side of the earth, forms at about 125 mi (roughly 200 km) altitude; the F2 layer exists at about 180 mi (roughly 300 km) over most, or all, of the earth, the dark side as well as the light side. Sometimes the distinction between the F1 and F2 layers is ignored, and they are spoken of together as the F layer. Communication by means of F-layer propagation can usually be accomplished between any two points on the earth at some frequencies between 5 MHz and 30 MHz.
The lowest ionized region is called the D layer. It exists at an altitude of about 30 mi (roughly 50 km), and is ordinarily present only on the daylight side of the planet. This layer absorbs radio waves at some frequencies, impeding long-distance ionospheric propagation.
Tropospheric Propagation
At frequencies above about 30 MHz (wavelengths shorter than about 10 m), the lower atmosphere bends radio waves towards the surface. Tropospheric bending occurs because the index of refraction of air, with respect to EM waves, decreases with altitude. Tropospheric bending makes it possible to communicate for hundreds of miles, even when the ionosphere will not return waves to the earth.
Ducting is tropospheric propagation that occurs somewhat less often than bending, but offers more dramatic effects. Ducting takes place when EM waves get “trapped” within a layer of cool, dense air sandwiched between two layers of warmer air. Like bending, ducting occurs almost entirely at frequencies above 30 MHz.
Still another tropospheric-propagation mode is called tropospheric scatter, or troposcatter. This phenomenon takes place because air molecules, dust grains, and water droplets scatter some of the EM field. We observe troposcatter most commonly at VHF and UHF. Troposcatter always occurs to some extent, regardless of weather conditions.
Tropospheric propagation in general, without mention of the specific mode, is sometimes called tropo.
Auroral Propagation
In the presence of unusual solar activity, the aurora (northern lights or southern lights) can return radio waves to the earth, facilitating auroral propagation. The aurora occurs at altitudes of about 40 to 250 mi (roughly 65 to 400 km). Theoretically, auroral propagation is possible, when the aurora are active, between any two points on the earth’s surface from which the same part of the aurora lie on a line of sight. Auroral propagation seldom occurs when either the transmitting station or the receiving station is located at a latitude less than 35° north or south of the equator.
Auroral propagation causes rapid and deep signal fading, which nearly always renders analog voice and video signals unintelligible. Digital modes work somewhat better, but the carrier frequency gets “spread out” or “smeared” over a band several hundred hertz wide as a result of phase modulation induced by auroral motion. This “spectral spreading” limits the maximum data transfer rate. Auroral propagation commonly takes place along with poor ionospheric propagation resulting from sudden eruptions called solar flares on the sun’s surface.
Meteor-Scatter Propagation
Meteors produce ionized trails that persist for a fraction of a second up to several seconds. The exact duration of the trail depends on the size of the meteor, its speed, and the angle at which it enters the atmosphere. A single meteor trail rarely lasts long enough to allow transmission of much data. However, during a meteor shower, multiple trails can produce almost continuous ionization for a period of hours. Ionized regions of this type can reflect radio waves at certain frequencies. Communications engineers call this effect meteor-scatter propagation, or sometimes simply meteor scatter. It can take place at frequencies far above 30 MHz and over distances ranging from just beyond the horizon up to about 1500 mi (roughly 2400 km). The maximum communications range depends on the altitude of the ionized trail, and also on the relative positions of the trail, the transmitting station, and the receiving station.
Moonbounce Propagation
Earth-moon-earth (EME) communications, also called moonbounce, is routinely carried on by amateur radio operators at VHF and UHF. This mode requires a sensitive receiver using a low-noise preamplifier, a large, directional antenna, and a high-power transmitter. Digital modes work far better than analog modes for moonbounce.
Signal path loss presents the main difficulty for anyone who contemplates EME communications. Received EME signals are always weak. High-gain directional antennas must remain constantly aimed at the moon, a requirement that dictates the use of steerable antenna arrays. The EME path loss increases with increasing frequency, but this effect is offset by the more manageable size of high-gain antennas as the wavelength decreases.
Solar noise can pose a problem; EME communications becomes most difficult near the time of the new moon, when the moon lies near a line between the earth and the sun. The sun constitutes a massive broadband generator of EM energy! Problems can also occur with cosmic noise when the moon passes near “noisy” regions in the so-called radio sky. The constellation Sagittarius lies in the direction of the center of the Milky Way galaxy, and EME performance suffers when the moon passes in front of that part of the stellar background.
The moon keeps the same face more or less toward the earth at all times, but some back-and-forth “wobbling” occurs. This motion, called libration (not “liberation” or “libation”!), produces rapid, deep fluctuations in signal strength, a phenomenon known as libration fading. The fading becomes more pronounced as the operating frequency increases. It occurs as multiple transmitted EM wavefronts reflect from various “lunagraphical” features, such as craters and mountains on the moon’s surface, whose relative distances constantly change because of libration. The reflected waves recombine in constantly shifting phase at the receiving antenna, sometimes reinforcing, and at other times canceling.
Transmission Media
Data can be transmitted over various media, which include cable, radio (also called wireless), satellite links (a specialized form of wireless), and fiberoptics. Cable, radio/TV, and satellite communications use the RF spectrum. Fiberoptics uses IR or visible light energy.
Cable
The earliest cables comprised plain wires that carried DC. Nowadays, data-transmission cables more often carry RF signals that can be amplified at intervals on a long span. The use of such amplifiers, called repeaters, greatly increases the distances over which data can be sent by cable. Another advantage of using RF is the fact that numerous signals can travel over a single cable, with each signal on a different frequency.
Cables can consist of pairs of wires, somewhat akin to lamp cords. But more often coaxial cable, of the type described and illustrated at the end of Chap. 10, is used. This type of cable has a center conductor that carries the signals, surrounded by a cylindrical, grounded shield that keeps signals confined to the cable, and also keeps external EM fields from interfering with the signals.
Radio
All radio and TV signals consist of EM waves traveling through the earth’s atmosphere or outer space. In a radio transmitting station, the RF output goes into an antenna system located at some distance from the transmitter. To get from the transmitter’s final amplifier to the antenna, the EM energy follows a transmission line, also called a feed line.
Most radio antenna transmission lines consist of coaxial cable. Other types of cable exist for special applications. At microwave frequencies, hollow tubes called waveguides can transfer the energy. A waveguide works more efficiently than coaxial cable at the shortest radio wavelengths.
Radio amateurs sometimes use a parallel-wire transmission line in which the RF currents in the two conductors are in phase opposition so their EM fields cancel each other out. This phase cancellation keeps the transmission line from radiating, guiding the EM field along toward the antenna.
Satellite Systems
At very high frequencies (VHF) and above, some communications circuits use satellites that follow geostationary orbits around the earth. If a satellite orbits directly over the equator at an altitude of approximately 35,800 km (22,200 mi) and travels from west to east, it follows the earth’s rotation, thereby staying in the same spot in the sky as seen from the surface. That’s why we call it a geostationary satellite.
A single geostationary satellite lies on a line of sight with a large set of locations that covers about 40% of the earth’s surface. Three such satellites, placed at 120° (⅓-circle) intervals around the earth, allow coverage of all human-developed regions. Only the extreme polar regions lie “out of range.” We can aim a dish antenna at a geostationary satellite, and once we’ve fixed the antenna in the correct position, we can leave it alone.
Another form of satellite system uses multiple “birds” in relatively low-altitude orbits that take them over, or nearly over, the earth’s poles. These satellites exhibit continuous, rapid motion with respect to the earth’s surface. If enough satellites of this type exist, the entire “flock” can work together, maintaining reliable communications between any two points on the surface at all times. Directional antennas aren’t necessary in these systems, which engineers call low earth orbit (LEO) networks.
Fiberoptics
We can modulate beams of IR or visible light, just as we can modulate RF carriers. An IR or visible light beam has a frequency far higher than that of any RF signal, allowing modulation by data at rates faster than anything possible with radio.
Fiberoptic technology offers several advantages over wire cables (which are sometimes called copper because the conductors usually comprise that metallic element). A fiberoptic cable doesn’t cost much, doesn’t weigh much, and remains immune to interference from outside EM fields. A fiberoptic cable doesn’t corrode as metallic wires do. Fiberoptic cables are inexpensive to maintain and easy to repair. An optical fiber can carry far more signals than a cable because the frequency bands are far wider in terms of megahertz or gigahertz.
In theory we can “imprint” the entire RF spectrum, from VLF through EHF, onto a single beam of visible light and transmit it through an optical fiber no thicker than a strand of human hair!
Receiver Fundamentals
A wireless receiver converts EM waves into the original messages sent by a distant transmitter. Let’s define a few important criteria for receiver operation, and then we’ll look at two common receiver designs.
Specifications
The specifications of a receiver quantify how well the hardware can actually do what we design and build it to do.
Sensitivity: The most common way to express receiver sensitivity is to state the number of micro-volts that must exist at the antenna terminals to produce a certain signal-to-noise ratio (S/N) or signal-plus-noise-to-noise ratio (S+N/N) in decibels (dB). The sensitivity depends on the gain of the front end (the amplifier or amplifiers connected to the antenna). The amount of noise that the front end generates also matters because subsequent stages amplify its noise output as well as its signal output.
Selectivity: The passband, or bandwidth that the receiver can “hear,” is established by a wideband preselector in the early RF amplification stages, and is honed to precision by narrowband filters in later amplifier stages. The preselector makes the receiver most sensitive within about plus-or-minus 10 percent (±10%) of the desired signal frequency. The narrowband filter responds only to the frequency or channel of a specific signal that we want to hear; the filter rejects signals in nearby channels.
Dynamic range: The signals at a receiver input can vary over several orders of magnitude (powers of 10) in terms of absolute voltage. We define dynamic range as the ability of a receiver to maintain a fairly constant output, and yet to keep its rated sensitivity, in the presence of signals ranging from extremely weak to extremely strong. A good receiver exhibits dynamic range in excess of 100 dB. Engineers can conduct experiments to determine the dynamic range of any receiver; commercial receiver manufacturers publish this specification as a selling point.
Noise figure: The less internal noise a receiver produces, in general, the better the S/N ratio will be. We can expect an excellent S/N ratio in the presence of weak signals only when our receiver has a low noise figure, a measure of internally generated circuit noise. The noise figure matters most at VHF, UHF, and microwave frequencies. Gallium-arsenide field-effect transistors (GaAsFETs) are known for the low levels of noise they generate, even at very high frequencies. We can get away with other types of FETs at lower frequencies. Bipolar transistors, which carry higher currents than FETs, generate more circuit noise than FETs do.
Direct-Conversion Receiver
A direct-conversion receiver derives its output by mixing incoming signals with the output of a tunable (variable frequency) local oscillator (LO). The received signal goes into a mixer along with the output of the LO. Figure 27-11 is a block diagram of a direct-conversion receiver.
27-11 Block diagram of a direct-conversion receiver.
For reception of on/off keyed Morse code, also called radiotelegraphy or continuous-wave (CW) mode, the LO, also called a beat-frequency oscillator (BFO), is set a few hundred hertz above or below the signal frequency. We can also use this scheme to receive FSK signals. The audio output has a frequency equal to the difference between the LO frequency and the incoming carrier frequency. For reception of AM or SSB signals, we adjust the LO to precisely the same frequency as that of the signal carrier, a condition called zero beat because the beat frequency, or difference frequency, between the LO and the signal carrier equals zero.
A direct-conversion receiver provides rather poor selectivity, meaning that it can’t always separate incoming signals when they lie close together in frequency. In a direct-conversion receiver, we can hear signals on either side of the LO frequency at the same time. A selective filter can theoretically eliminate this problem. Such a filter must be designed for a fixed frequency if we expect it to work well. However, in a direct-conversion receiver, the RF amplifier must operate over a wide range of frequencies, making effective filter design an extreme challenge.
Superheterodyne Receiver
A superheterodyne receiver, also called a superhet, uses one or more local oscillators and mixers to obtain a constant-frequency signal. We can more easily filter a fixed-frequency signal than we can filter a signal that changes in frequency (as it does in a direct-conversion receiver).
In a superhet, the incoming signal goes from the antenna through a tunable, sensitive front end, which is a precision weak-signal amplifier. The output of the front end mixes (heterodynes) with the signal from a tunable, unmodulated LO. We can choose the sum signal or the difference signal for subsequent amplification. We call this signal the first intermediate frequency (IF), which can be filtered to obtain selectivity.
If the first IF signal passes straight into the detector, we call our system a single-conversion receiver. Some receivers use a second mixer and second LO, converting the first IF to a lower-frequency second IF. Then we have a double-conversion receiver. The IF bandpass filter can be constructed for use on a fixed frequency, allowing superior selectivity and facilitating adjustable bandwidth. The sensitivity is enhanced because fixed IF amplifiers are easy to keep in tune.
Unfortunately, even the best superheterodyne receiver can intercept or generate unwanted signals. We call external false signals images; we call internally generated false signals birdies. If we carefully choose the LO frequency (or frequencies) when we design our system, images and birdies will rarely cause problems during ordinary operation.
Stages of a Single-Conversion Superhet
Figure 27-12 shows a block diagram of a generic single-conversion superheterodyne receiver. Individual receiver designs vary somewhat, but we can consider this example representative. The various stages break down as follows:
27-12 Block diagram of a single-conversion superheterodyne receiver.
• The front end consists of the first RF amplifier, and often includes LC bandpass filters between the amplifier and the antenna. The dynamic range and sensitivity of a receiver are determined by the performance of the front end.
• The mixer stage, in conjunction with the tunable local oscillator (LO), converts the variable signal frequency to a constant IF. The output occurs at either the sum or the difference of the signal frequency and the tunable LO frequency.
• The IF stages produce most of the gain. We also get most of the selectivity here, filtering out unwanted signals and noise, while allowing the desired signal to pass.
• The detector extracts the information from the signal. Common circuits include the envelope detector for AM, the product detector for SSB, FSK, and CW, and the ratio detector for FM.
• One or two stages of audio amplification boost the demodulated signal to a level suitable for a speaker or headset. Alternatively, we can feed the signal to a printer, facsimile machine, or computer.
Predetector Stages
When we design and build a superheterodyne receiver, we must ensure that the stages preceding the first mixer provide reasonable gain but generate minimal internal noise. They must also be capable of handling strong signals without desensitization (losing gain), a phenomenon also known as overloading.
Preamplifier
All preamplifiers operate in the class-A mode, and most employ FETs. An FET has a high input impedance ideally suited to weak-signal work. Figure 27-13 shows a simple generic RF preamplifier circuit. Input tuning reduces noise and provides some selectivity. This circuit produces a 5 dB to 10 dB gain, depending on the frequency and the choice of FET.
27-13 A tunable preamplifier for use with a radio receiver. This circuit uses an N-channel JFET.
We must ensure that a preamplifier remains linear in the presence of strong input signals. Nonlinearity can cause unwanted mixing among multiple incoming signals. These so-called mixing products produce intermodulation distortion (IMD), or intermod, that can spawn numerous false signals inside the receiver. Intermod can also degrade the S/N ratio by generating hash, a form of wideband noise.
Front End
At low and medium frequencies, considerable atmospheric noise exists, and the design of a front-end circuit is simple because we don’t have to worry much about internally generated noise. (Conditions are bad enough in the antenna!)
Atmospheric noise diminishes as we get above 30 MHz or so. Then the main sensitivity-limiting factor becomes noise generated within the receiver. For this reason, front-end design grows in importance as the frequency rises through the VHF, UHF, and microwave spectra.
The front end, like a preamplifier, must remain as linear as possible. The greater the degree of nonlinearity, the more susceptible the circuit becomes to the generation of mixing products and intermod. The front end should also have the greatest possible dynamic range.
Preselector
The preselector provides a bandpass response that improves the S/N ratio, and reduces the likelihood of overloading by a strong signal that’s far removed from the operating frequency. The preselector also provides image rejection in a superheterodyne circuit.
We can tune a preselector by means of tracking with the receiver’s main tuning control, but this technique requires careful design and alignment. Some older receivers incorporate preselectors that must be adjusted independently of the receiver tuning.
IF Chains
A high IF (several megahertz) works better than a low IF (less than 1 MHz) for image rejection. However, a low IF allows us to obtain superior selectivity. Double-conversion receivers have a comparatively high first IF and a low second IF to get the “best of both worlds.” We can cascade multiple IF amplifiers with tuned-transformer coupling. The amplifiers follow the mixer and precede the detector. Double-conversion receivers have two series, called chains, of IF amplifiers. The first IF chain follows the first mixer and precedes the second mixer, and the second IF chain follows the second mixer and precedes the detector.
Engineers sometimes express IF-chain selectivity by comparing the bandwidths for two power-attenuation values, usually –3 dB and –30 dB, also called 3 dB down and 30 dB down. This specification offers a good description of the bandpass response. We call the ratio of the bandwidth at –30 dB to the bandwidth at –3 dB the shape factor. In general, small shape factors are more desirable than large ones, but small factors can prove difficult to attain in practice. When we have a small shape factor and graph the system gain as a function of frequency, we get a curve that resembles a rectangle, so we can say that the receiver has a rectangular response.
Detectors
Detection, also called demodulation, allows a wireless receiver to recover the modulating information, such as audio, images, or printed data, from an incoming signal.
Detection of AM
A radio receiver can extract the information from an AM signal by half-wave rectifying the carrier wave and then filtering the output waveform just enough to smooth out the RF pulsations. Figure 27-14A shows a simplified time-domain view of how this process works. The rapid pulsations (solid curves) occur at the RF carrier frequency; the slower fluctuation (dashed curve) portrays the modulating data. The carrier pulsations are smoothed out by passing the output through a capacitor that’s large enough to hold the charge for one carrier current cycle, but not so large that it dampens or obliterates the fluctuations in the modulating signal. We call this technique envelope detection.
27-14 At A, envelope detection of AM, shown in the time domain. At B, slope detection of FM, shown in the frequency domain.
Detection of CW and FSK
If we want a receiver to detect CW, we must inject a constant-frequency, unmodulated carrier a few hundred hertz above or below the signal frequency. The local carrier is produced by a tunable beat-frequency oscillator (BFO). The BFO signal and the incoming CW signal heterodyne in a mixer to produce audio output at the sum or difference frequency. We can tune the BFO to obtain an audio “note” or “tone” at a comfortable listening pitch, usually 500 to 1000 Hz. This process is called heterodyne detection.
We can detect FSK signals using the same method as CW detection. The carrier beats against the BFO in the mixer, producing an audio tone that alternates between two different pitches. With FSK, the BFO frequency is set a few hundred hertz above or below both the mark frequency and the space frequency. The frequency offset, or difference between the BFO and the signal frequencies, determines the audio output frequencies. We adjust the frequency offset to get specific standard AF notes (such as 2125 Hz and 2295 Hz in the case of 170-Hz shift).
Slope Detection of FM and PM
We can use an AM receiver to detect FM or PM by setting the receiver frequency near, but not exactly at, the unmodulated-carrier frequency. An AM receiver has a filter with a passband of a few kilohertz and a selectivity curve such as that shown in Fig. 27-14B. If we tune the receiver so that the FM unmodulated-carrier frequency lies near either edge, or skirt, of the filter response, frequency variations in the incoming signal cause its carrier to “swing” in and out of the receiver passband. As a result, the instantaneous receiver output amplitude varies along with the modulating data on the FM or PM signal. In this system, known as slope detection, the relationship between the instantaneous deviation and the instantaneous output amplitude is nonlinear because the skirt of the passband is not a straight line (as we can see in Fig. 27-14B). Therefore, slope detection does not provide an optimum method of detecting FM or PM signals. The process can usually yield an intelligible voice, but it will ruin the quality of music.
Using a PLL to Detect FM or PM
If we inject an FM or PM signal into a PLL circuit, the loop produces an error voltage that constitutes a precise duplicate of the modulating waveform. A limiter, which keeps the signal amplitude from varying, can be placed ahead of the PLL so that the receiver doesn’t respond to changes in the signal amplitude. Weak signals tend to abruptly appear and disappear, rather than fading, in an FM or PM receiver that employs limiting.
Discriminator for FM or PM
A discriminator produces an output voltage that depends on the instantaneous signal frequency. When the signal frequency lies at the center of the receiver passband, the output voltage equals zero. When the instantaneous signal frequency falls below the passband center, the output voltage becomes positive. When the instantaneous signal frequency rises above center, the output voltage becomes negative. The relationship between the instantaneous FM deviation (which, as we remember, can result indirectly from PM) and the instantaneous output amplitude is linear. Therefore, the detector output represents a faithful reproduction of the incoming signal data. A discriminator is sensitive to amplitude variations, but we can use a limiter to get rid of this problem, just as we do in a PLL detector.
Ratio Detector for FM or PM
A ratio detector comprises a discriminator with a built-in limiter. The original design was developed by RCA (Radio Corporation of America), and works well in high-fidelity receivers and in the audio portions of old-fashioned analog TV receivers. Figure 27-14C illustrates a simple ratio detector circuit. The potentiometer marked “balance” should be adjusted experimentally to get optimum received-signal audio quality.
27-14 At C, a ratio detector circuit for demodulating FM signals.
Detection of SSB
For reception of SSB signals, most communications engineers prefer to use a product detector, although a direct-conversion receiver can do the job. A product detector also facilitates reception of CW and FSK. The incoming signal combines with the output of an unmodulated LO, reproducing the original modulating signal data. Product detection occurs at a single frequency, rather than at a variable frequency, as in direct-conversion reception. The single, constant frequency results from mixing of the incoming signal with the output of the LO.
Figures 27-14D and 27-14E are schematic diagrams of product-detector circuits, which can also serve as mixers in superhet receivers. In the circuit shown at D, diodes are used, so we do not get any amplification. The circuit shown at E employs a bipolar transistor biased for class-B mode, providing some gain if the incoming signal has been sufficiently amplified by the front end before it arrives at the detector input. The effectiveness of the circuits shown in Figs. 27-14D or 27-14E lies in the nonlinearity of the semiconductor devices. This nonlinearity facilitates the heterodyning necessary to obtain sum and difference frequency signals that result in data output.
27-14 At D, a product detector using diodes. At E, a product detector using an NPN bipolar transistor biased for class-B operation.
Postdetector Stages
We can obtain selectivity in a receiver by tailoring the frequency response in the AF amplifier stages following the detector, in addition to optimizing the RF selectivity in the IF stages preceding the detector.
Filtering
In a communications system, a human voice signal requires a band ranging from about 300 Hz to 3000 Hz for a listener to easily understand the content. An audio bandpass filter, with a passband of 300 Hz to 3000 Hz, can improve the intelligibility in some voice receivers. An ideal voice audio bandpass filter has little or no attenuation within the passband range but high attenuation outside the passband range, along with a near-rectangular response curve.
A CW or FSK signal requires only a few hundred hertz of bandwidth. Audio CW filters can narrow the response bandwidth to 100 Hz or less, but passbands narrower than about 100 Hz produce ringing, degrading the quality of reception at high data speeds. With FSK, the bandwidth of the filter must be at least as large as the difference (shift) between mark and space, but it need not (and shouldn’t) greatly exceed the frequency shift.
An audio notch filter is a band-rejection filter with a sharp, narrow response. Band-rejection filters pass signals only below a certain lower cutoff frequency or above a certain upper cutoff frequency. Between those limits, in the so-called bandstop range, signals are blocked. A notch filter can “mute” an interfering unmodulated carrier or CW signal that produces a constant-frequency tone in the receiver output. Audio notch filters are tunable from at least 300 Hz to 3000 Hz. Some AF notch filters work automatically; when an interfering AF tone appears, the notch finds and “mutes” it within a few tenths of a second.
Squelching
A squelch silences a receiver when no incoming signals exist, allowing reception of signals when they appear. Most FM communications receivers use squelching systems. The squelch is normally closed, cutting off all audio output (especially receiver hiss, which annoys some communications operators) when no signal is present. The squelch opens, allowing everything to be heard, if the signal amplitude exceeds a squelch threshold that the operator can adjust.
In some systems, the squelch does not open unless an incoming signal has certain pre-determined characteristics. This feature is called selective squelching. The most common way to achieve selective squelching is the use of a subaudible (below 300 Hz) tone generator or an AF tone-burst generator in the transmitter. The squelch opens only in the presence of signals modulated by a tone, or sequence of tones, having the proper characteristics. Some radio operators use selective squelching to prevent unwanted transmissions from “coming in.”
Fast-Scan TV
An analog fast-scan television (FSTV) receiver can follow a D/A converter for reception of advanced digital TV signals. This type of receiver has a tunable front end, an oscillator and mixer, a set of IF amplifiers, a video demodulator, an audio demodulator and amplifier chain, a picture CRT or display with associated peripheral circuitry, and a loudspeaker. Figure 27-15 is a block diagram of a receiver for FSTV.
27-15 Block diagram of a conventional FSTV receiver.
In the United States, conventional FSTV broadcasts were once made on channels numbered from 2 through 69. Each channel was 6 MHz wide, including video and audio information. Channels 2 through 13 were called the VHF TV broadcast channels. Channels 14 through 69 constituted the UHF TV broadcast channels. In digital television these days, the D/A “converter box” serves as the program selector and outputs all signals on a single analog channel.
Slow-Scan TV
A slow-scan television (SSTV) communications station needs a transceiver with SSB capability, a standard FSTV set or personal computer, and a scan converter that translates between the SSTV signal and either FSTV imagery or computer video data. The scan converter contains two data converters (one for receiving and the other for transmitting), some digital memory, a tone generator, and a TV detector. Scan converters are commercially available. Computers can be programmed to perform this function. Some amateur radio operators build their own scan converters.
Specialized Wireless Modes
Communications engineers have a long history of innovation, developing numerous exotic wireless modes. In recent years, new modes have emerged; we can expect more to come, each of which offers specific advantages under strange or difficult conditions. Four common examples follow.
Dual-Diversity Reception
A dual-diversity receiver can reduce fading in radio reception at high frequencies (approximately 3 to 30 MHz) when signals propagate through the ionosphere and return to earth’s surface. The system comprises two identical receivers tuned to the same signal and having separate antennas spaced several wavelengths apart. The outputs of the receiver detectors go into a single audio amplifier as shown in Fig. 27-16.
27-16 Block diagram of a dual-diversity radio receiving system.
Dual-diversity receiver tuning is a sophisticated technology—we might call it an art—and good equipment for this purpose costs a lot of money. Some advanced diversity-reception installations employ three or more antennas and receivers, providing superior immunity to fading, but further compounding the tuning difficulty and further increasing the expense.
Synchronized Communications
Digital signals require less bandwidth than analog signals to convey a given amount of information per unit of time. The term synchronized communications refers to any of several specialized digital modes in which the transmitter and receiver operate from a common frequency-and-time standard to optimize the amount of data that can be sent in a communications channel or band.
In synchronized digital communications, also called coherent communications, the receiver and transmitter operate in lock-step. The receiver evaluates each transmitted data bit for a block of time lasting for the specified duration of a single bit. This process makes it possible to use a receiving filter having extremely narrow bandwidth. The synchronization requires the use of an external frequency-and-time standard, such as that provided by the National Institute of Standards and Technology (NIST) radio station WWV in the United States. Frequency dividers generate the necessary synchronizing signals from the frequency-standard signal. A tone or pulse appears in the receiver output for a particular bit if, but only if, the average signal voltage exceeds a certain value over the duration of that bit. False signals caused by filter ringing, sferics, or ignition noise are generally ignored because they rarely produce sufficient average bit voltage.
Experiments with synchronized communications have shown that the improvement in S/N ratio, compared with nonsynchronized systems, is several decibels at low to moderate data speeds.
Multiplexing
Signals in a communications channel or band can be intertwined, or multiplexed, in various ways. The most common methods are frequency-division multiplexing (FDM) and time-division multiplexing (TDM). In FDM, the channel is broken down into subchannels. The carrier frequencies of the signals are spaced so that they don’t overlap. Each signal remains independent of all the others. A TDM system breaks signals down into segments of specific time duration, and then the segments are transferred in a rotating sequence. The receiver stays synchronized with the transmitter by means of an external time standard, such as the data from “shortwave” station WWV. Multiplexing requires an encoder that combines or “intertwines” the signals in the transmitter, and a decoder that separates or “untangles” the signals in the receiver.
Spread-Spectrum
In spread-spectrum communications, the transmitter varies the main carrier frequency in a controlled manner, independently of the signal modulation. The receiver is programmed to follow the transmitter frequency from instant to instant. The whole signal, therefore, “roams” up and down in frequency within a defined range.
In spread-spectrum mode, the probability of catastrophic interference, in which one strong interfering signal can obliterate the desired signal, is near zero. Unauthorized people find it impossible to eavesdrop on a spread-spectrum communications link unless they gain access to the sequencing code, also known as the frequency-spreading function. Such a function can be complex, and, of course, it must be kept secret. If neither the transmitting operator nor the receiving operator divulge the sequencing code to anyone, then (ideally) no unauthorized listener will be able to intercept it.
During a spread-spectrum contact between a given transmitter and receiver, the operating frequency can fluctuate over a range of several kilohertz, megahertz, or tens of megahertz. As a band becomes occupied with an increasing number of spread-spectrum signals, the overall noise level in the band appears to increase. Therefore, a practical limit exists to the number of spread-spectrum contacts that a band can handle. This limit is roughly the same as it would be if all the signals were constant in frequency, and had their own discrete channels. The main difference between fixed-frequency communications and spread-spectrum communications, when the band gets crowded, lies in the nature of the mutual interference.
A common method of generating spread-spectrum signals involves so-called frequency hopping. The transmitter has a list of channels that it follows in a certain order. The transmitter “jumps” or “hops” from one frequency to another in the list. The receiver must be programmed with this same list, in the same order, and must be synchronized with the transmitter. The dwell time equals the length of time that the signal remains on any given frequency; it’s the same as the time interval at which frequency changes occur. In a well-designed frequency-hopping system, the dwell time is short enough so that a signal will not be noticed by an unauthorized listener using a receiver set to a constant frequency, and also will not cause interference on any frequency. The sequence contains numerous dwell frequencies, so the signal energy is diluted to the extent that, if someone tunes to any particular frequency in the sequence, they won’t notice the signal.
Another way to obtain spread spectrum, called frequency sweeping, requires frequency-modulating the main transmitted carrier with a waveform that guides it “smoothly” up and down over the assigned band. The “sweeping FM” remains entirely independent of the actual data that the signal conveys. A receiver can intercept the signal if, but only if, its instantaneous frequency varies according to the same waveform, over the same band, at the same rate, and in the same phase as that of the transmitter. The transmitter and receiver in effect “roam all over the band,” following each other from moment to moment according to a “secret map” that only they know.
Quiz
Refer to the text in this chapter if necessary. A good score is at least 18 correct. Answers are in the back of the book.
1. Which of the following communications modes has a mark component at one carrier frequency and a space component at a different carrier frequency?
(a) CW
(b) FSK
(c) FSTV
(d) FM
2. Which of the following communications modes has a mark component in which the carrier is “full-on” and a space component in which the carrier is entirely absent?
(a) CW
(b) FSK
(c) FSTV
(d) FM
3. We can demodulate FM signals with
(a) a discriminator.
(b) a ratio detector.
(c) an envelope detector.
(d) All of the above
4. If we want to demodulate an AM signal, we’ll get the best results with
(a) a discriminator.
(b) a ratio detector.
(c) an envelope detector.
(d) a product detector.
5. We can accomplish spread-spectrum communications by means of
(a) ratio detection.
(b) product detection.
(c) frequency hopping.
(d) All of the above
6. Birdies can occur in a
(a) direct-conversion receiver.
(b) superheterodyne receiver.
(c) ratio detector.
(d) front end with poor dynamic range.
7. The dynamic range specification in a receiver tells us how well the system can handle signals
(a) in diverse modulation modes.
(b) over a wide range of frequencies.
(c) in the presence of high noise levels.
(d) from extremely weak to extremely strong.
8. Which of the following modes involves breaking a signal down into “pieces” of specific time duration, transmitting the “pieces” in a repeating sequence, and reassembling them back into the original signal at the receiver?
(a) FSK
(b) TDM
(c) SSTV
(d) CW
9. Suppose that an RF carrier has a frequency of 830 kHz. We can effectively modulate this carrier with information containing frequency components up to about
(a) 83.0 Hz.
(b) 830 Hz.
(c) 8.30 kHz.
(d) 83.0 kHz.
10. We might expect to observe a dead spot in communications reception when the direct and reflected waves arrive at the receiving antenna
(a) with a 0° phase difference.
(b) with a 180° phase difference.
(c) with a 90° phase difference.
(d) in any condition other than phase coincidence.
11. The earth’s atmosphere generally exhibits a decreasing index of refraction, with respect to radio waves, as the altitude above the surface increases. At some radio frequencies, this property gives rise to
(a) ionospheric reflection.
(b) troposcatter.
(c) tropospheric bending.
(d) auroral propagation.
12. Suppose that we modulate a VHF carrier in the SSB mode with AF data having frequency components up to 20 kHz. What’s the approximate bandwidth of the SSB signal?
(a) 10 kHz
(b) 20 kHz
(c) 40 kHz
(d) 80 kHz
13. In a LEO satellite network, each individual satellite follows an orbit
(a) approximately 22,200 miles above the earth’s surface.
(b) that keeps it in a single spot above the earth’s surface.
(c) that keeps it constantly between the earth and the moon.
(d) that takes it over, or nearly over, the earth’s poles at low altitude.
14. We can generate a DSB, suppressed-carrier signal with a
(a) frequency modulator.
(b) phase modulator.
(c) balanced modulator.
(d) slope modulator.
15. Figure 27-17 illustrates a circuit designed to perform
27-17 Illustration for Quiz Questions 15 through 17.
(a) ratio detection.
(b) modulation.
(c) oscillation.
(d) product detection.
16. Based on the general knowledge of electronics that we’ve gained so far in this course, we know that the components marked X in Fig. 27-17
(a) pass signals but not DC.
(b) pass DC but not signals.
(c) ensure that the transistor receives pure DC from the battery.
(d) limit the current through the transistor E-B and B-C junctions.
17. What appears at the output terminals of the circuit of Fig. 27-17, assuming that we choose the correct component values and operate the system properly?
(a) An FM signal
(b) An SSB signal
(c) A DSB, suppressed-carrier signal
(d) An AM signal
18. Figure 27-18 is a schematic diagram for a weak-signal amplifier that we might find in the front end of a radio communications receiver. What’s wrong?
27-18 Illustration for Quiz Questions 18 through 20.
(a) We should use a P-channel JFET, not an N-channel JFET.
(b) We should install a blocking capacitor between the gate resistor and the center tap of the inductor.
(c) We should replace the capacitor at the output terminals with an RF choke.
(d) We should remove the capacitor between the source and ground.
19. In the circuit of Fig. 27-18, the LC circuit marked X
(a) optimizes the bias on the JFET.
(b) prevents strong signals on the desired frequency from overloading the system.
(c) provides selectivity at the receiver’s front end.
(d) minimizes the S/N ratio.
20. In the circuit of Fig. 27-18, the resistor marked Y
(a) optimizes the bias on the JFET.
(b) prevents strong signals from overloading the system.
(c) provides selectivity.
(d) minimizes the S/N ratio.