Previous Page
Next Page

Chapter 2: Radio Frequency

This chapter provides the average reader with a crash course in the basics of radio frequency (RF) terminology and hardware. A basic understanding of the fundamental concepts and theories in RF electronics is an invaluable asset to the wireless hacker. At the end of the day, if your hardware isn't working, then no amount of clever software is going to help.

RF TERMINOLOGY

To start off, let's talk about what is meant by radio frequency or RF. Technically speaking, RF refers to any signal between the frequencies of 3 Hz and 300 GHz. More practically, however, RF refers to signals from about 3 MHz up to 300 GHz. Signals that fall into this range of frequencies are capable of traveling through space in the form of electromagnetic waves. The distance over which these signals can travel depends on factors such as the signal's frequency and atmospheric conditions.

Communications Systems

The history of wireless communications goes back to the late 1800s when a German physicist by the name of Heinrich Rudolf Hertz first discovered the existence of electromagnetic waves. This discovery marked the creation of the first radio. Although Hertz's radio was very primitive, it established that a signal could be generated at one location and detected at another location without the use of wires. Technology later advanced to the point where Morse code could be transmitted using radio waves rather than relying on telegraph wires. Wireless communication then progressed to the transmission of human voice and audio using radio waves, and today high-speed data communications over wireless links are used every day.

Components of a Communications System

Modern communication systems are generally constructed using the same set of fundamental components. Although the designs have changed over the years, the basic components are very similar to those used for the early radios. Figure 2-1 shows a block diagram illustrating these basic components as implemented in both a traditional analog (voice) communications system and a typical 802.11 wireless communications system (or radio). Both systems contain largely the same components for the RF front-end and transceiver. The primary differences are in the format of the incoming and outgoing data and the components that interface the data source to the transceiver as designated by the dashed boxes surrounding these components in the figure.

Image from book
Figure 2-1: Block diagram of basic communications system

The data source is the component that is generating the information to be transmitted and received by the system. For an analog radio, this could be either a microphone in the case of a communications radio or music in the case of a broadcast radio station. For a digital communications system, the data source is any type of digital bitstream. Of particular interest for readers of this book is the bitstream originating from the bottom level of the TCP/IP stack where the raw data is passed to the physical link layer.

For digital radios, the next component of the radio is the baseband processing. This is where the digital bitstream coming from the data source is converted into an analog baseband signal through a process called modulation. (The process of modulation is not the same thing as a standard analog to digital conversion. There are various techniques for generating the modulated baseband signal that will be discussed later in the section on modulation.) Now that the data has been converted into an analog signal, the rest of the signal path becomes very similar to that used in more traditional radios. The baseband analog signal is typically the lowest frequency signal in the radio and is at too low a frequency for RF transmission, but this issue is taken care of in the transceiver.

The transceiver handles the process of converting the low-frequency baseband signal into a higher-frequency RF signal though a process called upconversion. Within the transceiver, an RF carrier signal is generated at the frequency that will be used for the final RF signal. This RF carrier is then combined with the baseband signal to upconvert the low-frequency baseband to the higher RF carrier frequency. Figure 2-2A shows both the low-frequency baseband signal and the RF carrier signal that are used in the upconversion process. Figure 2-2B shows the result of an ideal upconversion where the modulated RF signal is identical to the original baseband signal; except, it is now centered at a much higher frequency. In addition to upconverting outgoing signals, the transceiver also downconverts incoming RF signals to low-frequency baseband signals. The downconversion process is the inverse of the upconversion process.

Image from book
Figure 2-2: The process of upconverting a modulated baseband signal into a modulated RF signal

Although the RF signal exiting the transceiver is now at a suitable frequency for wireless transmission, it is still too weak to travel over any appreciable distance. Likewise, any signal being received by the antenna is too weak to be sent directly to the transceiver. The RF front-end serves two functions: It amplifies signals coming from the transceiver to a power level suitable for transmission, and it amplifies weak signals coming from the antenna to a level that can be detected by the transceiver. The quality and performance of the RF front-end is what determines a radio's overall RF performance. Specifications such as output power and receiver sensitivity are directly determined by the front-end components.

The last component in the radio is the antenna itself, whose purpose is to interface the electric currents flowing in the radio's circuitry to electromagnetic waves in free space. Depending on the radio's intended application, the type of antenna used can vary widely. The quality and performance of the antenna used in a radio has as much impact on its total performance as that of the RF front-end. For this reason, it is important to understand how to properly utilize the type of antenna used in any particular application.

Radio Frequency Signals

Before getting into the higher-level aspects of RF communications, let's cover a couple of basic principles and theories. To begin, we'll discuss some of the fundamental properties of analog signals. This is important even for digital wireless communications because the majority of RF components are-and will remain for a long time-analog in nature.

The basic building block for all analog signals is a single sinusoidal tone. A sinusoidal tone is a signal whose amplitude variation is defined by the trigonometric sine function (typically, however, the cosine function is used when mathematically expressing a sinusoidal function in communications theory). Sinusoids are considered to be functions of time, and an example of a sinusoidal function is shown in Equation 1, where f is the sinusoidal frequency (in Hertz) and $$ represents the phase shift of the sinusoid. Figure 2-3 illustrates a sinusoidal voltage in the time domain.

Image from book
Image from book
Figure 2-3: Time-domain waveform of a sinusoidal voltage
Note 

Even though this equation is defined in terms of voltage, this doesn't mean sinusoids are restricted to voltages. A sinusoid can be defined in terms of any unit of measure.

Every sinusoid has three basic properties that completely describe its characteristics: amplitude, frequency, and phase. Of these three properties, amplitude and frequency are probably the easiest to understand. Amplitude refers to how large of an excursion is generated, or how strong a signal, and is represented by the A coefficient in Equation 1. The frequency of a sinusoidal signal refers to how many cycles of the repeating sine function occur per second. For example, a 2.4-GHz signal has 2,400,000,000 cycles of the sine function every second. The phase of a sinusoid is a somewhat elusive concept, but is most easily thought of as a shifting of the sinusoid's waveform along the x-axis (usually time). Figure 2-4 illustrates two sinusoids with a phase shift between them of 90 degrees.

Image from book
Figure 2-4: Phase shift between two sinusoids

There are two methods for analyzing any analog signal: time-domain analysis and frequency-domain analysis. Time-domain analysis is when a signal is plotted as a function of time, as shown in Figures 2-3 and 2-4. On the other hand, frequency-domain analysis is made possible through the use of the Fourier Transform, which allows a time-domain signal to be separated into its individual sinusoidal components. Figure 2-5 shows the RF signal envelope of an 802.11a signal in the time-domain, and Figure 2-6 shows the same signal in the frequency-domain. Frequency-domain analysis is a more intuitive method for examining and interpreting analog signals than time-domain analysis because it clearly shows all of the spectral components of a signal. This is advantageous because most modulations encode data in the frequency (spectral) domain.

Image from book
Figure 2-5: Time-domain plot of RF voltage (envelope)
Image from book
Figure 2-6: Frequency-domain plot of 802.11a 54-Mbit/sec signal

One of the characteristics of analog signals is how much spectrum they occupy, or in other words, how wide they are when viewed in the frequency-domain. This characteristic is referred to as a signal's bandwidth. For example, the 802.11a signal shown in Figure 2-6 has a bandwidth of approximately 16.6 MHz. Note that only the region of the signal with the higher power level is considered when calculating a signal's bandwidth; the remaining spectral content is noise and distortion byproducts of the main signal. Because the graph in Figure 2-6 is plotted in decibels (dB), the spectral content outside of the 16.6-MHz bandwidth is much weaker than the actual 16.6-MHz signal. Typically, the more data that a signal contains, the more bandwidth it will occupy. For example, a broadcast FM radio signal has a bandwidth of about 150 kHz (or 0.150 MHz), which is considerably smaller than the typical WLAN signal bandwidth of 16.6 MHz.

Electromagnetic Waves

The existence of electromagnetic waves is what makes transmission of RF signals over wireless links possible. Electromagnetic waves are time-varying electric and magnetic fields that are able to propagate, or travel, through space. The way in which electromagnetic waves propagate depends on several factors, the two most important being the frequency of the signal and the environment through which the wave is traveling.

The frequency of the signal determines the wavelength of the electromagnetic wave. A signal's wavelength describes the amount of distance traveled by the signal between adjacent peaks in the signal and can be calculated by dividing the speed of light (for the medium through which the wave is traveling) by the frequency of the signal (as shown in Equation 2). Wavelength dictates many aspects of how a propagating electromagnetic wave will behave. Long wavelengths, on the order of 160–20 meters (corresponding to frequencies between 1.9 to 14 MHz), benefit from atmospheric phenomena that allow them to travel for great distances, oftentimes to opposite sides of the world. Shorter wavelengths (higher frequencies) tend to travel in straight lines and are blocked by obstructions such as walls and buildings. You can visualize this effect by thinking of an electromagnetic wave as a beam of light coming from a flashlight (in fact, light is a type of electromagnetic wave). This is commonly referred to as the line-of-sight characteristic of a signal.

Image from book

Another characteristic of electromagnetic waves is that they will reflect off of conductive surfaces. The ability of a surface to reflect an electromagnetic wave depends on its conductivity as well as its size as compared to a signal's wavelength. Since microwave signals (let's say above 1 GHz) have rather short wavelengths, they are easily reflected by metallic objects. Because of this, microwave signals are commonly affected by a phenomenon called multipath interference. Multipath interference occurs when multiple copies of a signal arrive at the same location but have taken different paths to get there. One way this can happen is by a signal being reflected off of multiple surfaces as it travels through space. Multipath interference is more common in indoor and urban environments and is a major problem that severely affects the performance of wireless communication systems. This degredation in system performance is a result of the blurring together of the multiple signals generated by multipath interference. This blurring is caused by each signal arriving at the receiving antenna at slightly different times since each signal had to travel a slightly longer or shorter path than the others. Imagine trying to listen to a conversation and hearing multiple echoes of the same conversation at the same volume as the conversation itself.

Units of Measure

When working with electronic and RF components, there are several units of measure that always seem to come up. This section provides a brief description of each of these units so no one gets left behind in the following discussions.

Voltage

Voltage is a measure of the difference in electric potential between two conductors and is measured in volts (V). Electric potential can be thought of as the ability of an electric field to cause an electric current to flow through a conductor. Greater differences in electric potential (in other words, higher voltages) are capable of generating larger electric currents. A common analogy is to compare electric potential to the amount of pressure inside a water pipe. The higher the pressure, the more water can be forced through the pipe.

Current

Current is a measure of the number of electrons that flow through a conductor in a finite amount of time and is measured in amperes (or amps). One amp is approximately 6.241×10 18 electrons per second. To continue with the prior analogy, the flow of electric current is analogous to the amount of water flowing through a pipe.

Power

Power is a measure of how much energy is absorbed or generated in a finite amount of time. Power is typically measured in watts and is equal to 1 Joule of energy transferred per second. In terms of electrical circuits, one watt is equal to the current flowing through a circuit (measured in amps) multiplied by the amount of voltage supplied to the circuit (measured in volts) as shown in Equation 3. In the field of RF electronics, pretty much everything is measured in terms of watts.

Image from book

The Decibel

The decibel is a convenient method for expressing the ratio between two numbers and is indicated by the abbreviation dB. It is important to realize that the decibel is a representation of a ratio rather than of a single number. It is possible to express a specific measurement or value in terms of decibels, but its measurement has to be referenced to a unit. First, we'll look at expressing a regular ratio and then move on to the subject of referenced decibels.

There are two methods for converting a ratio into decibels and which method to use depends on the type of numbers being expressed by the ratio. If the ratio represents voltages or currents, then the correct conversion equation is Equation 4. If the ratio to be converted represents power, then the correct conversion equation is Equation 5. It is important to remember the difference between the methods, as using the wrong conversion will result in an incorrect answer that is always off by a factor of 2.

Image from book
Image from book

For someone familiar with electronics, the distinction between Equations 4 and 5 is probably clear. For the layperson, this is a possible point of great consternation. Luckily, it can be safely assumed that for the purposes of this text, Equation 5 is almost always the correct conversion method because RF and communications system engineers almost always think of things in terms of power. Gain is always thought of in terms of power gain. Loss is always thought of in terms of power loss. Signal strength is always thought of in terms of power.

For example, if someone states that a device has a power gain of 20 dB, this means that the power exiting the device is 10 (20/10) or 100 times greater than the power that entered the device. Or if a device has 3 dB of loss, it means that the power leaving the device is 10 (3/10) or 2 times smaller than the power entering the device.

One of the benefits of using decibels is that it simplifies many of the calculations necessary when working with RF systems. This is because the addition of two ratios once they have been converted into decibels is the same as multiplying the ratios themselves. An example of this is shown in Equations 6 and 7. Equation 6 shows that the gain (in decibels) of three series amplifiers is the sum of the three gains (in decibels). Equation 7 shows the same calculation, except that the gains are expressed in terms of regular (non-decibel) numbers. Since most quantities in RF are expressed in decibels to begin with, the method used in Equation 6 is typically the most straightforward.

Image from book
Image from book

There are times when it is necessary to express a specific value in decibels. This is accomplished through the use of referenced decibels. The first step is to determine which unit the value is to be referenced to. Typically, this is just the unit scale used to measure the value. For example, if the value to be expressed is 4 volts, then the reference value is 1 volt. This is not always the case though. It is sometimes more convenient to use other values; for example, the reference value commonly used for RF power measurements is 1 milliwatt.

The second step is to divide the value to be converted into dB by the reference value chosen. Notice that by dividing the two values, a ratio between the value and the reference unit has been created. This ratio is then converted into decibels just like before, except there is one more important step to take.

The reference value used in the conversion must be indicated in order for the result to be meaningful. This is done by adding a suffix letter to the dB unit designator. Table 2-1 contains a list of the most common types of referenced decibel units along with their reference values.

Table 2-1: Examples of Referenced Decibel Values

Unit

Reference Value

Example

dBm

1 milliwatt

100 mW = 100 mW / 1 mW = 100 = 20 dBm
1 uW = 1 uW / 1 mW = 1/1000 = 30 dBm

dBV

1 volt

2 Volts = 2 V / 1 V = 2 = 6 dBV

dBi

Gain of isotropic radiator

3 dB above isotropic gain = 3 dBi

Efficiency

Efficiency is a measure of how well a device converts energy from one form into another. In the context of a wireless communications system, there are many components that perform various types of energy conversion. The efficiency with which these components are able to convert energy is an important factor for two main reasons: It determines how long the battery in your mobile wireless device (laptop or PDA) is going to last, as well as how hot the device gets while it is in the process of draining your battery.

In a perfect world, the efficiency of every component would be 100 percent. That is, all of the energy that goes into a device is perfectly converted to the desired energy type. An example of this would be an amplifier that perfectly converts the energy it is taking from your battery into RF energy that is delivered to the antenna. You can be assured, however, that this is not how things work in the real world. That energy being sucked from your battery isn't perfectly converted into the RF energy you want. In fact, RF components typically have efficiencies that are on the order of 10 percent to 30 percent. This means that for every watt of power your battery supplies, only 0.16 to 0.30 watts of RF power are actually produced. The remaining power goes into making things nice and warm.

Efficiency is typically expressed as a percentage and is defined as the amount of energy that leaves a system in the desired form divided by the total amount of energy entering the system. It can be applied to any component that converts one form of energy to another. The two most common RF devices to calculate the efficiencies of are amplifiers and antennas. An example of how to calculate the efficiency of an RF amplifier is shown in Equation 8. Simply divide the amount of RF power generated by the amplifier by the amount of DC power supplied to the amplifier and then multiply by 100.

Image from book

Gain and Loss

Gain and loss are measures of how signal power levels are affected by various components of an RF system. If the signal power exiting an RF component is greater than the signal power entering the component, that component can be said to have gain. Conversely, if the signal power exiting a component is lower than the signal power entering the component, it is said to have loss. The gain of a device is calculated by dividing the output power (in watts) by the input power (in watts) as shown in Equation 9. This ratio can then be converted into decibels by using Equation 5, shown earlier in the chapter. If the power levels are already expressed in decibels, then gain in decibels is calculated by simply subtracting the input power from the output power as in Equation 10. It is also interesting to note that gain and loss are the complement of each other. When expressed in decibels, a component with gain has a negative loss, and a component with loss has a negative gain.

Image from book
Image from book

Modulation

Modulation is the process of embedding data that is to be transmitted by a communications system onto an analog carrier that will then be used to transport the data. This is done by converting the data to be transmitted from its native format (either analog or digital) into an analog signal that is suitable for effective RF transmission. The method by which this process is accomplished varies depending on the type of data being modulated as well as the type of medium through which the RF signal will be traveling.

When discussing modulation, there are several terms that describe the various signals used during the modulation process. The term baseband is used to refer to the lowest-frequency signal in an RF system. In a digital system, the baseband signal is typically the signal that is being passed back and forth between the Digital Signal Processor (DSP) and the transceiver. In analog modulations, the baseband signal is the actual data itself. The RF carrier is a single sinusoidal signal whose frequency is the same as the desired RF output signal. Once the baseband signal has been upconverted to the RF transmission frequency, it is then referred to as the modulated RF carrier.

The various types of modulations may be separated into two broad categories: analog and digital. The two predominant types of analog modulation are amplitude modulation and frequency modulation. There are numerous types of digital modulation, but those of interest to readers of this book are PSK, CCK, and QAM. While entire textbooks have been written on the subject of these modulations, it is possible to take a quick look at how they work and the basic characteristics of each.

Analog Modulation Techniques

Analog modulation is the process of converting an analog input signal into a signal that is suitable for RF transmission. It is performed by varying the amplitude and/or phase of an RF carrier signal based on an analog input signal's time-varying properties. Analog modulation techniques are ideally suited for signals that are inherently analog in nature, such as the human voice and music. Although analog modulation can be used to transmit data, it is inefficient when compared to digital modulation techniques.

Pulse Modulation (PM) Starting with the most basic of all the modulation types is pulse modulation. This modulation is the type used by Benjamin Franklin for his telegraphy system. Pulse modulation is only capable of conveying either an "on" or an "off" state by switching the RF carrier signal on or off. While this modulation technique could be used to transmit binary data, it has the drawback of being extraordinarily inefficient for respectable data rates. Pulse modulation is still commonly used for low-data-rate telemetry signals and Morse code.

Amplitude Modulation (AM) This modulation is one of the most common analog modulation techniques. It is accomplished by varying the amplitude of an RF carrier signal based on the amplitude of an analog input signal. As the amplitude of the analog input signal varies in time, the amplitude of the RF signal is made proportional to the time-varying analog input signal amplitude. Figure 2-7 shows an example analog input signal, and Figure 2-8 shows the resulting modulated RF signal. The shape of the modulated RF signal is oftentimes referred to as the signal envelope, which describes the average power level of the RF signal as a function of time rather than the signal itself. Amplitude modulation is commonly used for long-distance voice communications, broadcast radio, and also in early analog cellular networks.

Image from book
Figure 2-7: Baseband audio signal
Image from book
Figure 2-8: Resulting AM modulated RF signal

Frequency Modulation (FM) This is another classic modulation technique. In a frequency modulated signal, it is the frequency of the RF carrier signal that varies as a function of time. The amount of variation in the RF carrier's frequency is determined by the amplitude of the input baseband signal. Figure 2-9 shows an example of an analog input signal, and Figure 2-10 shows the resulting frequency modulated RF carrier.

Image from book
Figure 2-9: Baseband input signal
Image from book
Figure 2-10: Resulting FM modulated RF signal

Digital Modulations

Digital modulation is the process of converting a digital bitstream into an analog signal suitable for RF transmission. This process is usually accomplished using digital signal processing due to the levels of complexity involved with modern modulations. A digital bitstream is input into the digital signal processor (DSP) and the analog output signal is generated using a digital-to-analog converter (DAC). It is worth noting that the digital data is not being converted directly to an analog signal. The DSP is analyzing the digital data and synthesizing an appropriate analog signal to represent the digital data based upon the type of modulation being implemented.

The first step performed by the DSP is to divide the digital bitstream into small equally sized groups of bits called symbols. The number of bits contained in, or represented by, a symbol depends on the type of modulation being used, but most modern modulation techniques can represent anywhere from 1 to 6 bits per symbol. These groups of bits are then used to form a sequence of symbols. Each of these symbols represents a unique analog output from the DAC. This process of separating the bitstream into symbols using a 16-QAM modulation is shown in the top portion of Figure 2-11.

Image from book
Figure 2-11: Process of dividing a digital bitstream into individual symbols using 16-QAM modulation

In order to conveniently represent the analog output of the DAC in a graphical manner, a type of graph called a constellation diagram is often used. A constellation diagram graphically depicts the various magnitudes and phases of the generated analog signal. The points on a constellation diagram represent unique combinations of magnitude and phase that, in turn, represent the various symbols used in a particular modulation.

The DSP steps through the sequence of symbols one at a time and for each symbol synthesizes the corresponding analog signal. This is shown in Figure 2-11 by the constellation diagrams for each symbol. The transition between each symbol is shown in Figure 2-12. The rate at which the DSP steps through the sequence of symbols is called the symbol rate. This rate is typically governed by the wireless standard being implemented. Since the number of bits represented by each symbol is known, the raw throughput can be calculated by multiplying the number of bits represented by each symbol by the symbol rate.

Image from book
Figure 2-12: Constellation diagram showing the transition between symbols

Now that we've covered the fundamental aspects of digital modulations, let's take a look at the various types of digital modulation.

Phase Shift Keying We'll start our discussion of digital modulations with phase shift keying (PSK). PSK is one of the simplest digital modulation techniques and is also one of the most robust. Simplicity and robustness do come at a price, however, as the data rate achievable by a PSK signal is rather low compared to other modulation techniques.

In a PSK-modulated signal, the phase of an RF carrier is varied among specific phases depending on the symbol being represented. Two common types of PSK modulation are binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK). BPSK utilizes two discrete phase states, and QPSK utilizes four discrete phase states. Since QPSK can represent twice as many symbols (as it has twice as many phase states) as BPSK, QPSK is capable of twice the data rate of BPSK. Constellation diagrams are shown in Figure 2-13.

Image from book
Figure 2-13: Constellation diagrams for BPSK (on left) and QPSK (on right)

There is a variation of PSK called differential PSK that is often implemented in wireless communications. The basic concept of DPSK is the same as in PSK except that rather than having each phase state represent a particular symbol, the transition between phases is used to represent each symbol. This reduces the amount of complexity required in the receiver's DSP to demodulate a signal. Table 2-2 shows a list of the absolute phases used in QPSK to represent each symbol, as compared to the amount of relative phase shift used in DQPSK to represent the same symbol.

Table 2-2: Different Symbol Representations of QPSK and DQPSK

Symbol

QPSK (Absolute Phase)

DQPSK (Phase Change)

00

–135

0

01

+135

+90

11

+45

+180

10

–45

–90

Complementary Code Keying (CCK) This is a modulation technique that utilizes spread-spectrum techniques coupled with the unique mathematical properties of complementary sequences to achieve higher data rates than ordinarily possible with plain spread-spectrum communications. The complementary sequences used in CCK modulation change the manner in which the symbols used in the modulation represent the data as compared to a regular spread-spectrum system. The mathematical properties associated with this process allow a CCK signal to be transmitted at the same symbol rate as a conventional spread-spectrum signal but with a much higher actual data rate. However, CCK is of less interest these days since it is largely being phased out in favor of OFDM-based systems.

The primary application of CCK modulation was in the 802.11b standard for the 5.5 and 11 Mbit/sec data rates. It was chosen because it allows for higher data rates while still using spread-spectrum DQPSK modulation of the lower speed 1 and 2 Mbit/sec 802.11 legacy standard. This allowed 802.11b networks to achieve faster data rates while still being compatible with older 802.11 legacy networks.

Quadrature Amplitude Modulation (QAM) This technique is a complex digital modulation capable of extremely high data rates. These high data rates are possible because of the large number of possible symbols that can be created using this modulation technique. There are various types of QAM, but the two most common are 16-QAM and 64-QAM, each of which is named after the number of symbols used in the modulation. Each symbol in a 16-QAM modulation represents 4 bits, and each symbol in a 64-QAM modulation represents 6 bits.

Symbols are constructed by varying both the magnitude and phase of the baseband signal. Each unique amplitude and phase combination represents a symbol. In most QAM signals, when these symbols are plotted on a constellation diagram, they are visible as a rectangular grid. Figure 2-14 shows the constellation diagrams for both a 16-QAM signal and a 64-QAM signal.

Image from book
Figure 2-14: Constellation diagrams for 16-QAM and 64-QAM signals

Spread Spectrum and Multiplexing

Spread spectrum and multiplexing are two methods for sharing a fixed amount of bandwidth between multiple users. These techniques are utilized because there is only a certain amount of available RF spectrum. In fact, the commercial communications bands are quite small compared to the number of users competing for the spectrum. The techniques used to help alleviate this problem have undergone continuous development and revision. This section will cover the most prevalent of these techniques today.

Spread Spectrum

The concept of spread-spectrum communications was originally developed as a means for the obfuscation (or hiding) and scrambling of a communications signal. Early development was predominately related to the application of this technique toward military communications.

Spread spectrum operates by taking an ordinary communication signal and then spreading it across a much wider bandwidth than that occupied by the original signal. It is from this spreading process that the technique derives its name. There are two unique characteristics of a signal once it has been spread.

First, the spreading process spreads the power level of the signal over a wider range of frequencies than it initially occupied. This reduces the amount of signal power present at any one particular frequency, but it does not change the total amount of power present in the entire signal. If the spreading is wide enough, the signal can actually seem to "disappear" into the noise, which has merit in military communications because once a signal falls below the noise floor, it is very difficult to even detect its presence.

Second, it is virtually impossible to recover the original signal without knowing exactly how it was spread to begin with. This means that even if someone is able to detect the communications signal, they won't be able to extract any data from it. However, if the original spreading technique is known, then a spread-spectrum signal is very easy to detect and unspread.

Even though they are beneficial for other reasons, these same properties are what make spread-spectrum techniques appealing for wireless communications. Now that you know the basic ideas behind spread spectrum, let's take a look at two of the prevalent spread-spectrum techniques.

Frequency Hopping Spread Spectrum (FHSS) This technique operates by rapidly changing the frequency at which a communications signal is being transmitted. Because the transmission frequency is changing at a rapid rate, the signal is effectively spread over a greater bandwidth. In order to successfully receive a signal that has been transmitted using FHSS, the frequency that a receiver is listening to has to move in tandem with the transmitter. In order for the receiver to successfully track the transmitter, it has to know the sequence of frequencies that the transmitter will be using, the amount of time that the transmitter will use each frequency, and the current location of the transmitter in the sequence of frequencies. The most common commercial application of Frequency Hopping Spread Spectrum is the Bluetooth wireless standard.

Direct Sequence Spread Spectrum (DSSS) This is the most commonly used spread-spectrum technique. DSSS works by combining an ordinary communications signal with pseudorandom noise in the spreading process. The resulting signal appears to be random noise, but when the same pseudorandom noise is used to despread the signal, the original signal is extracted.

The core of DSSS is the pseudorandom noise used in the spreading and despreading process. This noise is generated from a sequence of pseudorandom bits called a PN sequence. The key characteristic of the PN sequence is that it is not a truly random sequence and is, in fact, completely deterministic. A pseudorandom algorithm is used to generate the PN sequence, and due to the nature of any algorithm, if the same starting condition is employed, the algorithm will always generate the same output. This means that a receiver can generate the exact same noise signal used by the transmitter if it knows the algorithm and initial condition used. Without this knowledge, however, the pseudorandom noise generated by the PN sequence will appear to be random.

One of the useful characteristics of DSSS is that without the correct despreading PN sequence, the signal appears as random noise. Taking this one step further, if multiple DSSS signals are transmitted in the same communications channel but each of them uses different PN sequences, then the resulting signal will still appear to be random noise. Now, this is where the magic begins to happen. Let's say there are four different DSSS signals that are all transmitted at the same frequency but with different PN sequences. Before dispreading, the combined signal still appears like random noise, but if a PN sequence matching the one used to spread one of the original four signals is applied to the combined received signal, then the signal originally spread with that PN sequence will "pop" out of the noise. This is true for each of the original four signals. If the corresponding PN sequence is used, that signal will be extracted from the received signal. This property of DSSS effectively allows multiple users to share the same communications channel simultaneously.

DSSS is, in fact, a highly sophisticated technique with an elaborate mathematical foundation. To effectively exploit the properties just described, the PN sequences employed by each of the users must be generated in a manner that satisfies numerous mathematical requirements. The method by which these sequences are generated varies depending on the wireless standard being implemented. Code Division Multiple Access is probably one of the most famous applications of DSSS. Other systems that incorporate DSSS are 802.11 and the Global Positioning System (GPS). The exploitation of the unique properties of DSSS signals is at the core of these systems and this is what makes their operation possible.

Multiplexing

Multiplexing is the process of dividing a single communication channel into subcomponents so that the channel can be shared among numerous users and/or sources. The manner in which the channels are divided depends on the type of multiplexing being utilized. In the following sections, the most relevant multiplexing schemes are discussed.

Frequency Division Multiplexing (FDM) This is the simplest form of multiplexing. In this form of multiplexing, a separate frequency is used for every signal. Because any modulated signal has a bandwidth associated with it, the available spectrum is often divided into channels with each channel having slightly more bandwidth than the bandwidth of the signal that is to fit inside the channel. Figure 2-15 illustrates the centering of RF signals inside different channels. These channels are then individually assigned to each user. One of the major problems with FDM is its inefficient usage of the frequency spectrum. Each user has their own dedicated channel, which means users can't share the same channel. Once the system runs out of available channels, additional users must wait until another user disconnects.

Image from book
Figure 2-15: Using different channels for different signals

Time Division Multiple Access (TDMA) This technique multiplexes each channel in the temporal dimension by dividing the channel into a finite number of timeslots. Each timeslot is a short segment of time that is allocated to an individual user during which that user is allowed to transmit and receive data. Once every user has been given a chance to communicate with the system, the system starts cycling through the timeslots again. TDMA is depicted graphically in Figure 2-16. By multiplexing a single channel in the temporal dimension, multiple users are able to share the same amount of bandwidth. TDMA was the multiplexing approach used by the old 2G digital cellular networks.

Image from book
Figure 2-16: The temporal multiplexing scheme used in TDMA

Code Division Multiple Access (CDMA) This technique is a direct implementation of Direct Sequence Spread Spectrum. Each user on a CDMA system has their own unique PN sequence that they employ during the spreading process. All of the PN sequences being used on a particular channel are chosen such that they are statistically uncorrelated. The implications of being statistically uncorrelated are that when many users are transmitting on the same frequency, the sum of all these signals generates uncorrelated random noise. The basestation will separately despread each of the individual signals using the PN sequence associated with each user. CDMA also exploits the phenomena of process gain to elevate each user's signal out of wideband noise when the correct PN sequence is applied, allowing many users to share the same frequency channel without having to synchronize with each other.

Orthogonal Frequency Division Multiplexing (OFDM) This is, in its purest form, a technique for achieving higher throughput wireless communications than normally achievable by traditional modulation techniques. This is done by dividing the high-speed digital bitstream being transmitted into several lower-speed bitstreams operating in parallel. Each of these bitstreams is then modulated onto separate subcarriers using standard modulation techniques. The frequencies of the subcarriers are carefully chosen such that they are orthogonal with each other. As a result of the subcarriers being at orthogonal frequencies, crosstalk and interference amongst the various subcarriers is prevented.

A signal that has been generated using OFDM techniques has a higher spectral efficiency than a signal containing the same data but generated with traditional modulation. The primary benefit of OFDM is that it eliminates many of the problems associated with the high symbol rates required in order to achieve high data rates with traditional modulations. OFDM also operates at a much slower symbol rate than non-OFDM systems at comparable data rates. Because of the slower symbol rate, the guard interval (amount of blank or whitespace time) required between symbols is much less than the symbol time (how long each symbol is transmitted). This means that a greater percentage of the time is spent transmitting data rather than waiting as compared to non-OFDM systems. Since the data rate of each subcarrier is slower than the combined data rate, the symbol rates required are lower, which reduces the effects of multipath interference. This characteristic of OFDM allows for the high throughputs offered by 802.11a and 802.11g.


Previous Page
Next Page