Chapter 1. Introduction

The modern notion of wireless communication relies on the transmission and reception of electromagnetic waves. The concept was theorized by Maxwell and demonstrated in practice by Hertz in 1888 [151]. Others contributed to the early demonstration of wireless communication, including Lodge, Bose, and de Moura.

The earliest examples of wireless communication used what is now known as digital communication. The term digital comes from digitus in Latin, which refers to a finger or toe. Digital communication is a form of communication that involves conveying information by selecting one symbol from a set at any given time. For example, by extending just one finger, a hand can convey one of five symbols. Extending two fingers at a time, a hand can convey one of 5 × 4 = 20 symbols. Repeating the hand gestures quickly allows multiple symbols to be sent in succession. This is the essence of digital communication.

Digital communication using electromagnetic waves involves varying the parameters of continuous-time signals (or analog signals) to send a sequence of binary information, or bits. The most common kind of wireline communication system in the 1800s was the telegraph, which used Morse code to send digital messages consisting of letters, numbers, stops, and spaces across the country, and even the ocean, over wires. A wireless telegraph was patented by Marconi in 1896, which is generally accepted as the first wireless (electromagnetic) digital communication system. The first transatlantic wireless Morse code message was sent by Marconi in 1901 [48]. The history of wireless digital communication is as old as wireless itself.

Though interest in wireless telegraphy continued, digital communication gave way to analog communication as the primary modulation method used in wireless applications until the 1980s. Using analog communication, the parameters of a waveform are varied continuously based on a continuous-time signal at the input. An early example of analog communication is the original telephone system, developed in the late 1870s [55], in which acoustic speech waves were converted via a microphone to electrical signals that could be amplified and propagated on a wire. Early examples of wireless analog communication, still in use today, include AM (amplitude modulation) and FM (frequency modulation) radio, and older broadcast TV (television). Analog communication has been widely used in wireless communication systems, but it is now being replaced by digital communication.

The primary reasons that digital communication has now overtaken analog communication are the prevalence of digital data and advancements in semiconductor technologies. Digital data was not common before the development of computers and computer networks. Nowadays, everything stored on a computer or exchanged over the Internet is digital, including e-mail, voice calls, music streaming, videos, and Web browsing among others. Advances in integrated circuits have led to increasing numbers of transistors in a given amount of semiconductor area, which has increased the potential of digital signal processing. While not required for digital communication, leveraging digital signal processing allows for much better transmitter and receiver algorithms. In wireline telephony, digital communication circuits began to completely replace analog circuits in the network backbone in the 1960s, in part because of the noise resilience of digital signals when transmitted over long distances (repeaters are less sensitive to noise than amplifiers). Similar developments in wireless communication, however, did not start in earnest until the 1980s. The reason, it seems, is that it was only in the 1980s that integrated circuit technology had developed to the point where it could be considered for use in portable wireless devices. About the same time, the compact disc started replacing the tape and vinyl record.

Digital communication is now a fundamental part of wireless communication. In fact, almost all current and next-generation wireless communication systems (actually all developing standards as well) make use of digital communication. Wherever there is currently a wire, there is a proposal to eliminate that wire via wireless. There are a multitude of commercial, military, and consumer applications of wireless digital communication.

In the 1990s, there was a transition of broadcast radio from analog to digital technology. In 1995, the digital audio broadcasting (DAB) standard, also known as Eureka 147, was developed [333]. DAB is used in Europe and other parts of the world, coexisitng in some cases with traditional AM and FM emissions. It uses a digital modulation technique known as COFDM (coded orthogonal frequency-division multiplexing) to broadcast multiple digital radio streams [304]. COFDM is a particular form of OFDM, which is discussed extensively in this book.

The United States uses a different digital method known as HD Radio (a trademarked name), which was approved by the FCC (Federal Communications Commission) in 2002 as the AM and FM digital broadcasting system to transmit digital audio and data along the existing analog radio signals [317, 266]. HD Radio uses a proprietary transmission technique, which also uses OFDM but fits in the gaps between existing FM broadcast stations. HD Radio started rolling out in force in 2007 in the United States. Digital coding and modulation techniques permit compact-disc-quality stereo signals to be broadcast from either satellites or terrestrial stations. In addition to audio quality improvement, digital audio broadcasting can provide other advantages: additional data services, multiple audio sources, and on-demand audio services. Just like today’s analog AM and FM radio, HD Radio requires no subscription fee. HD Radio receivers are factory installed in most vehicles at present. Therefore, owners of new cars immediately have access to the HD Radio audio and data services offered [317].

In the 1990s, the DVB (digital video broadcasting) suite of standards was initiated for digital and high-definition digital television broadcasting [274, 275]. DVB standards are deployed throughout much of the world, except in the United States. Like DAB, DVB also uses an OFDM digital modulation technique. There are several different flavors of DVB specified for terrestrial, satellite, cable, and handheld applications [104].

The United States chose to follow a different approach for high-definition digital broadcasting that produces a digital signal with a similar spectrum to the analog NTSC signal. The ATSC (Advanced Television Systems Committee) digital standard employs 8-VSB (vestigial sideband) modulation and uses a special trellis-encoder (one of the few examples of trellis-coded modulation in wireless systems) [86, 276, 85]. ATSC systems require directional antennas to limit the amount of multipath, since equalization is relatively more difficult compared with the OFDM modulation used in the DVB standard. In 2009, after over a half-century of use, the analog NTSC systems were replaced by ATSC in the United States.

The first generation of cellular communication devices used analog communication, in particular FM modulation, for the wireless link between mobile users and the base stations. The technology for these systems was conceived in the 1960s and deployed in the late 1970s and early 1980s [243, 366, 216]. The use of analog technology gave little security (it was possible to eavesdrop on a call with the right radio gear), and limited data rates were supported. Many similar, but not compatible, first-generation systems were deployed around the same time, including AMPS (Advanced Mobile Phone System) in the United States, NMT (Nordic Mobile Telephony) in Scandinavia, TACS (Total Access Communication System) in some countries in Europe, Radiocom 2000 in France, and RTMI (Radio Telefono Mobile Integrato) in Italy. Japan had several different analog standards. The plurality of standards deployed in different countries made international roaming difficult.

The second and subsequent generations of cellular standards used digital communication. Second-generation systems were conceived in the 1980s and deployed in the 1990s. The most common standards were GSM (Global System for Mobile Communications) [123, 170], IS-95 (Interim Standard 1995, also known as TIA-EIA-95) [139, 122], and the combination IS-54/IS-136 (known as Digital AMPS). GSM was developed in a collaboration among several companies in Europe as an ETSI (European Telecommunications Standards Institute) standard. It was adopted eventually throughout the world and became the first standard to facilitate global roaming. The IS-95 standard was developed by Qualcomm and used a new (at the time) multiple access strategy called CDMA (Code Division Multiple Access) [137]; therefore, IS-95 was also known as cdmaOne. IS-95 was deployed in the United States, South Korea, and several other countries. The IS-54/IS-136 standard was developed to provide a digital upgrade to the AMPS system and maintain a certain degree of backward compatibility. It was phased out in the 2000s in favor of GSM and third-generation technologies. The major enhancements of second-generation systems were the inclusion of digital technology, security, text messaging, and data services (especially in subsequent enhancements).

The third generation (3G) of cellular standards, deployed in the 2000s, was standardized by 3GPP (3rd Generation Partnership Project) and 3GPP2 (3rd Generation Partnership Project 2). UMTS (Universal Mobile Telecommunications System) was specified by 3GPP as the 3G technology based on the GSM standard [193, 79]. It used a similar network infrastructure and a higher-capacity digital transmission technology. The evolution of cdmaOne led to CDMA2000, which was standardized by 3GPP2 [364, 97]. Notably, both UMTS and CDMA2000 employ CDMA. The major advance of third-generation standards over the second generation was higher voice capacity (the ability to support more voice users), broadband Internet access, and high-speed data.

The fourth generation of cellular standards was the object of much development, and much debate (even over the definition of “fourth generation”). In the end, two systems were officially designated as fourth-generation cellular systems. One was 3GPP LTE (Long Term Evolution) Advanced release 10 and beyond [93, 253, 299, 16]. The other was WiMAX (Worldwide Interoperability for Microwave Access), a subset of the IEEE 802.16 m standard [194, 12, 98, 260]. Though WiMAX was deployed earlier, 3GPP LTE became the de facto 4G standard. A major departure from third-generation systems, fourth-generation systems were designed from the ground up to provide wireless Internet access in a large area. 3GPP LTE is an evolution of 3GPP that supports larger-bandwidth channels and a new physical layer based on OFDMA (orthogonal frequency-division multiple access) where subcarriers can be dynamically assigned to different users. OFDMA is a multiple-access version of OFDM (orthogonal frequency-division multiplexing), which is discussed in Chapter 5. 3GPP LTE Advanced adds other new capabilities, including more support for MIMO (multiple input multiple output) communication enabled by multiple antennas at the base station and handset, and thus supports higher data rates. WiMAX is based on the IEEE 802.16 standard. Essentially, the WiMAX Forum (an industry consortium) is specifying a subset of functions for implementation, and appropriate certification and testing procedures will ensure interoperability. WiMAX also employs OFDMA, though note that earlier versions used a slightly different access technique based on OFDM. Fourth-generation systems make more use of multiple antennas via MIMO communication, which is discussed in Chapter 6. The fourth generation of cellular systems promises higher data rates than previous systems along with network enhancements such as simplified backhaul architectures.

Research on the fifth generation of cellular standards has begun in 3GPP. At the time of the writing of this book, various technologies are being considered to further improve throughput and quality and reduce latency and costs. There is great interest in continuing to push MIMO communication to its limits [321]. Massive MIMO promises hundreds of antennas at the base station to support more users simultaneously [223, 185, 42], and full-dimensional MIMO uses horizontal and vertical beamforming to support more users [238]. Millimeter wave MIMO systems making use of spectrum above 30GHz are also being considered for the fifth generation of cellular systems [268, 262, 269]. Research on all of these topics is ongoing [45, 11].

The most widely used WLAN standards are developed within the IEEE 802.11 working group [279, 290]. IEEE 802 is a group that develops LAN and MAN (metropolitan area network) standards, focusing on the physical (PHY), media access control (MAC), and radio link protocol (link) layers, considered Layer 1 and Layer 2 in typical networking literature [81]. The IEEE 802.11 working group focuses on WLANs. The Wi-Fi Alliance is an organization for certifying IEEE 802.11 products to guarantee interoperability (often Wi-Fi is used interchangeably with IEEE 802.11, though they are not exactly the same). Different subgroups of IEEE 802.11 are associated with different letters, such as IEEE 802.11b, IEEE 802.11a, IEEE 802.11g, and IEEE 802.11n.

The original IEEE 802.11 standard supported 0.5Mbps (megabit-per-second) data rates with a choice of two different physical layer access techniques, either frequency-hopping spread spectrum or direct-sequence spread spectrum in the 2.4GHz ISM band. IEEE 802.11b provides data rates of 11bps by using Complementary Code Keying modulation, extending the direct-sequence spread spectrum mode. IEEE 802.11a and IEEE 802.11g provide data rates of 54Mbps in the 5.8GHz and 2.4GHz bands, respectively, using OFDM modulation, which is discussed in Chapter 5.

IEEE 802.11n is a high-throughput extension of IEEE 802.11g and IEEE 802.11a that uses MIMO communication, combined with OFDM, to provide even higher data rates [360, 257]. MIMO enables a new class of modulation techniques, some of which can be used to send multiple data streams in parallel, and others that provide higher reliability as described further in Chapter 6. More advanced high-throughput extensions of IEEE 802.11 were developed as IEEE 802.11ac and IEEE 802.11ad. Two letters are used since single letters have been exhausted through other extensions of the standard. IEEE 802.11ac focuses on sub-6GHz solutions [29], and IEEE 802.11ad focuses on higher-frequency, in particular the 60GHz millimeter wave unlicensed band, solutions [258, 268]. Compared with IEEE 802.11n, IEEE 802.11ac supports more advanced MIMO capability (up to eight antennas) and multiuser MIMO communication, where the access point communicates with several users at the same time. IEEE 802.11ad is the first WLAN solution at millimeter wave, providing gigabit-per-second (Gbps) peak throughputs. The next generation of WLAN is currently in development under the name IEEE 802.11ay; it will support multiuser operation, targeting 100Gbps data rates and an extended transmission distance of 300–500m.

Most PANs are developed within the IEEE 802.15 working group [308]. The Bluetooth standard, IEEE 802.15.1a and later extensions, is perhaps the most familiar protocol. It is most commonly used for wireless headset connections to cell phones, wireless keyboards, and wireless computer mice. Another PAN standard is IEEE 802.15.4, known as ZigBee, intended for low-power embedded applications like sensor networks, home monitoring and automation, and industry controls [227]. IEEE 802.15.3c was a high-data-rate extension of 802.15 to the millimeter wave unlicensed band (around 57GHz to 64GHz), which was not as successful as WirelessHD [356], which was developed by an industry consortium [268]. These systems provide high-bandwidth connections in excess of 2Gbps for applications such as wireless HDMI (High-Definition Multimedia Interface) and wireless video display connections. The boundaries between WLAN and PAN are starting to blur, with IEEE 802.11ad taking over many of the functions offered by 60GHz PAN. It is likely that such developments will continue with IEEE 802.11ay.

Satellites, in the context of telecommunications, act as repeaters to help both point-to-point and point-to-multipoint transmissions of signals. Traditionally, communication satellites provide a wide range of applications, including telephony, television broadcasting, radio broadcasting, and data communication services [94]. Compared to other systems, communication satellite systems stand out because of their broad coverage, especially their ability to provide services to geographically isolated regions or difficult terrains. For example, mobile satellite services would target land mobile users, maritime users [180], and aeronautical users [164].

Satellites provide both long-distance (especially intercontinental) point-to-point or trunk telephony services as well as mobile telephony services. In 1965, Intelsat launched the first commercial satellite, named Early Bird, to provide intercontinental fixed telephony services. Communication satellite systems are able to provide worldwide mobile telephone coverage, also via digital communication technology [286]. The first GEO satellite providing mobile services, Marisat, was launched into orbit in 1976. Other examples of systems include Iridium, Inmarsat, and Globalstar. Satellite phones are inherently more expensive because of the high cost of putting a satellite in orbit and their low capacity. Satellite phones are useful in remote areas and for sea-based communication; their use in populated areas has been eclipsed by cellular networks.

Television accounts for about 75% of the satellite market for communication services [221]. Early satellite TV systems used analog modulation and required a large receiving dish antenna. In 1989, TDF 1 was launched as the first television direct broadcasting satellite. Now most satellite TV programming is delivered via direct broadcast satellites, which use digital communication technology. Some examples of current communications satellites used for TV broadcasting applications are Galaxy and EchoStar satellites in the United States, Astra and Eutelsat Hot Bird in Europe, INSAT in India, and JSAT satellites in Japan.

A recent application of satellite broadcast is high-definition radio. In the last 20 years, satellite radio has taken off in many areas [355]. The initial applications of satellites in radio were to provide high-fidelity audio broadcast services to conventional AM or FM broadcast radio stations. Now they are widely used for transmitting audio signals directly to the users’ radio sets. In satellite radio systems like SiriusXM [88], based on Sirius and XM technology [247, 84], digital communication is used to multicast digital music to subscribers. Other information may also be bundled in the satellite radio transmissions such as traffic or weather information.

A final application of satellites is for data communication. Satellite systems provide various data communication services, including broadcast, multicast, and point-to-point unidirectional or bidirectional data services [105]. Example services include messaging, paging, facsimile, data collection from sensor networks, and of course wireless Internet access. Unidirectional or broadcast data communication services are often provided by VSAT (very small aperture terminal) networks [4, 66, 188], using GEO satellites. VSAT networks work well for centralized networks with a central host and a number of geographically dispersed systems. Typical examples are small and medium-size businesses with a central office and banking institutions with branches in different locations. VSAT networks are also used for wireless Internet access in rural areas.

High-altitude platform (HAP) stations are a hybrid technology that combines the benefits of terrestrial and satellite communication systems. Examples of HAP are unmanned airships and manned/unmanned aircraft flying in the stratosphere just above the troposphere, at an altitude of about 17km or higher [76, 18, 103]. HAP stations may fill the gap between satellite-based communication systems, which are expensive and put high demands on the subscriber units because of the large distance to the satellites, and the terrestrial transmitters, which suffer from limited coverage. They may also be an alternative to cellular systems for telephony and wireless Internet access in parts of the world that lack cellular infrastructure.

Ad hoc networking capability is a core part of most PANs. With Bluetooth, for example, devices self-organize into a piconet with one device acting as the master and the other devices slaved to that master. The master coordinates transmissions among the various devices. WLANs also support ad hoc capability for communication between devices, and also a more formal mesh capability in IEEE 802.11s [61]. Cellular networks are starting to support device-to-device communication where devices can exchange data directly without going through the base station [89, 110]. This is not a completely self-organized ad hoc operation, though, because the devices may coordinate key network operations like device discovery through the base station.

A recent application of mobile ad hoc networking is vehicular ad hoc networking (commonly known as VANETs in the literature) [329]. As illustrated in Figure 1.6, VANETs involve both vehicle-to-vehicle communication and vehicle-to-infrastructure communication and are a key ingredient in connected and automated vehicles. A difference between VANETs and other ad hoc networks is in the overlying applications. Safety is a primary application of VANETs. For example, the dedicated short-range communication protocol [41, 232, 177] allows vehicles to exchange messages with position and velocity information for applications such as forward collision warning. Next-generation connected vehicles will exchange even more information. For example, sharing perceptual data among neighboring vehicles can extend a vehicle’s perception range beyond its visual line of sight [69]. This data can be fused to create a bird’s eye view of neighboring traffic, which can assist both automated and human drivers in difficult driving tasks such as overtaking and lane changing [234]. VANETs, especially at millimeter wave, continue to be an active area of research [337].

Energy networks provide another potential application of wireless communication in the form of sensor networks. The electric power grid is based on hundred-year-old technology where power is sent into the network and consumption is measured by electric meters, which are read infrequently. Sensors can be used to enable what is called a smart grid, which supports features like demand-based pricing and distributed power generation [58, 293]. Many aspects of the smart grid are enabled through wireless meters. Smart grids can be implemented with a host of different wireline or wireless technologies. There are many research challenges in smart grid technology, including control, learning, and system-level issues.

RFID (radio frequency identification) is a special type of communication that is used in applications such as manufacturing, supply chain management, inventory control, personal asset tracking, and telemedicine [361, 183, 62]. An RFID system consists of RFID tags, which are given to products and objects for identification purposes, and RFID readers. Readers broadcast queries to tags in their radio range for information control, and tags reply with stored identification information, typically using energy from the broadcast query to power the RFID circuit and transmitter [64]. Since no active transmission is involved, the power consumption for communication is very low [5]. RFID may be used in a sensor network as both a sensor and a means of communication to detect, for example, if the RFID tag (or the object that is tagged) is physically present in a given location. RFID has been standardized by EPCglobal and the ISO (International Organization for Standardization). The battery-free design of the typical RFID tag makes its design different from that of conventional communication systems.

Modern underwater communication systems use digital transmission [311, 315, 265]. From a signal processing perspective, underwater communication requires the use of sophisticated adaptive receiver techniques [311]. The reason is that, relatively speaking, the underwater propagation channel changes and presents many multipaths. Most radio frequency wireless systems are designed with a kind of block invariance where time variation can be neglected in short processing intervals.

This assumption may not be appropriate for underwater communication due to the rapid channel variations. The main applications of underwater communication are found in the military, for example, ship-to-ship, ship-to-shore, and ship-to-sub, though there are commercial applications in the petroleum industry, such as autonomous underwater vehicles. Communicating underwater is a growth industry for the United States Navy. Two-way underwater digital communication between submarines and the AUTEC (Atlantic Undersea Test and Evaluation Center) range-control station in the Bahamas has been successfully demonstrated [144]. Sensor networks are also applied underwater for oceanographic data collection, environment monitoring, explorations, and tactical surveillance [9]. Many of the concepts developed in this book can be applied to underwater communication systems, with some modifications to account for variability of the propagation channel.

Though signal processing includes digital and analog techniques, DSP dominates most of the application scenarios. Therefore, an analog signal to be processed is discretized and quantized before manipulation. For example, the receiver in a wireless communication system has to apply some processing to the received signal to remove noise, cancel interference, or eliminate the distortion due to the propagation through the wireless channel; at the transmitter side, signal processing is used to generate the waveform to be transmitted and maximize the range or the amount of information per time unit that can be sent. The current trend is to perform all these operations in digital, placing an analog-to-digital converter (ADC) or a digital-to-analog converter (DAC) as close as possible to the receive or transmit antenna respectively. Figure 1.8 shows an example of a basic communication system using a signal processing approach, making use of analog and digital techniques.

Signal processing has many applications in other fields such as:

• Speech and audio, for speaker recognition, text-to-speech conversion, speech recognition, speech or audio compression, noise cancellation, or room equalization.

• Image and video, for image and video compression, noise reduction, image enhancement, features extraction, motion compensation, or tracking of objects.

• Medicine, for monitoring and analysis of biosignals.

• Genomics, for interpretation of genomic information.

• Finance, to analyze financial variables mainly for prediction purposes.

• Radar, for detecting targets and estimating their position and velocity.

Signal processing is a discipline at the intersection of signal processing and applied mathematics. It did not emerge as an independent field of study until the mid-twentieth century [239]. By that time Norbert Wiener had proposed a random process model for the information source. He also invented the Wiener filter, which provides a statistical estimate of an unknown process from an observed noisy process. The landmark paper “A Mathematical Theory of Communication,” written by Claude Shannon in 1948 [302], established the foundations of communication theory by analyzing a basic digital communication system from a signal processing perspective, using Wiener’s idea to model information signals. The sampling theorem proposed by Harry Nyquist in 1928 and proved by Shannon in 1949 in his paper “Communication in the Presence of Noise” addressed the problem of sampling and reconstruction of continuous signals, a milestone in DSP. For subsequent years, however, analog signal processing continued to dominate signal processing applications, from radar signal processing to audio engineering [239]. The publication in 1965 by Cooley and Tukey of an algorithm for the fast implementation of the Fourier transform (now known as FFT) led to the explosion of DSP, making it possible to implement convolution much more efficiently. Speech coding for telephone transmission was at that time a very active signal processing area, which started to benefit from adaptive algorithms and contributed to the success of DSP. Since that time, DSP algorithms have continued to evolve, leading to better performance and the expansion of the range of applications that benefit from them. Wireless communication is not an exception; the incredible increase in performance and data rates experienced in recent years in many communication systems was made possible by the increased complexity of DSP techniques.

A signal processing approach tackles problems from a system perspective, including models for the input and output signals at every block in the system. The different blocks represent the different processing stages, which can be realized with an analog device or a numerical algorithm implemented in a digital processor, as can be seen in Figure 1.8. There exists a trade-off between the complexity and the performance of the models used for the signals and the analog components of the system: more accurate models provide an excellent tool for the simulation and practical evaluation of the system, but they increase complexity and simulation time and make the theoretical analysis of the problems difficult. Statistical characterization of signals using random process theory and probability provides useful models for the signal carrying the information and also for the noise and the interfering signals that appear in a wireless communication system.

Signal processing theory also provides mathematical tools to relate the different signals in a system, using concepts from calculus, linear algebra, and statistics. Chapter 3 reviews in detail the fundamental signal processing results that can be used in the design and analysis of wireless communication systems. Linear time-invariant systems are used extensively in wireless communication to model different devices in the system such as filters or equalizers. Many of the features of a communication system are better understood in the frequency domain, so Fourier analysis is also a basic tool for wireless engineers. Digital communication systems leverage multirate theory results as well, since multirate filters lead to efficient implementations of many of the operations usually performed in a digital transmitter or receiver. Finally, fundamental results in linear algebra are the basis for many signal processing algorithms used for different tasks at the receiver such as channel equalization.

A digital signal processing approach to wireless communications, the so-called software-defined radio (SDR) concept, makes sense for many reasons, such as ease of reconfigurability (software download) or simultaneous reception of different channels and standards, as shown in Figure 1.9. Digitizing the communication signal at the output of the receive antenna may not be feasible, however, because of technical (a very high sampling frequency) or cost (too-high power consumption at the ADC) reasons. Therefore, a trade-off between analog signal processing and DSP is usually found in practical communication systems, which usually include an analog stage to downconvert the signal followed by a digital stage, as illustrated in Figure 1.9. Later chapters of this book provide several examples of functional block diagrams corresponding to current communication systems that make use of this approach.

There are already a number of textbooks on related topics of wireless communication and digital communication. Most other textbooks on wireless communication are targeted toward graduate students in communications, building on the foundations of graduate courses in random processes and digital communication. Unfortunately, undergraduate students, graduate students in other areas, and practicing engineers may not have taken the typical graduate prerequisites for those textbooks. Other textbooks on digital communication are targeted toward one- or two-semester graduate courses, attempting to present digital communication in its most general form. This book, however, focuses on a subset of digital communication known as complex pulse-amplitude modulation, which is used in most commercial wireless systems. In addition, this book describes in detail important receiver signal processing algorithms, which are required to implement a wireless communication link. While most concepts are presented for a communication system with a single transmit and single receive antenna, they are extended at the end of the book to MIMO communication systems, which are now widely deployed in practice.

For communications engineers, this book provides background on receiver algorithms like channel estimation and synchronization, which are often not explained in detail in other textbooks. It also provides an accessible introduction to the principles of MIMO communication. For signal processing engineers, this book explains how to view a communication link through a signal processing lens. In particular, input-output relationships are built upon principles from digital signal processing so that the entire system can be represented in terms of discrete-time signals. Critical background on communication system impairments and their models is provided, along with an approachable introduction to the principles of wireless channel modeling. For analog, mixed-signal, and circuit designers, this book provides an introduction into the mathematical principles of wireless digital communication. The formulations are simplified from what is found in other textbooks, yet what is presented is immediately practical and can be used to prototype a wireless communication link [147].

Coverage in this book is intentionally narrow. No attempt is made to present a framework that includes every possible kind of digital communication. The focus is specifically on complex pulse-amplitude modulated systems. Nor is any attempt made to provide optimum receiver signal processing algorithms for all different channel impairments. Rather the emphasis is on using simpler estimators like linear least squares, which work in practice. The foundations developed in this book are a great platform for further work in wireless communication.

This chapter serves as an introduction to wireless communication, providing a detailed overview of the myriad applications. It also provides some historical background on signal processing and makes the case for using signal processing to understand wireless communications.

Chapter 2 provides an overview of digital communication. The review is built around a canonical block diagram for a digital communication system to provide context for developments in subsequent chapters. Then the components of that diagram are discussed in more detail. First the types of distortion introduced by the wireless channel are reviewed, including additive noise, interference, path loss, and multipath. The presence of the wireless channel introduces many challenges in the receiver signal processing. Then a brief overview is provided of source coding and decoding, with examples of lossless and lossy coding. Source coding compresses data, reducing the number of bits that need to be sent. Next, some background is provided on secret-key and public-key encryption, which is used to keep wireless links secure from eavesdroppers. Then channel coding and decoding are reviewed. Channel coding inserts structured redundancy that can be exploited by the decoder to correct errors. The chapter concludes with an introduction to modulation and demodulation, including baseband and passband concepts, along with a preview of the impact of different channel impairments. Subsequent chapters in the book focus on modulation and demodulation, correcting channel impairments, modeling the channel, and extending the exposition to multiple antennas.

Chapter 3 provides a review of signal processing fundamentals, which are leveraged in subsequent parts of the book. It starts with an introduction to the relevant continuous-time and discrete-time signal notation, along with background on linear time-invariant systems, the impulse response, and convolution. Linear time-invariant systems are used to model multipath wireless channels. The chapter continues with a review of several important concepts related to probability and random processes, including stationarity, ergodicity, and Gaussian random processes. Next, some background is provided on the Fourier transform in both continuous and discrete time, as well as on signal power and bandwidth, as it is useful to view communication signals in both the time and frequency domains. The chapter continues with derivation of the complex baseband signal representation and complex baseband equivalent channel, both of which are used to abstract out the carrier frequency of a communication signal. It then provides a review of some multirate signal processing concepts, which can be developed for efficient digital implementation of pulse shaping. The chapter concludes with background on critical concepts from linear algebra, especially the least squares solution to linear equations.

Chapter 4 introduces the main principles of complex pulse-amplitude modulation. The main features of the modulation are provided, including symbol mapping, constellations, and the modulated signal’s bandwidth. Then the most basic impairment of additive white Gaussian noise is introduced. To minimize the effects of additive noise, the optimal pulse-shaping design problem is formulated and solved by Nyquist pulse shapes. Assuming such pulse shapes are used, then the maximum likelihood symbol detector is derived and the probability of symbol error is analyzed. The topics in this chapter form a basic introduction to digital communication using pulse-amplitude modulation with perfect synchronization and only the most basic impairment of additive noise.

Chapter 5 describes other impairments introduced in wireless communication. It starts with an overview of symbol synchronization and frame synchronization for flat-fading channels. This involves knowing when to sample and the location of the beginning of a frame of data. It then proceeds to present a linear time-invariant model for the effects of multiple propagation paths called frequency selectivity. Several mitigation strategies are described, including linear equalization. Because the distortion introduced by the frequency-selective channel varies over time, the chapter then describes approaches for channel estimation. The channel estimate is used to compute the coefficients of the equalizer. Alternative modulation strategies that facilitate equalization are then introduced: single-carrier frequency-domain equalization (SC-FDE) and OFDM. Specific channel estimation and carrier frequency offset correction algorithms are then developed for single-carrier and OFDM systems. Most of the algorithms in this chapter are developed by formulating a linear system and taking the least squares solution. The chapter concludes with an introduction to propagation and fading-channel models. These statistical models are widely used in the design and analysis of wireless systems. A review is provided of both large-scale models, capturing channel variations over hundreds of wavelengths, and small-scale models, incorporating variations over fractions of a wavelength. Ways to quantify frequency selectivity and time selectivity are introduced. The chapter concludes with a description of common small-scale fading-channel models for both flat and frequency-selective channels.

Chapter 6 concludes the book with a concise introduction to MIMO communication. Essentially, the key concepts developed in this book are re-examined, assuming a plurality of transmit and/or receive antennas. Most of the development is built around flat-fading channels, with extensions to frequency selectivity through MIMO-OFDM provided at the end. The chapter starts with an introduction to the different configurations of multiple antennas in SIMO (single input multiple output), MISO (multiple input single output), and MIMO configurations. It then describes the basics of receiver diversity for SIMO systems, including antenna selection and maximum ratio combining, including their impact on the probability of vector symbol error. Next it explains some approaches for extracting diversity in MISO communication systems, including beamforming, limited feedback, and space-time coding. The chapter then introduces the important MIMO technique known as spatial multiplexing. Extensions to precoding, limited feedback, and channel estimation are also described. The chapter concludes with an overview of MIMO-OFDM, which combines MIMO spatial multiplexing with the ease of equalization in OFDM systems. Key ideas like equalization, precoding, channel estimation, and synchronization are revisited in this challenging setting of MIMO with frequency-selective channels. MIMO and MIMO-OFDM are used in many commercial wireless systems.

The concepts developed in this book are ideally suited for practical implementation in software-defined radio. The author has developed a companion laboratory manual [147], which is sold as part of a package with the National Instruments Universal Software Radio Peripheral. That laboratory manual features seven experiments that cover the main topics in Chapter 4 and Chapter 5, along with a bonus experiment that explores the benefits of error control coding. Of course, the concepts can be demonstrated in practice in other ways, even using a speaker as a transmit antenna and a microphone as a receive antenna. Readers are encouraged to simulate algorithms when possible, along with working example and homework problems.

• Most major wireless communication systems use digital communication. Advantages of digital over analog include its suitability for use with digital data, robustness to noise, ability to more easily support multiple data rates and multiple users, and easier implementation of security.

• Digital signal processing is well matched with digital communication. Digital signal processing makes use of high-quality reproducible digital components. It also leverages Moore’s law, which leads to more computation and reduced power consumption and cost.

• This book presents the fundamentals of wireless digital communication as seen through a signal processing lens. It focuses on complex pulse-amplitude modulation and the most common challenges faced when implementing a wireless receiver: additive noise, frequency-selective channels, symbol synchronization, frame synchronization, and carrier frequency offset synchronization.

(a) Choose a current cell phone from three of these manufacturers: Nokia, Samsung, Apple, LG, Huawei, Sony, Blackberry, Motorola, or another of your choosing. Describe the wireless and cellular technologies and the frequency bands supported by each one.

(b) Name at least three mobile service providers in your country. Which cellular technologies are currently supported by the networks?

(c) Which of those three mobile service providers charge for data and what are the charges for a typical consumer plan (not business)? Why do you think some providers have stopped offering unlimited data plans?

2. Wireless Device Comparison Fill in the following table for three cellular devices manufactured by the three companies in the table:

3. Visible Light Communication (VLC) Do some research on VLC, an alternative to wireless communication using RF (radio frequency) signals. This is a topic that we do not cover much in the book, but the topics from the book can be used to understand the basic principles. Be sure to cite your sources in the answers below. Note: You should find some trustworthy sources that you can reference (e.g., there may be mistakes in the Wikipedia article or it may be incomplete).

(a) Which part of the IEEE 802 LAN/MAN Standards Committee deals with VLC?

(b) What is the concept of VLC?

(c) What is the bandwidth of a typical VLC application?

(d) Explain how VLC might be used for secure point-to-point communication.

(e) Explain how VLC might be used for indoor location-based services.

(f) Explain why VLC might be preferred on aircraft for multimedia delivery.

(g) Explain how VLC could be used in intelligent transportation systems.

4. Sensor Networking There are many kinds of wireless networks, such as wireless sensor networks, that have important applications in manufacturing. They are often classified as low-rate wireless personal area networks. This is a topic that we do not cover much in the book, but the topics from the book can be used to understand the basic principles. Be sure to cite your sources in the answers below. Note: You should find some trustworthy sources that you can reference (e.g., there may be mistakes in the Wikipedia article or it may be incomplete).

(a) What is a wireless sensor network?

(b) What is IEEE 802.15.4?

(c) What is ZigBee?

(d) How are IEEE 802.15.4 and ZigBee related?

(e) What communication bands are supported by IEEE 802.15.4 in the United States?

(f) What is the bandwidth of the communication channel specified by IEEE 802.15.4? Note: This is bandwidth in hertz, not the data rate.

(g) What is the typical range of an IEEE 802.15.4 device?

(h) How long should the battery last in an IEEE 802.15.4 device?

(i) How are sensor networks being used to monitor highway bridges?

5. Wireless and Intellectual Property The wireless industry has been plagued with lawsuits over intellectual property. Identify a recent case of interest and describe the parties and their positions. Then describe in at least half a page your opinion about the role of intellectual property in wireless communications.