Introduction

Looking at classical communication theory, we see that it mainly deals with point-to-point links disturbed by thermal (Gaussian) noise. More recently the challenges of mobile communication have introduced features such as adverse, time-varying propagation conditions which create channel variations that are difficult to predict. Radio systems, as we find them in reality, have to cope with additional problems. Maybe the most characteristic feature of modern radio communication is that virtually no radio link or radio system operates in isolation, and is thus never alone in its allocated frequency band. Other radio transmitters, near and far, constantly cause interference. Interference is in many cases the limiting factor to the performance of the system. Since the days of Marconi, the proliferation of wireless communications has caused a tremendous increase in the utilization of the frequency spectrum. A key problem area, as was already noted in Chapter 1, is how to manage the frequency spectrum to avoid, or at least minimize, the adverse effects of interference. Can interference be avoided completely, or are there efficient methods to minimize the performance degradation? The ether, where we transmit our signals, is, whether intended or not, a broadcast medium. In some geographical regions, a large number of wireless networks have to coexist as illustrated in Figure 2.1. The blessing of wireless communication is that it allows for quickly establishing arbitrary new connections between a large number of users. In Figure 2.1, we consider three transmitters transmitting information to three different receivers indicated by the solid black arrows. We call these paths the active communication links. As the radio spectrum is shared by all users the transmissions of the three transmitters in Figure 2.1 give rise to interference. These unwanted cross-links, the interference links, are indicated by the light gray arrows in the figure.

The properties of the interference will depend on the waveforms and transmitter powers selected by the interfering transmitters as well as the propagation conditions on the cross-links. The impact on the performance of the active communication link will depend not only on the waveforms, the powers and the propagation conditions in the active link but also on the performance of the radio receiver, e.g. how good the receiver is at suppressing the unwanted signals. The performance experienced by the user in this network will depend on the type of service that is provided.

Historical perspective on radio resource management

As J. C. Maxwell had predicted in the 1850s, wireless transmission of electrical energy was feasible. Several decades later Heinrich Hertz managed to experimentally verify Maxwell's daring ideas with his award-winning experiment in 1888. He was able to demonstrate that his 600 MHz transmitter was capable of producing a spark in his simple receiver a few meters away in his laboratory. Although several scientists and inventors would like to claim the fame of inventing radio as we know it today, it took an engineer to bring this groundbreaking research into practical use. The Italian pioneer Guglielmo Marconi was the first to make practical and commercial use of the so-called Hertzian waves. After some initial experiments on his father's estate in 1895, his wireless apparatus gradually became a commercial success. It eventually made Marconi the first, but certainly not the last, millionaire in the wireless business. From humble beginnings, transmitting messages a few hundred meters in his first experiments, in 1901 he was finally able to demonstrate wireless communication across the Atlantic Ocean from Poldhu in Cornwall, England, to Newfoundland, Canada. In the decades to follow, wireless communications became an essential technology onboard ships. The early 1920s saw the advent of radio broadcasting, bringing wireless receivers into every home. We know what happened later—wireless has created a deep impact in our daily lives through success stories such as TV broadcasting, worldwide shortwave communication, satellite communications, and in recent decades mobile telephony and wireless and mobile Internet access.

The latest chapter in this story started to be written in the early 1980s with the commercial success of automated mobile telephony and mobile data. Examples of so-called first generation mobile telephone systems are the NMT system in Scandinavia (1981), AMPS in the USA (1984), TACS in the UK (1984) and other systems. These systems were targeting limited markets, terminals were expensive and they never reached very high user penetrations. The first-generation systems were analog designs—only the switching logic relied on digital technology.

It took another decade and the introduction of global standards for digital mobile systems to put a cellular phone in almost every person's hands.

Overview

In wireless networks, multiple terminals need to communicate at the same time and a medium access control (MAC) protocol allows several terminals to transmit over the wireless channel and to share its capacity.MAC protocols multiplex several data streams of different terminals to share the same channel and deal with issues such as addressing, how a terminal obtains a channel when it needs one, and so forth.

The design of MAC protocols closely relates to the condition of the physical channels. Initially MAC protocols were designed for wired communications where multiple computers need to transmit data packets at the same time in a local area network (LAN). With wired networks, the physical medium can be copper or fiber optics, which are in general very reliable with abundant bandwidth. Packet loss in wired networks is mainly due to collisions and the MAC designs are relatively simple.

The MAC design in wireless networks is much more challenging. The difficulties lie in the following aspects. With wireless communications, a radio signal may experience reflection, diffraction or scattering before reaching its receiver. Any of them will deteriorate the signal and incur variation of signal quality in time, frequency and space. Another main issue is the broadcast nature of wireless channels. For reliable transmission against fading, strong radio transmission power needs to be used by the transmitter. This incurs strong interference with other terminals in the vicinity. The stronger the transmission power or the closer the neighboring terminals, the stronger the interference will be. Because of fading and interference, wireless networks are more vulnerable compared to wired ones. Usually the bit error rate of wired networks is better than 10−6 and that of wireless ones is worse than 10−3. The difficulty also lies in the fact that wireless terminals usually have to operate in half-duplex mode. This is because transmission power is in general much stronger than reception power, and with full-duplex operation the leakage of transmission power to the receiver component will incur very strong self-interference and therefore the terminal will not be able to receive packets or detect a collision when it is sending.

MAC schemes can be divided into two categories, contention-free and contention-based protocols. A contention-free MAC protocol requires a central controller to coordinate the resource allocation and the central controller can be a base station in a cellular network or an access point in a wireless local area network.

Radio spectrum is a finite resource, and therefore has to be shared by multiple users. Such sharing or reuse of radio spectrum is a dominant feature of present-day wireless systems where a tremendous number of devices have wireless connectivity. If you use mobile Internet, particularly in a city, it is highly likely that the frequency channel assigned to you is also used by someone very near you. This sharing of radio spectrum inevitably causes interference between users due to the broadcast nature of the wireless medium. Managing the interference between the wireless links is one of the most important problems in wireless networks.

We have already discussed some basics of interference management in Chapter 5. Let's briefly recall the cellular principle here. In a cellular system, the whole service area is divided into cells, and each cell is covered by a base station or an access point. The mobile terminals in the same cell are exempt from the interference problem because they are given orthogonal resources, e.g., frequency channel, time, spreading code, or waveform. However, interference still takes place between the neighboring cells. Thus, what matters in the cellular systems is inter-cell interference. Frequency planning is applied to the cells so that the proper reuse distance can be maintained between the cells of the same frequency channel assignment.

The static frequency reuse presented in Chapter 5 is of course not the only method of managing interference in practical wireless systems. In fact, immense research work has been performed and is still ongoing in the field of inter-cell interference management. In this chapter, we will go deeper into this subject. First, the categories and elements of interference management techniques will be introduced in Section 7.1. Then, in the subsequent sections, the three categories of interference management techniques, namely interference avoidance, interference randomization and interference cancellation, will be further discussed with examples. Finally, interference management for small cells and heterogeneous networks is presented in Section 7.5. Note that interference management is a vast research area. This chapter only scratches the surface of the field.

Classification of interference management techniques

We will give a brief overview of interference management before looking into specific techniques.

Communication infrastructures

Most of this book has covered how to provide effective wireless access services and the efficient utilization of the spectrum resources. In the previous chapters we have also looked at energy aspects. Now we will widen the scope even more as we look at the total resource consumption in a wireless access infrastructure. This will include networks, switches/routers and access ports and the terminals, as described in Chapters 2 and 9. In this chapter, the tradeoff between the resource consumption incurred by adding more access points, i.e. a more expensive infrastructure, and increased capacity and a higher QoS provided to the users, will be reviewed in more detail. We will make comparisons among such diverse quantities as frequency spectrum allocation, equipment, physical infrastructure including towers and antennas, real estate, power consumption, user equipment and maintenance. Such comparisons are naturally made in monetary or economic terms. The total cost of running a wireless access system comprises the cost of all these individual components. In economic terms, this is usually referred to as the supply side of operating a network. This cost of providing (supplying) the network has to be compared with revenues that the system provider, for example an operator, can derive from the users, i.e. what the users are willing to pay for the service. The “will” in turn is based on user demand and user satisfaction. Supply and demand are tightly coupled and together determine if operating the network will be a profitable endeavor, i.e. that the revenue will exceed the investment.

Although not always easy, it is usually possible to compute the cost of network operation, as we will see later in the chapter. However, understanding the demand and willingness to pay is difficult to model and will depend on how the services are provided and which actors are providing them. The telecommunications industry has been a global one for over a century, with many actors. Whereas previous chapters have outlined the supply chain of actual communication (transport) services, the picture has been more complicated in the past as the actual content transmitted and the services that have been provided by and through the network have been interwoven with the network service itself. Figure 11.1 gives a rough description of the actors on the scene and their interrelations.

The world has seen astonishing developments in wireless communications. From the early days when wireless was seen as a new and complex technology that required skilled operators to work, to a situation where wireless has become a truly pervasive technology with devices in everyone's pocket. Voice communications, including mobile telephony, have dominated the first century of wireless communications. The technical challenges have been dominated by the struggle of the engineer against nature—how to facilitate communications over long distances and how to overcome adverse radio propagation conditions. With the advent of digital communications, we have over recent decades seen marvelous advances in this area, with technologies such as error control coding, digital signal processing, advanced antenna technologies and others. Meanwhile, the number of wireless users has skyrocketed. In addition we now witness wireless Internet access becoming a dominant technology for all kinds of IT services. A necessary prerequisite for this development is that wireless access is abundant and becomes almost free. The consequence is that data rates in wireless communications have increased dramatically during the last decade. The industry predicts an exponential increase of data traffic that would correspond to a 1000-fold increase in traffic between 2010 and 2020. It has become obvious that traditional measures for increasing data rates in the wireless links, e.g. coding and signal processing, are not going to save the day since these techniques now operate close to their theoretical limits, regardless of their complexity. Instead, much of the focus of the engineering work has shifted to what can be seen as the social struggle for scarce resources. The proper management of resources such as frequency spectrum, energy consumption, and to a large extent monetary investments in infrastructure (base stations and the like) is now a key issue. The design objectives have changed from “how can we provide high quality communications in a single radio link?” to “how can we create sustainable systems that provide affordable high quality wireless communications for billions of users?” The latter question is mainly one of Radio Resource Management (RRM), which is the main theme of this book. The book approaches this problem in the following way.

Introduction

While semiconductor processing speed has been increasing exponentially, doubling almost every two years according to Moore's law, processor power consumption also continues to grow by 150% every two years [K. Lahiri et al., 2002]. By contrast, advances in battery technology have not kept pace, with capacity increasing at a modest rate of 10% every two years [K. Lahiri et al., 2002]. This leads to an increasingly large gap between power thirst and battery capacity. Information and communication technology (ICT) plays an important role in global greenhouse gas emissions since the amount of energy consumed by ICT is increasing dramatically to meet rapidly growing broadband mobile service requirements. For example, the power consumption for a macro base station can be 1400 watts and the corresponding energy costs can reach $3200 per annum with 11 tons of CO2 emissions. It has been shown that nowadays the total energy used by the infrastructure of cellular networks, wired networks and Internet takes up more than 2% of worldwide electrical energy consumption [GeSI, 2008]. The radio network itself adds up to 80% of an operator's entire energy consumption [EE Times, 2007]. In addition, this amount of energy is expected to increase rapidly in the coming years. Energy efficiency, therefore, is increasingly important for wireless mobile communications.

In this chapter we introduce some basic energy-efficient communication technologies. We start by studying node-level energy-efficient design, as improvements at the wireless node for energy-efficient radio transmission will translate into savings for the entire network. For an individual pair of wireless transceivers, the relation between power consumption, channel fading, path loss, modulation, coding, data rate and implementation factors are discussed thoroughly. To be specific, we will first analyze the energy consumption of different components of wireless transmitters. Then we will introduce the link energy efficiency metric that characterizes how efficiently energy is used in communication systems. Based on the energy efficiency metric, we will introduce how a communication pair can be designed in the most energy-efficient way. First, we will consider only radio transmission power consumption and show how the transmitter can be designed optimally to achieve the maximum energy efficiency. In practice, electronic circuits also consume a certain amount of operating power and this will significantly change the design of energy-efficient transmitters. We will study how this electronic circuit power consumption affects energy-efficient transmission.