103,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Serie: Wiley - IEEE
- Sprache: Englisch
Massive Connectivity Learn to support more devices and sensors in Internet of Things applications through NOMA and machine-type communication Non-orthogonal multiple access (NOMA) has held much interest due to its ability to provide a higher spectral efficiency--such as more bits per unit bandwidth in Hertz--than other, orthogonal multiple access schemes. The majority of this research focuses on the application of NOMA to downlink channels (from base station to users) in cellular systems as its use for uplink (users to base station) is somewhat circumscribed. However, NOMA has recently been employed in contention-based uplink access, which has shown an improvement in performance that allows an increase in the number of users that can be supported. As a result, NOMA is promising for machine-type communication (MTC) in 5G systems and beyond, making it a key enabler of the Internet of Things (IoT). Massive Connectivity provides an in-depth, comprehensive view of the benefits and drawbacks of uplink NOMA random access (RA) systems. This text offers a basic introduction and description of uplink NOMA RA systems before considering the possibilities for evolution of the scheme as attempts are made to derive the most benefits and overcome any weaknesses. The book further presents key performance analysis while also highlighting game-theoretic views. In essence, by describing the essential properties of stable and high-throughput yielding RA systems, the book demonstrates that uplink NOMA can fulfill these required properties. Massive Connectivity readers will also find: * An extensive literature survey on RA systems and their applications since the 1970s * Recent advances in random access for massive connectivity * Retransmission control algorithms for NOMA random access * Discussion of how uplink NOMA random access systems can be integrated into the existing long-term evolution (LTE) or upcoming 5G cellular networks Massive Connectivity is a useful reference for field engineers and academics, as well as experts for random access systems for IoT applications.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 249
Veröffentlichungsjahr: 2022
Ähnliche
Table of Contents
Cover
Title Page
Copyright
Dedication
Author Biography
Preface
1 Introduction
1.1 Machine‐Type Communication
1.2 Non‐orthogonal Multiple Access
1.3 NOMA for MTC
1.4 An Overview of Probability and Random Processes
2 Single‐User and Multiuser Systems
2.1 A Single‐User System
2.2 Multiuser Systems
2.3 Further Reading
Note
3 OMA and NOMA
3.1 Orthogonal Multiple Access
3.2 Non‐Orthogonal Multiple Access
3.3 Power and Rate Allocation
3.4 Code Division Multiple Access
3.5 Further Reading
4 Random Access Systems
4.1 ALOHA Systems
4.2 Throughput Analysis
4.3 Analysis with a Finite Number of Users
4.4 Analysis with an Infinite Number of Users
4.5 Fast Retrial
4.6 Multiuser Detection
4.7 Further Reading
5 NOMA‐Based Random Access
5.1 NOMA to Random Access
5.2 Multichannel ALOHA with NOMA
5.3 Opportunistic NOMA
5.4 NOMA‐Based Random Access with Multiuser Detection
5.5 Further Reading
Note
6 Application of NOMA to MTC in 5G
6.1 Machine‐Type Communication
6.2 A Model with Massive MIMO
6.3 NOMA for High‐Throughput MTC
6.4 Layered Preambles for Heterogeneous Devices
6.5 Further Reading
Notes
7 Game‐Theoretic Perspective of NOMA‐Based Random Access
7.1 Background of Game Theory
7.2 Random Access Game
7.3 NOMA‐ALOHA Game
7.4 Fictitious Play
7.5 Evolutionary Game Theory and Its Application
7.6 Further Reading
Notes
Bibliography
Index
End User License Agreement
List of Tables
Chapter 5
Table 5.1 The transmit power for each region and the predetermined power lev...
Chapter 6
Table 6.1 Absolute values of cross‐correlation of layered preambles,
and
Chapter 7
Table 7.1 Bimatrix of matching pennies game (Players 1 and 2 are referred to...
Table 7.2 Bimatrix of two‐person random access game.
Table 7.3 Bimatrix of two‐person NOMA‐ALOHA game.
Table 7.4 The other player's action based on the feedback from BS.
List of Illustrations
Chapter 2
Figure 2.1 Signal space diagram for QPSK.
Figure 2.2 Signal space diagram for
signal vectors or codewords of length
Figure 2.3 Sphere packing to find the capacity of AWGN channel.
Figure 2.4 Shannon capacity of AWGN channel in terms of SNR.
Figure 2.5 Downlink and uplink transmissions in a cellular system consisting...
Figure 2.6 Capacity region for two‐user broadcast channel.
Figure 2.7 Capacity region for two‐user multiple access channel.
Chapter 3
Figure 3.1 An illustration of TDMA with
users.
Figure 3.2 Achievable rates of two‐user TDMA in downlink and uplink.
Figure 3.3 An illustration of FDMA with
users.
Figure 3.4 A few subcarriers in the time domain when
.
Figure 3.5 Channel allocation for two users in multiple access schemes: (a) ...
Figure 3.6 A signal constellation of the superposition of two QPSK signals....
Figure 3.7 Opportunistic access for multiuser diversity in uplink communicat...
Figure 3.8 Outage probability (from simulations) and its upper‐bound in (3.5...
Figure 3.9 Surfaces of the minimum total transmission power,
, of OMA (the ...
Figure 3.10 Minimum total transmission powers,
, of OMA and NOMA: (a)
for...
Figure 3.11 Minimum total transmission powers,
, of OMA and NOMA for variou...
Chapter 4
Figure 4.1 An illustration of uncoordinated transmissions by four users.
Figure 4.2 An illustration of S‐ALOHA with two users.
Figure 4.3 Maximum throughput of S‐ALOHA as a function of
with
when the ...
Figure 4.4 Throughput curves of three different ALOHA protocols as functions...
Figure 4.5 Two curves as functions of state
,
and
, when
,
, and
.
Figure 4.6 Two functions for the drift for a large
with
:
and
.
Figure 4.7 The average number of backlogged packets,
, by taking the averag...
Figure 4.8 An illustration of fast retrial for multichannel ALOHA with four ...
Figure 4.9 Throughput curves of CRA and multichannel ALOHA as functions of t...
Figure 4.10 Throughput curves of CRA and multichannel ALOHA as functions of
Chapter 5
Figure 5.1 NOMA‐ALOHA with two active users of different power levels.
Figure 5.2 Throughput curves of S‐ALOHA and NOMA‐ALOHA protocols as function...
Figure 5.3 Performance of NOMA for different numbers of power levels,
: (a)...
Figure 5.4 Throughput of NOMA‐ALOHA when
.
Figure 5.5 Throughput of NOMA‐ALOHA for different number of subchannels,
, ...
Figure 5.6 Throughput for different values of
when
,
, and
.
Figure 5.7 Average transmission power for different values of
when
,
,
Figure 5.8 Average transmission power for different numbers
of subchannels...
Figure 5.9 An illustration of the system model with
RBs to support both SD...
Figure 5.10 Throughput of the conventional approach per channel as a functio...
Figure 5.11 Throughput of the NOMA‐based approach per channel as a function ...
Figure 5.12 Performance of the conventional approach and the NOMA‐based appr...
Figure 5.13 Performance of the conventional random access approach and the N...
Figure 5.14 Performance of the NOMA‐based approach in terms of throughput fo...
Figure 5.15 The normalized transmit power in dB of the NOMA‐based approach i...
Figure 5.16 Performance of the NOMA‐based approach for various values of
w...
Figure 5.17 An illustration of layered CRA based on power‐domain NOMA.
Figure 5.18 An illustration of dividing a cell into
regions to reduce the ...
Figure 5.19 Average numbers of FA/MD errors when
and
(thus,
users per ...
Chapter 6
Figure 6.1 An illustration of four‐step random access in MTC.
Figure 6.2 An illustration of two‐step random access in MTC.
Figure 6.3 Two phases (i.e. preamble transmission and data transmission phas...
Figure 6.4 A multi‐drawer cabinet as a received signal in a massive MIMO sys...
Figure 6.5 An illustration of the CoPD approach where simultaneous preamble ...
Figure 6.6 Throughput comparison between conventional and CoPD approaches wi...
Figure 6.7 Two sets of preambles for type‐1 and type‐2 devices with differen...
Figure 6.8
for various values of
to keep
.
Figure 6.9 Probabilities of MD of active type‐2 devices with/without error p...
Figure 6.10 Probabilities of MD of active type‐2 devices with/without error ...
Chapter 7
Figure 7.1 The mixed strategy NE,
, for different values of the reward of s...
Figure 7.2 The mixed strategy that maximizes the average payoff for differen...
Figure 7.3 The maximum average payoff and the average payoff of the mixed st...
Figure 7.4 The mixed strategy NE,
, for different values of the reward of s...
Figure 7.5 The mixed strategy NE,
, for different numbers of users when
....
Figure 7.6 Trajectory of the empirical probability of two‐person NOMA‐ALOHA ...
Figure 7.7 A trajectory of the state of the replicator equation in (7.67) wi...
Guide
Cover
Table of Contents
Title Page
Copyright
Dedication
Author Biography
Preface
Begin Reading
Bibliography
Index
End User License Agreement
Pages
ii
iii
iv
v
xi
xiii
xiv
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
167
168
169
170
171
173
174
175
176
177
IEEE Press445 Hoes LanePiscataway, NJ 08854
IEEE Press Editorial BoardSarah Spurgeon, Editor in Chief
Jón Atli Benediktsson
Andreas Molisch
Diomidis Spinellis
Anjan Bose
Saeid Nahavandi
Ahmet Murat Tekalp
Adam Drobot
Jeffrey Reed
Peter (Yong) Lian
Thomas Robertazzi
Massive Connectivity
Non‐Orthogonal Multiple Access to High Performance Random Access
Jinho ChoiDeakin UniversityAustralia
Copyright © 2022 by The Institute of Electrical and Electronics Engineers, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New Jersey.Published simultaneously in Canada.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per‐copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750‐8400, fax (978) 750‐4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748‐6011, fax (201) 748‐6008, or online at http://www.wiley.com/go/permission.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762‐2974, outside the United States at (317) 572‐3993 or fax (317) 572‐4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.
Library of Congress Cataloging‐in‐Publication Data:
Names: Choi, Jinho, author.Title: Massive connectivity : non‐orthogonal multiple access to high performance random access / Jinho Choi.Description: Hoboken, New Jersey : Wiley‐IEEE Press, [2022] | Includes bibliographical references and index.Identifiers: LCCN 2022015208 (print) | LCCN 2022015209 (ebook) | ISBN 9781119772774 (cloth) | ISBN 9781119772781 (adobe pdf) | ISBN 9781119772798 (epub)Subjects: LCSH: Wireless communication systems.Classification: LCC TK5103.2 .C456 2022 (print) | LCC TK5103.2 (ebook) | DDC 621.384–dc23/eng/20220609LC record available at https://lccn.loc.gov/2022015208LC ebook record available at https://lccn.loc.gov/2022015209
Cover Design: WileyCover Image: © Fit Ztudio/Shutterstock
To my family
Author Biography
Jinho Choi was born in Seoul, Korea. He received B.E. (magna cum laude) degree in electronics engineering in 1989 from Sogang University, Seoul, and M.S.E. and Ph.D. degrees in electrical engineering from Korea Advanced Institute of Science and Technology (KAIST) in 1991 and 1994, respectively. He is with the School of Information Technology, Burwood, Deakin University, Australia, as a Professor. Prior to joining Deakin in 2018, he was with Swansea University, United Kingdom, as a Professor/Chair in Wireless, and Gwangju Institute of Science and Technology (GIST), Korea, as a Professor. His research interests include the Internet of Things (IoT), wireless communications, and statistical signal processing. He authored two books published by Cambridge University Press in 2006 and 2010. Prof. Choi received a number of best paper awards including the 1999 Best Paper Award for Signal Processing from EURASIP. He is on the list of World's Top 2% Scientists by Stanford University in 2020 and 2021. Currently, he is an Editor of IEEE Wireless Communications Letters and a Division Editor of Journal of Communications and Networks (JCN). He has also served as an Associate Editor or Editor of other journals including IEEE Trans. Communications, IEEE Communications Letters, JCN, IEEE Trans. Vehicular Technology, and ETRI journal.
Preface
Wireless connectivity is an indispensable technology of our lives today. From smartphones to connected vehicles to remote controlled drones, many devices rely on wireless connectivity. In addition, there are a variety of wireless connectivity technologies including WiFi, Zigbee, and cellular systems. These technologies exist all around us and help us a lot in our daily life. They have been developed through various stages and will continue to evolve. Advances in wireless technology have resulted in a variety of new applications. For example, with the invention of radio, mankind began to quickly share information through radio news from the 1920s, and in the twenty‐first century, most of mankind enjoys wireless Internet services.
Wireless communication utilizes the radio spectrum, which is part of the electromagnetic spectrum with frequencies between 30 Hz and 300 GHz, and most wireless systems including cellular use microwave bands (1–100 GHz). For example, WiFi uses 2.4 and 5 GHz and fifth generation (5G) systems use sub‐6 GHz frequency bands as well as millimeter bands (i.e. 24.25 GHz and above). In order to support an ever‐increasing number of users and diverse applications, as technology advances, it is expected to increase the frequency so that a wider bandwidth is available. However, the bandwidth is a key limiting factor and scarce resource. Therefore, multiple access schemes to share a given bandwidth among users are always important to efficiently exploit the limited bandwidth. Orthogonal multiple access (OMA) schemes are currently employed for existing cellular systems. For example, the Global System for Mobile Communications (GSM) system, which is a second generation (2G) system, employs time division multiple access (TDMA) that allocates orthogonal time slots to different users. While OMA schemes are straightforward to be implemented and have been successfully employed in a number of wireless communication systems, there can be quite different approaches that can provide a higher spectral efficiency than them. Interestingly, such schemes had been discussed in the information theory literature since the 1970s under various names such as superposition coding, successive interference cancellation, and so on.
Code division multiple access (CDMA), which has been employed for a 2G system, i.e. Interim Standard 95 (IS‐95), is a multiple access scheme for multiuser communications where multiple users coexist and share the same radio spectrum. In CDMA, coexisting users are differentiated by signature sequences that are not orthogonal. As a result, CDMA can be seen as a non‐orthogonal multiple access (NOMA) and is expected to have higher spectral efficiency than an OMA scheme such as TDMA. In practice, in order to obtain high spectral efficiency in CDMA, an interference canceller, which was considered difficult to implement when IS‐95 was introduced, is required together with precise power control.
Power‐domain NOMA is another NOMA scheme where the power allocation is integrated with successive interference cancellation. Therefore, in power‐domain NOMA, interference cancellation is essential in the receiver, and high spectral efficiency can be expected through this. This makes power‐domain NOMA a strong multiple access technology candidate in the next generation cellular system.
Prior to the Internet‐of‐Things (IoT), it is not an exaggeration to say that most wireless connectivity technologies except for telematics were developed for human‐type communication (HTC) services. As IoT applications become more popular, we are witnessing growth in machine‐type communication (MTC) used in supporting the connectivity of numerous sensors and devices. MTC is now part of cellular systems such as 5G, and expected to play a more important role in next generation systems. As the number of things such as sensors and devices in MTC continues to increase, we expect to face various challenges due to limited spectrum. Like HTC, power‐domain NOMA can be applied to MTC so that more devices can be supported with limited spectrum, which is the main topic of this book.
After discussing the well‐known models for single‐user and multiuser systems, this book discusses the details and key differences between OMA and NOMA. We then describe key principles of random access and extend random access to NOMA‐based random access. We will also discuss how NOMA can be applied to MTC and take a closer look at how NOMA can improve the performance of MTC. After explaining NOMA from a communication perspective on their applications to MTC, at the end of this book, we will show how we can interpret these NOMA‐based random access systems using game theory.
Although NOMA was not a new notion at all, I remember that there were skeptical views in the early studies of applying NOMA to wireless communication systems, which may be due to fear to the unknown or unconventional. I sincerely hope that these skeptical views will be turned into positive ones during the last decade of active research on NOMA, and would like to thank numerous researchers who have contributed to NOMA.
Finally, I would like to offer very special thanks to my wife, Kila, and two children, Seji and Wooji, for their understanding and love.
Jinho Choi
Melbourne, AustraliaMarch, 2022
1Introduction
Two main topics are covered in this book. One is machine‐type communication (MTC) and the other is non‐orthogonal multiple access (NOMA). Each topic has its own foundations and applications. In this chapter, we briefly explain each of them, and then explain why both topics should be covered in this book.
1.1 Machine‐Type Communication
It may not be easy to imagine our daily life and business without the Internet although it began to appear as a backbone network in the 1970s to interconnect a small number of academic and military networks. The Internet is a network of networks and allows to exchange information between servers, computers, mobile phones, and so on restlessly. The Internet‐of‐Things (IoTs) is a natural extension of the Internet as machines, devices, and sensors are connected to the Internet to exchange information without human intervention in a number of applications such as smart factory.
As the number of devices connected to the Internet grows, their connectivity becomes important. Private and public networks can be used for their connectivity. For example, for smart home applications, a private network can be used at home to allow a small number of devices to be connected. For smart city applications, a large‐scale public network would be preferable. Thus, the deployment of IoT networks depend on applications.
MTC is to support communications between machines or devices without human intervention. Unlike human‐type communication (HTC), MTC mainly focuses on uplink transmission rather than downlink transmission (this is one of the main differences between MTC and HTC, where it can be seen that MTC's design principles must be different from those of HTC) and will support sporadic traffic in the form of short data packets. As a result, in order to keep signaling overhead low, the random access channel (RACH) procedure is used for MTC in Long‐Term Evolution (LTE) systems. In the fifth generation (5G) system, a new random access scheme consisting of two steps, which is more efficient than the RACH procedure in LTE consisting of four steps, has been standardized.
MTC can provide connectivity for a large number of devices in a cell, paving the way for IoT applications to interact with devices deployed over a large area via cellular systems. This means that MTC becomes essential in various IoT applications such as smart cities and intelligent transport systems.
Furthermore, the global number of connected devices is expected to exceed 500 billion by 2030, while the human population is predicted to be 8.5 billion by the United Nations (UN). This means that IoT devices will outnumber human population by approximately 60‐folds in 2030, and these devices will be used in a variety of IoT applications requiring heterogeneous connectivity demand. As a result, MTC will play a more prominent role in 5G and beyond (i.e. the sixth generation (6G)) and thus new MTC schemes need to be developed to meet the diverse requirements for future IoT applications.
1.2 Non‐orthogonal Multiple Access
Various multiple access schemes have been used to support multiple users in a cellular system. For example, time division multiple access (TDMA) is adopted in the global system for mobile communications (GSM), which is regarded as a second generation (2G) system. In the third generation (3G) to 5G systems, orthogonal frequency division multiple access (OFDMA) is used. In general, most multiple access schemes used in cellular systems are orthogonal multiple access (OMA) schemes that allocate orthogonal channel resources to different users. To increase the spectral efficiency, NOMA schemes have been considered, where multiple users share the same channel resource.
NOMA has been extensively studied for cellular systems since the 2010s. In particular, for downlink transmissions, various NOMA schemes are studied using the difference in propagation loss between users near the center of the cell (where a base station (BS) is located) and users far from the center. The resulting NOMA is often referred to as power‐domain NOMA.
It is necessary to distinguish between power‐domain NOMA and NOMA in a broad sense. For example, code‐division multiple access (CDMA) and interleave‐division multiple access (IDMA) can be seen as NOMA schemes, because multiple users' signals can co‐exist in a shared radio resource block, where one user signal can see the other users' signals as interfering signals. In CDMA, each user's signal is spread by a dedicated sequence, which is called the spreading sequence. Due to spreading sequences, CDMA has a bandwidth expansion. In particular, the bandwidth of CDMA increases by a factor of the processing gain or the length of the spreading sequence, while IDMA is a generalization of CDMA with forward error correcting codes. On the other hand, power‐domain NOMA does not use spreading sequences. As a result, there is no bandwidth expansion and a high spectral efficiency can be achieved.
In power‐domain NOMA, however, the transmit power levels and transmission rates should be carefully decided so that successive interference cancellation (SIC) can be used to remove other users' signals once they are decoded.
1.3 NOMA for MTC
In general, power‐domain NOMA for downlink requires coordinated transmissions by a BS in terms of transmit powers and rates. Thus, it seems difficult to use power‐domain NOMA for uplink as coordinated transmission by distributed users is not easy to implement. In other words, the gain of NOMA can be offset by excessive signaling overhead to perform coordinated transmissions by distributed users. As a result, the use of NOMA for random access seems quite challenging. On the contrary, NOMA is well‐suited to random access as we will illustrate with two users.
Suppose that two users want to access a given channel without coordination. If two users always transmit their signals, they experience collisions and no user succeeds to transmit. Thus, they need to transmit with a certain probability. To this end, each user is to decide the access probability, denoted by , , for user . The probability that at least one user succeeds to transmit a packet is given by
If , , which is maximized when and the maximum of is , which is also the maximum average number of successfully transmitted packets. To consider random access with NOMA, we can assume two different power levels, and with , and the receiver is able to decode both the signals if one user transmits a signal with transmit power and the other with . Let be the probability that a user chooses the high transmit power when transmitting (with probability ). Then, the average number of successfully transmitted packets is given as follows:
where the first term is the average number of successfully transmitted packets when only one user transmits and the second term is the average number of successfully transmitted packets when two users transmit simultaneously. It is easy to show that maximizes . Then, we have . Thus, maximizes , which is 1. In other words, the maximum average number of successfully transmitted packets can be doubled if NOMA is used for uncoordinated transmissions of two users in uplink transmissions.
As more power levels are considered, the average number of successfully transmitted packets can increase. However, this comes at the cost of high transmit power by devices.
In this book, we mainly focus on the principles of NOMA and the application of NOMA to MTC. In particular, we discuss how NOMA can help improve the performance of random access in MTC once we present key ideas of random access including its stability. Game theory will also be used to understand the nature of random access where users compete for common radio resources in MTC.
1.4 An Overview of Probability and Random Processes
Prior to the main parts of this book, we present an overview of probability and random processes in this section, which can be used to see the required background in terms of probability and random processes. The reader is referred to well‐known textbooks such as Papoulis and Pillai (2002); Ross (1995); Mitzenmacher and Upfal (2005) if not well equipped with theory of probability and random processes.
1.4.1 Review of Probability
A sample space is the set of all possible outcomes (or events) of an experiment. Let be an event , which is a subset of . A probability measure is a mapping from to the real line with the following properties:
,
For a countable set of events,
, if
, for
, then
The joint probability of two events and is expressed as , where the conditional probability of given is expressed as
The two events and are independent if and only if
and this implies .
In addition, for any two events and , we have
where the equality holds if . Thus, for a set of events, it can be shown that
which is called the union bound.
1.4.2 Random Variables
A random variable is a mapping from an event in to a real number, denoted by . We first consider continuous random variables. The cumulative distribution function (cdf) of is defined as
and the probability density function (pdf) is defined as
where the subscript on and identifies the random variable. If the random variable is obvious, the subscript is often omitted.
Note that we use capital letters to denote random variables in this book if necessary. For example, is a random variable, while
