Full Chapter The Economics of Telecommunication Services An Engineering Perspective Pramode Verma PDF
Full Chapter The Economics of Telecommunication Services An Engineering Perspective Pramode Verma PDF
Full Chapter The Economics of Telecommunication Services An Engineering Perspective Pramode Verma PDF
https://textbookfull.com/product/multi-photon-quantum-secure-
communication-pramode-k-verma/
https://textbookfull.com/product/health-economics-an-
international-perspective-4th-edition-barbara-mcpake/
https://textbookfull.com/product/local-public-finance-and-
economics-an-international-perspective-harry-kitchen/
https://textbookfull.com/product/engineering-economics-decisions-
and-solutions-from-eurasian-perspective-svetlana-igorevna-
ashmarina/
Atrial Fibrillation from an Engineering Perspective
Leif Sörnmo
https://textbookfull.com/product/atrial-fibrillation-from-an-
engineering-perspective-leif-sornmo/
https://textbookfull.com/product/coulson-and-richardsons-
chemical-engineering-fourth-edition-volume-3a-chemical-and-
biochemical-reactors-and-reaction-engineering-r-ravi/
https://textbookfull.com/product/ecosystem-services-economics-
and-policy-stephen-muddiman/
https://textbookfull.com/product/understanding-automotive-
electronics-8-e-an-engineering-perspective-eighth-edition-
william-b-ribbens/
https://textbookfull.com/product/telecommunication-for-health-
care-1982-1st-edition-brown/
Textbooks in Telecommunication Engineering
Pramode Verma
Fan Zhang
The Economics
of Telecommunication
Services
An Engineering Perspective
Textbooks in Telecommunication Engineering
Series Editor
Tarek S. El-Bawab, PhD
Nile University
Giza, Egypt
Telecommunications have evolved to embrace almost all aspects of our everyday life,
including education, research, health care, business, banking, entertainment, space,
remote sensing, meteorology, defense, homeland security, and social media, among
others. With such progress in Telecom, it became evident that specialized telecom-
munication engineering education programs are necessary to accelerate the pace of
advancement in this field. These programs will focus on network science and engineer-
ing; have curricula, labs, and textbooks of their own; and should prepare future engineers
and researchers for several emerging challenges. The IEEE Communications Society’s
Telecommunication Engineering Education (TEE) movement, led by Tarek S. El-Bawab,
resulted in recognition of this field by the Accreditation Board for Engineering and
Technology (ABET), November 1, 2014. The Springer’s Series Textbooks in Telecom-
munication Engineering capitalizes on this milestone, and aims at designing, developing,
and promoting high-quality textbooks to fulfill the teaching and research needs of this
discipline, and those of related university curricula. The goal is to do so at both the
undergraduate and graduate levels, and globally. The new series will supplement today’s
literature with modern and innovative telecommunication engineering textbooks and will
make inroads in areas of network science and engineering where textbooks have been
largely missing. The series aims at producing high-quality volumes featuring interactive
content; innovative presentation media; classroom materials for students and professors;
and dedicated websites. Book proposals are solicited in all topics of telecommunication
engineering including, but not limited to: network architecture and protocols; traffic
engineering; telecommunication signaling and control; network availability, reliability,
protection, and restoration; network management; network security; network design,
measurements, and modeling; broadband access; MSO/cable networks; VoIP and IPTV;
transmission media and systems; switching and routing (from legacy to next-generation
paradigms); telecommunication software; wireless communication systems; wireless,
cellular and personal networks; satellite and space communications and networks;
optical communications and networks; free-space optical communications; cognitive
communications and networks; green communications and networks; heterogeneous
networks; dynamic networks; storage networks; ad hoc and sensor networks; social
networks; software defined networks; interactive and multimedia communications and
networks; network applications and services; e-health; e-business; big data; Internet of
things; telecom economics and business; telecom regulation and standardization; and
telecommunication labs of all kinds. Proposals of interest should suggest textbooks that
can be used to design university courses, either in full or in part. They should focus
on recent advances in the field while capturing legacy principles that are necessary for
students to understand the bases of the discipline and appreciate its evolution trends.
Books in this series will provide high-quality illustrations, examples, problems and
case studies. For further information, please contact: Dr. Tarek S. El-Bawab, Series
Editor, Department of Electrical and Computer Engineering, Nile University, Egypt,
[email protected]; or Mary James, Senior Editor, Springer, [email protected]
The Economics of
Telecommunication Services
An Engineering Perspective
Pramode Verma Fan Zhang
School of Electrical & Computer Two-bit Beijing Technical
Engineering Beijing, China
University of Oklahoma
Norman, OK, USA
This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
v
vi Preface
rather than information based goods and services. Information manifests itself both
as material goods and as a non-material entity defying the traditional approaches
to characterizing its economic impact and potential. Second, information goods
and services are driven, largely, by technology. Customer preferences do play a
part of course, but it is at a much finer level, and more in form rather than in
substance. In other words, the impact of technology influences the trajectory of the
evolution of information goods and services to a far greater extent than customers’
anticipation of and need for such evolution. The third reason is based on the fact
that technology is now recognized as a factor of production, supplementing the
traditional three factors of production labor, capital, and manpower. Unlike the other
three, however, technology is known to have no limits. And within the realm of
technology, information technology is, by far, the fastest growing component.
A notable barrier to motivation in understanding the economics of telecom-
munication-based goods and services in the past has been the fact that, until
recently, it was treated as a regulated monopoly and owned, in most countries,
by the state itself. Regulation of an industry, especially price regulation, distorts
its pricing structure and any attempt to relate pricing to the underlying cost.
Fortunately, a competitive landscape in the telecommunications space is emerging
quickly. This will increasingly demand an understanding of the cost associated with
telecommunication services.
The main emphasis in this book is on understanding the macroscopic parameters
that affect the quality of service, capacity, and scalability along with the underlying
cost of circuit and packet switched networks that offer a range of heterogeneous
services with potentially varying performance parameters. Quality of service param-
eters that characterize circuit and packet switched networks from a macroscopic
perspective are the grade of service and latency, respectively. Delivered traffic within
the user’s measure of acceptable quality of service is what the customer would pay
for. The network service provider needs to ensure that the resources provisioned
in the network are consistent with the capital and operational expenses it intends
to incur in providing the service. Absent this consistency, the service provider will
either cede its customers to competitors or go bankrupt. Indeed, this phenomenon
was widespread during the telecommunications debacle some 15 years ago.
The pricing model proposed in this book is based on the cost of displaced
opportunity as opposed to the cost of the elements of the network engaged in
delivering a particular service. The displaced opportunity is characterized by the
revenue associated with the service that the network could have alternatively
delivered most efficiently using an identical level of its resources.
This book also introduces the use of game theory in pricing services in a
competitive marketplace. Our belief is that as telecommunication transitions from
the realm of regulated monopolies to competitive businesses, game theory is going
to play an increasingly important role in the pricing of telecommunication services.
This will be especially true in a multi-vendor environment where each vendor offers
a number of services, each vendor’s services supported by its common transport
structure from which all its services are derived. The book has also used game theory
to balance the interests of the customer, the service provider, and the regulator.
Preface vii
The book is suitable for use by the senior undergraduate, graduate students of
telecommunications engineering, researchers, and practitioners in telecommunica-
tion engineering. However, it is primarily aimed at the practicing telecommunication
engineer who is tasked with pricing telecommunication services in a competitive
environment. It is the authors’ hope that the book will bridge the gap between
the science of economics as practiced by economists and the practice of pricing
as carried out by network service providers.
ix
Contents
xi
xii Contents
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
8 Pricing of Telecommunication Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
8.1 The Role of Pricing in Network Services . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
8.2 Traditional Methods of Pricing Telecommunication Services . . . . . 72
8.3 Characteristics of Communication Services . . . . . . . . . . . . . . . . . . . . . . . . 73
8.4 Other Considerations in Pricing Network Services. . . . . . . . . . . . . . . . . 75
8.5 Related Work on Pricing Network Services . . . . . . . . . . . . . . . . . . . . . . . . 76
8.5.1 Pricing for Regulated Telecommunication Services . . . . . . . 76
8.5.2 Pricing Internet Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
8.5.3 Recent Literature on Pricing Multi-Service
Communication Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
8.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
9 Pricing of Circuit Switched Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
9.1 Resource Consumption Based Pricing for Circuit Switched
Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
9.2 Pricing for Multi-hop Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
9.2.1 Pricing for a Single-hop Traffic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
9.2.2 Pricing for Two-hop Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
9.2.3 Pricing for Multi-hop Traffic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
9.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
9.4 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
10 Pricing of Packet Switched Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
10.1 Pricing Based on Bounded Delays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
10.1.1 Pricing for Single-hop Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
10.1.2 Pricing for Two-hop and Multi-hop Networks . . . . . . . . . . . . . 93
10.2 Pricing Based on Bounded Jitter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
10.2.1 Pricing for Single-hop Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
10.2.2 Pricing for Multi-hop Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
10.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
11 Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
11.1 Need for Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
11.2 Forms of Regulatory Safeguards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
11.2.1 Rate of Return Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
11.2.2 Price Cap Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
11.3 Emerging Regulatory Apparatus in the USA . . . . . . . . . . . . . . . . . . . . . . . 101
11.4 A Constant Revenue Model for Blocked and Lost Networks . . . . . . 102
11.4.1 Mathematical Construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
xiv Contents
xvii
Chapter 1
Characteristics and Characterization
of Information Networks
The contemporary society has been aptly named an Information Society. Our ability
to transform, communicate, and act on information has made dramatic changes,
especially over the last quarter century. These changes have affected the manner
in which we conduct businesses, lead our personal lives, and run government
operations, in fundamental ways. We anticipate even more changes in the future,
possibly at an increasingly rapid rate.
But what is information? Information occurs when uncertainty is removed. Math-
ematically, we characterize information by the magnitude with which uncertainty
has been removed. As a simple example, if the outcome—head or tail—resulting
from the toss of a coin is known, we have secured one bit of information. Similarly,
if one end of a transmission line transmits a 0 or a 1 and it is received at the
other end correctly, the transmission line has conveyed one bit of information. The
assumption, of course, is that both instances of the bit, namely, the 0 or the 1 are
equally likely to occur. If this were not the case, the information received will be
less than one bit. In the extreme case, if the receiving end had a priori knowledge of
the bit (to be received), the information received would be zero bits.
We might ask ourselves if information is a tangible commodity or an abstract
concept. Information behaves most of the time like a tangible commodity. It can be
bartered or sold like a commodity. The information recorded in this book is priced
by the publisher. The price might be different depending on whether the book is
made available in an electronic format downloadable from the publisher’s website
or be in a paperback or hardcover edition.
Information can substitute for material goods in definitive ways and its value can
be assessed as a material goods. For example, using better signaling techniques,
railroads can carry more passenger and/or goods saving the cost of steel or labor in
laying another track. Teleconferencing techniques, similarly, can displace the cost of
travel and lodging while saving time for participants at the same time. The resulting
savings are measurable, so the cost associated with the transport of information can
be justified using economics metrics.
Information, however, is different from material goods in many ways. For
example, it can be easily duplicated or stored at little cost. If information is stolen,
there is no specific marker that would indicate that it is somehow diminished. This
statement is generally, but not entirely, correct. For example, quantum cryptography
is based on the fact that, for information coded in quantum bits, it is possible to
detect if it has been observed or measured by a third party. This is quite unlike
a situation where coins of gold removed from a safe could provide evidence of
thievery. In a situation where a competitive bid from a vendor A is known to its
competitor before the latter puts in its bid, vendor A may never be aware of the loss,
yet its compromise could have a material impact. In addition to ease of duplication,
even more compelling is the fact that the cost of distributing information has fallen
down dramatically over the past few decades, and is likely to continue in the future.
Information has more value when it is made available at the right place at the
right time. Telecommunication networks offer the vehicle over which information
travels and reaches its intended destination. The management of telecommunication
networks from an economic perspective is the primary intent of this book. The
emphasis is on the economics of telecommunications networking.
Information may exist in more than one format. Speech, data, and moving or still
images are examples of information in different formats. The characterization of
information in each of these formats could be entirely different even though the
underlying entity, namely, the bit, is the same. This might require some further
elaboration.
Human speech, at its origin, exists in an analog format. It exhibits the singular
characteristics of the specific speaker in a variety of ways. Retaining the tonal
quality, the dynamic range of the speech, and the pauses or the gaps between words
and syllables with fidelity across a network that serves as the medium between the
two communicating parties is a large task. If these requirements are overlooked,
human conversation will lose its subjective impact. Pauses and gaps between words
or syllables, if not communicated with fidelity, might create a wrong impact on the
listener. The resources needed in the network that would convey the “naturalness”
associated with speech are very different from, say, conveying a transcript of the
conversation written in words and transported as data between two human beings. A
major objective of this book is to highlight the impact of resources needed to convey
information with varying parameters of performance across a telecommunication
network.
Speech is, most generally interactive and has, therefore, requirements related
to the absolute delay it can tolerate without noticeable degradation. Additionally,
stochastically varying delay or jitter is another requirement that affects the quality
1.3 Analog and Digital Information 3
of conversation. Even though the frequency range of speech might be limited, the
corresponding bandwidth requirement, when it is transported over a common user
network, in particular, a packet switched network, could be very different depending
on the degree to which the naturalness, as measured by jitter, is required to be
communicated. The exploration of the amount of resources required in a network
in communicating information to varying levels of quality is an important objective
of this book. In other words, we will address the notion of quality from a subjective
perspective while addressing its impact on the resources within the network in a
quantifiable manner.
communication between and among a pair (or more) of entities. From a general
perspective, communication between two entities can be defined as a programmed
response to a stimulus originating from one entity and conveyed to an appropriate
destination. Communication thus includes processing and storage in addition to the
commonly understood function of transmission.
As stated in Sect. 1.3, information in any mode can be converted into digital
format. Data is defined as machine originated information in digital format. Speech
transported over the network exists in digital format; however, it does not qualify as
data. This distinction is important from the network’s perspective because two ends
transmitting and receiving voice signals may lose synchronization for, say, 40 ms
without noticeably degraded performance from the users’ perspective. Such a loss
of communication will be unacceptable if data were transferred, however. In this
case, either the network or the endpoints must be capable of recovering the lost data
by other means, such as retransmission.
This section primarily addresses the impact of the size of the network on its
economics. Stated differently, to what extent, if any, does the classical rule of
economics—economy of scale—apply to communication networks. For the pur-
poses of this section, the structure of the network, the performance parameters
associated with transmission links, and other parameters related to network elements
are entirely omitted. The analysis presented below is from a comparative standpoint
only. It is not our intent here to compute the cost vs. size characterization of a
network in absolute terms.
Network externality implies that the value of a network is higher if there are
more users attached to it. A larger number of users increases the access space of
each individual user. Metcalfe’s Law [2], which is named after the inventor of
Ethernet, states that the value of a network is quadratically related to the number
of its users. This law is based on the observation that in a network with n users, each
user can make n − 1 other connections. This will result in a total of n(n − 1) possible
associations. The total value of the network is thus proportional to n(n − 1), that is,
approximately, n2 . This notion gives a large user base a competitive advantage for a
network because each of the large network’s users can communicate with a greater
number of other users. The Metcalfe’s Law obviously values each connection as
being of equal value to the network as a whole. A newly added node (or customer),
in order words, offers an equal benefit (of communication) to all the existing
network nodes.
The Metcalfe’s Law is an empirical observation made and heuristically validated
during the frenzy associated with the explosion in the telecommunications industry
fueled by deregulation in the mid-1990s. During this period, also characterized as
the Internet bubble period, entrepreneurs, venture capitalists, and engineers believed
in the steady and speedy commercial growth of the Internet. The Metcalfe’s Law
1.4 Networks: The Externality Factor 5
gave a quantitative explanation for the Internet boom with reasons like, “first move
advantage,” “Internet time,” and “network effects.” At that time, many companies
invested heavily in new fiber infrastructures not only at a backbone level but also at
the metropolitan level. Dense Wavelength Division Multiplexing made it possible
to transport up to 160 light waves on a single strand of fiber with a combined bit
rate into the range of terabits per second [4]. Ethernet technologies and the Internet
Protocol made the connectivity services to be supported on the fiber infrastructure
very inexpensive; a fact that likely is responsible for an overinvestment in the
telecommunication infrastructure during the 1990s [5].
Metcalfe’s law assigns equal value to all connections. This is a flaw because not
all possible connections are equally valuable for each user. For example, in a large
network such as the Internet, there are millions of potential connections between
users. In general, connections are not all used with the same intensity, and most
of them are not used at all. As a result, assigning equal value to all connection
is not justified and a revision of Metcalfe’s law is proposed in [6], based on the
assumption that not all connections are equally valuable to the users. The assignment
of value to each connection is based on the ZIPF’s Law [3] which is an empirical
rule. It says, for example, that in a long English language text, the most popular
words, or the most used letters, are roughly related to frequencies of occurrence
as: 1, 1/2, 1/3, . . . . This logic can be extended to a communication network with
n users. For each user, the value of connections to other users will be proportional
to 1 + 1/2 + 1/3 + · · · + 1/n, which approaches roughly ln n. There are other n − 1
users who get similar value from the network and the value of the network is thus
proportional to n ln n in the revised Metcalfe’s Law.
The revised Metcalfe’s law shows that the value of the network grows faster than
its size in linear terms and has a form of n ln n. The n ln n valuation describes a
slower growth (than what the classical Metcalfe’s Law would predict) in the value
of dot-com companies, and explains the Internet bubble from another angle.
In this section, we have reviewed the network externalities: The value of the
network increases as n ln(n), where n is its user base. This is faster than the linear
growth while the cost of the network is, at most, linear to its user base. A network
provider has, thus, a great incentive to price its services attractively to increase the
user base.
Consider two networks: network N1 and Network N2 with user bases of n1 and n2 .
Using the (classical) Metcalfe’s Law, let the parameters V N1 and V N2 capture their
respective values. We can write, V N1 = kn21 , and V N2 = kn22 , where k is a constant.
The combined value of the networks as separate entities can be given as:
V N = k(n1 + n2 )2 (1.2)
From Eqs. (1.1) and (1.2), it can be easily seen that, irrespective of the values
of n1 and n2 , the value of the combined network is always higher. This should
offer motivation to networks of any (relative) sizes to merge. However, in practice,
networks with a large user base have not been observed to court their much smaller
competitors with the intent of a possible merger. In the following sub-section, we
examine if the relative values of n1 and n2 affect the motivation for merger, assuming
that n1 + n2 is a constant.
As in the preceding section, let there be two networks, N1 and N2 , with user bases of
n1 and n2 , respectively. A combined network N will have a user base of n = n1 + n2 .
We wish to answer the following question: If n were to remain constant, what is the
relationship between n1 and n2 that will maximize the value of N? In the following
analyses, we address this question for both the classical and the modified Metcalfe’s
Law.
First, using the classical Metcalfe’s Law, we note that the value V1 of the network
N1 is given as, V N1 = kn21 , where k is a constant. Similarly, V N2 = kn22 .
The value of the combined network N can be given as, V N = k(n1 + n2 )2 . The
additional value created due to merger is,
higher than that of the larger network. This would explain the motivation of smaller
networks to merge with larger entities.
The analysis can be extented to include the modified Metcalfe’s Law as follows.
Following the same notations as before, the sum of the values of the two networks
prior to merging is: k(n1 ln n1 + n2 ln n2 ). Similarly the value of the combined
network is: kn ln n
It can be easily seen that,
n ln n > n1 ln n1 + n2 ln n2 (1.4)
since
n ln n = n1 ln n + n2 ln n (1.5)
Hence the merger benefits combining two networks accrues under the modified
Metcalfe’s Law as well.
In order to show that optimal value of merging is realized when the merging
entities have equal value, we proceed as follows.
Following the same approach as before, an optimum will occur when,
∂[n1 ln n1 + n2 ln n2 ]
=0 (1.6)
∂n1
or
∂[n1 ln n1 + (n − n1 ) ln(n − n1 )]
=0 (1.7)
∂n1
∂[n1 ln n1 ]
= ln n1 + 1 (1.8)
∂n1
∂[ln n] 1
= (1.9)
∂n n
and,
∂[(n − n1 ) ln(n − n1 )]
= −[1 + ln(n − n1 )] (1.10)
∂n1
or n1 = n/2 which proves that the maximum value of merger is realized when the
merging entities have equal value.
Underlying the motivation for merger of two separate networks, discussed in the
last section, was the fact that each user presented a uniform demand on the network
and contributed an identical value to the network. Merger of heterogenous networks
is a more complex issue from an analytical standpoint. We address this question in
Appendix B.
Problem Two networks with numbers of subscribers equal to 2000 and 3000,
respectively, merge. Find the relative increase in their respective values using (a)
the original Metcalfe’s law, and (b) its enhanced version.
Solution (a)
Prior to Merger
• Value of network1 = k20002
• Value of network2 = k40002
After Merger
• Value of the combined network = k60002
2
• Total relative value increase = 20006000
2 +40002 = 180%
After Merger
• Value of the combined network = k6000 ln 6000 = k52197.1
• Total relative value increase = 15201.8+33176.2
52197.1
= 108%
• Total increase in value: k(52197.1 − 15201.8 − 33176.2) = k3819.1
Distributing the value increase in proportion to their original values, value
increase of network1 = k 15201.8+33176.2
15201.8∗3819.1
or k1200.1. Network2 gains a value of
k2619.
If the increase in value were to be awarded equally, each network would gain a
value of k1909.55.
Problems
1.1 State Metcalfe’s Law. How does it relate the value of a network to the number
of subscribers it serves? Give one reason that would lead you to suspect that it might
be too aggressive.
1.2 Two networks with numbers of subscribers equal to 2000 and 3000, respec-
tively, merge. Find the relative increase in their respective values using the original
Metcalfe’s Law and its revised version.
1.3 The naturalness of voice in telephony is governed by two parameters—
fidelity and dynamic range. Which elements of the PCM system govern these two
parameters? How do we arrive at 1.544 Mbps as the line rate of the T1 system?
Explain the use of the framing bit in a T1 frame. (This question is not reviewed in
the book. However, it will be a good exercise to address it in the class room or have
students look up its answer in any book on telecommunication.)
1.4 Cable television channels today are priced in bundles. At the same time, it
is well known that most subscribers watch just a few channels in the bundle,
typically just about 10–20 channels in a 150-channel package. In a recent survey,
most consumers (>90%) expressed their desire to pay for only those channels they
regularly watch. It is well known that the popularity of the different channels is
vastly different, and these channels receive different amounts from the cable service
provider depending on the eye balls they attract. The Internet-based service pricing
for a single channel (e.g., Netflix) is about $10/month. State the likely consequences
of a la carte pricing of cable channels on the consumers as well as the service
providers. What would happen to the channels that are generally unpopular with
the masses? (This question is not reviewed in the book. However, it will be good
exercise to address it in the class room.)
10 1 Characteristics and Characterization of Information Networks
References
1. M.R. Aaron, PCM Transmission in the exchange plant. Bell Syst. Tech. J. 41(1), 90–141 (1962)
2. C. Shapiro, H.R. Varian, Information Rules (Harvard Business Press, Cambridge, 1999), p. 184
3. C.D. Manning, H. Schütze, Foundations of Statistical Natural Language Processing (MIT Press,
New York, 1999), pp. 24
4. NTT(2010-03-25), World Record 69-Terabit Capacity for Optical Transmission over a Single
Optical Fiber, Press release (2010)
5. C. Courcoubetis, R. Weber, Pricing Communication Networks: Economics, Technology, and
Modelling, West Sussex, England (Wiley, Hoboken, 2003)
6. B. Briscoe, A. Odlyzko, B. Tilly, Metcalfe’s law is wrong—communications networks increase
in value as they add members-but by how much? IEEE Spectr. 43(7), 34–39 (2006)
Chapter 2
Drivers of the Telecommunication
Industry
2.1 Bandwidth
simply the digital bandwidth in terms of bits per second that can be squeezed into
one Hertz of analog bandwidth. Spectral efficiency is most important in wireless
communication where the available analog bandwidth in a given space is limited
and cannot be duplicated as is the case of communication in a guided medium, such
as a fiber-optic cable. In the latter, multiple cables can be laid side by side increasing
the bandwidth proportionately.
The analog bandwidth and the digital carrying capacity of a transmission channel
are related by the classical equation developed by Claude Shannon in 1948 [1] and
is expressed as follows:
S
C = W log2 1 + (2.1)
N
where C is the channel capacity in bits per second, W is the bandwidth in Hertz, and
S and N are the power levels associated with the signal and the noise, respectively.
Equation (2.1) represents an upper bound of channel capacity under ideal
conditions. If we consider an analog telephone channel with the following band-
width and signal-to-noise characteristics: W = 3000 Hz, S/N = 30 dB or 103 then,
C = 3000log2 (1 + 1000) = 29, 880 b/s.
Capacity of optical fibers has dramatically improved over time. Current tech-
nology for commercial applications allows each fiber to have several independent
transmission channels with each channel having a transmission capacity of up to
40 Mb/s. Dense wavelength division multiplexing techniques can lead to each fiber
having several tens of independent channels allowing a single fiber to carry up to
a few Tb/s of information. The spectral efficiency has correspondingly grown from
about 1 bit/s/Hz to 10 bits/s/Hz over the past 15 years.
Needless to say, the dramatic improvement in the carrying capacity of fiber-optic
technology has brought down the transmission cost by orders of magnitude over the
corresponding period. Without such a reduction, the fast evolving cloud technology
allowing information to be stored and processed remotely and made available on
demand anywhere would not be economical.
correspondingly. The positive outlook for the chip industry continues unabated till
now. Sometime in 2017, semiconductor manufacturers shipped processors using the
10-nm chip-manufacturing technology [3].
We note in passing that the computing industry as a whole appears to be
driven by the Moore’s Law. As a result, anticipating the increase in processing
power, the application writers continue to create applications that are ever more
functional and that would depend on the anticipated increase in processing power.
The technology layers in between the raw processing power and the application,
the operating system, software languages, firmware, and middleware, continue to be
driven accordingly. This virtual lockstep among the different facets of the computing
industry has advanced the information industry as a whole to new heights where the
end user is looking for new applications and devices at regular intervals. This is a
unique phenomenon in the computing and telecommunication world, and is truly
global in scope.
Processing power along with the transmission capacity constitute the raw
material for telecommunication networking. A continuing reduction in their costs
would imply that telecommunication networking would displace other components
of business or personal endeavors to the extent that a cross elasticity exists between
these components and telecommunication-based applications. Indeed, this has been
the case over the past several decades. While forecasting of the evolution of
technology is fraught with risks, we do posit that the combination of reducing
costs of bandwidth and processing technologies, when combined with human
ingenuity in conceiving novel applications, will continue to enhance the relevance
of telecommunication networking in our personal and business world. Businesses
would do well by recognizing this trend and plan for correspondingly enhanced
investment in their telecommunication networking needs.
Like any other industry, the telecommunication industry has two major segments:
the customer segment and the supplier segment. Customers are further segmented
into business customers and consumers. Each of these segments buys products and
services. The supplier segment is further segmented into service providers and
equipment providers. Each of these sub-segments offers services or equipment to
both business customers and consumers. The segments are illustrated in Fig. 2.1. As
shown, the fact that the same service provider provides telecommunication service
to both the retail consumer and the business enterprise implies that there is an
inherent economy of scale associated with telecommunication services. Likewise
for equipment providers.
Innate characteristics of telecommunication products and services suppliers are
that they (suppliers) are global in scope as far as the footprints of their offerings
are concerned. The growing mobility of people in both the personal space and
14 2 Drivers of the Telecommunication Industry
Services
the business space would imply that planning for launching telecommunication
products and services should consider the global scope of the business.
As shown in Fig. 2.1, service providers offer services to both the business
and retail customers. Similarly, equipment providers supply equipment to both
the customer segments. Suppliers in the telecommunication business, in other
words, do not differentiate themselves by specializing in a specific market segment
characterized as the consumer or the enterprise market.
Sections 2.1 and 2.2 have captured the fundamental drivers of telecommunication
services. These factors are based on the increasing capability of the underlying
technology while at the same time reducing its costs. Another important factor
driving the demand for telecommunication services is the fast pace at which
user friendly applications are emerging. The combination of these two factors is
responsible, in large part, for expansion of the telecommunication services market.
This section captures the relationship between the value of telecommunication
services as perceived by the customer and the price paid for it. Economists present
the rationale for customer i buying a certain amount of a specific service by its utility
function, ui . If x is the amount of a service and the perceived value of the service by
the customer i is vi , then the utility of the service purchased is represented by,
Title: Majatalo
Language: Finnish
Kirj.
I. S. Turgenjeff
Suomentanut
Alexander Halonen
Majatalo.
Kaikin puolin hyvä olisi ollut Akim, tahi kuten häntä kutsuttiin
herrastalossa, jossa hän usein käväsi ja varsinkin sunnuntaisin
puolipäivämessun jälkeen, Akim Semenovitsh, — kaikinpuolin olisi
hän ollut hyvä, ellei häntä olisi seurannut eräs heikkous, joka jo
monen ihmisen on maailmassa perikatoon saattanut ja loppujen
lopuksi tuhosi hänenkin elämänsä — nimittäin, heikkous
naissukupuoleen nähden. Akimin rakkaus meni äärimmäisyyksiin,
hänen sydämensä ei mitenkään voinut vastustaa naisen katsetta;
hän suli siitä, niinkuin ensimäinen syksylumi auringonpaisteesta… ja
tavallisesti hän joutui maksamaan ylimääräisen tunteellisuutensa
tähden.