Study of the Telecommunication Networks Performance and Reliability

Indicators

NUGZAR CHEDIA

Department of Engineering and Construction, Faculty of Technology,

Batumi Shota Rustaveli State University,

12 Pirosmani Street, Batumi,

GEORGIA

MARINA CHKHARTISHVILI

Department of Engineering and Construction, Faculty of Technology,

Batumi Shota Rustaveli State University,

12 Pirosmani Street, Batumi,

GEORGIA

Abstract: - It is currently urgent to improve the technical characteristics and increase the efficiency of

telecommunication networks because of the global problem of ensuring the reliable transmission, processing

and protection of commercial information in telecommunication networks. The aim of this study is to assess the

reliability indicators of telecommunication networks and analyse the level of their development. The aim was

achieved through the methods that were used to assess the reliability of the entire network: the generalized

method of analytical assessment of the telecommunication network reliability; connection-based metrics;

scaling of telecommunication networks, and application of unified reliability indicators to routine control,

operation and maintenance. As a result of the research, the indicators of telecommunication networks were

studied with and without using connection-based metrics. As a result, the network metrics were also divided

into distribution and generation/transmission reliability and telecommunication network metrics – into

connection-, performance-, and condition-based metrics, which showed improvements in network performance

and reliability. As a prospect for further research, studying and improving the telecommunication network

reliability indicators is suggested.

Key-Words: - metrics, transmission system, power grid efficiency, information security.

Received: October 8, 2022. Revised: May 12, 2023. Accepted: June 14, 2023. Published: July 19, 2023.

1 Introduction

Critical infrastructure, such as telecommunication

networks, is an important driving force for

sustainable social and economic development.

Demand for telecommunication services is growing

with the continuous progress of society. As a result,

telecommunications permeate almost all aspects of

daily life in most parts of the world. Besides, they

play an increasingly important role in our society.

Failure of a telecommunications network can entail

huge losses — from economic to human lives,

depending on the reliability category.

Additionally, these crucial networks are

consistently exposed to inevitable risks,

encompassing natural calamities like earthquakes,

tsunamis, and volcanic eruptions, as well as human-

induced disruptions such as malicious cyberattacks,

terrorist incidents, and accidents.

The current and relevant example is the problem

with the electricity supply in Ukraine caused by the

military operations on its territory. Since the autumn

of 2022, the unified energy system of Ukraine has

been regularly subject to missile attacks, which has

led to high unreliability and instability of

telecommunication networks. In October 2022, tens

of millions of consumers faced a blackout, and they

are still subject to stabilization and sometimes

emergency power outages, which entails the

problem of access to telecommunications networks.

During the ice disaster in southern China in 2008,

[1], ice on overhead power lines and facilities

caused power outages in some cities and districts,

leaving about 30 million people without electricity.

The Wenchuan earthquake occurred in southwest

China in the same year, [2]. Eight districts were

completely disconnected during a specific period,

28,765 kilometers of optical cables were damaged,

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

204

Volume 18, 2023

and 142,078 electric poles collapsed. In 2013, two

transmission lines in the Arkansas power system

were cut, and one was disconnected by one person,

which caused a disconnection of electricity for

10,000 customers, [3]. In 2015, the Black Energy

malware hacked the Ukrainian power system,

leaving 700,000 households without electricity, [4].

The given examples indicate the need to ensure the

reliable operation of telecommunication networks.

Measurement of reliability is the main and critical

issue for the implementation of quantitative

management to prevent and mitigate potential losses

or damage. Reliability cannot be quantified without

metrics. Consequently, quantitative management

cannot be achieved. Clearly defined indicators of

network reliability enable quantifying,

understanding and reporting failure probabilities.

Besides, the system has the purpose of operation

and maintenance, [5], [6].

Research on telecommunication network

reliability indicators began several decades ago. A

system of reliability indicators and indicator

standards were established.

The aim of this study is to assess the reliability

indicators of telecommunication networks and

analyse the level of their development.

The research objectives are the following:

The development of a methodology for assessing

the telecommunication network's structural and

topological indicators will allow the evaluation of

the structural and topological indicators of the

telecommunication network. This will be achieved

by normalizing the structural reliability of the

network and decomposing the network into

information directions.

Analysis of the dependence of structural reliability

indicators on the structure of information

directions. The study involves the analysis of the

assessment of the dependence of structural

reliability indicators of information directions on

their structure, in particular, on the length and

number of paths included in the information

direction.

Identification of the main directions for further

research. The study also aims to identify the main

directions for further research in the field of

reliability of telecommunication networks.

2 Literature Review

In the 1980’s, researchers began to study the

reliability of energy distribution. In, [3], the authors

wrote two fundamental books. They formally

defined various reliability indices such as SAIFI,

SAIDI, CAIFI, CAIDI, and ASAI in these two

books, as shown below (Formulas 1-5).

, (1)

, (2)

(3)

, (4)

, (5)

In the formula, CI is the total number of customers

interrupted, N is the total number of customers

served, CMI is the total duration of customer

interruptions, and CN is the total number of

customers continuously interrupted during the

reporting period. In ASAI, the required customer

time is defined as the average number of clients

served over 12 months multiplied by 8760.

Besides, were summarized other indicators of

distribution reliability, such as the Average System

Interruption Frequency Index (ASIFI), the Average

System Interruption Duration Index (ASIDI), the

Customer Total Average Interruption Duration

Index (CTAIDI), Customers Experiencing Multiple

Interruptions Index (CEMIn) and Momentary

Average Interruption Frequency Index (MAIFI),

expressed as follows (Formulas 6-10), [3].

(6)

(7)

(8)

(9)

(10)

where — load of kVA connections interrupted

because of each disconnection , is the total

serviced load of kVA connections, is the recovery

time for one disconnection event, is the

total number of consumers who had n or more stable

disconnections, and TMI is the total number of

momentary interrupted customers.

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

205

Volume 18, 2023

Later, Brooks noticed that many consumers

were negatively affected by more subtle voltage

disturbances, such as dips and surges. They propose

four indicators for estimating the magnitude of the

root mean square (rms) change and the combination

of magnitude and duration:

1) the System Average RMS Variation

Frequency Index (SARFIx);

2) the System Instantaneous Average RMS

Variation Frequency Index (SIARFIx);

3) the System Momentary Average RMS

Variation Frequency Index (SMARFIx);

4) System Temporary Average RMS

(Variation) Frequency Index (STARFIx) (Formulas

11-14).

(11)

(12)

(13)

(14)

where x – rms voltage threshold value, – the total

number of consumers experiencing short-term

voltage deviations, – the total number of

consumers experiencing momentous voltage

deviations within 0.5–30 cycles, and –

the number of consumers experiencing transient

voltage deviations during the set duration,

represents the total number of consumers

experiencing transient voltage deviations during 3–

60 seconds.

In, [7], the authors classified the above

indicators into three categories: by users, load, and

system quality indicators. The first category

includes SAIFI, SAIDI, CAIFI, CAIDI, ASAI,

CTAIDI, CEMIn and MAIFI. The second category

contains ASIFI and ASIDI. The third category

includes SARFIx, SIARFIx, SMARFIx and

STARFIx.

While reliability metrics continue evolving in

academic circles, utilities and professional

associations work closely with academic researchers

to standardize metrics. In the late 1980’s, the IEEE

(Institute of Electrical and Electronics Engineers)

created the Working Group on Distribution

Reliability (WGDR). IEEE keeps updating this

standard. Currently, the standard is used by many

network operators in several countries.

We use the examples described in, [4], to clarify

the previously described measures. These indicators

can be illustrated by considering a part of a power

distribution system with six load point buses. This

basic system displays only the data (Table 1)

necessary to explain the metrics calculation. The

number of customers connected to these buses and

the average load is shown in Table 1, which serves

4,000 customers with a total load of 8 MW. The

consequences of downtime are shown in Table 1,

considering four system failures in a given calendar

year.

Table 1. Detailed information about the distribution

system

Loading point

Number of

consumers

Average

connected load

1000

5000

800

3600

600

2800

800

3400

500

2400

300

1800

Total

4000

19000

The above data are the basis for direct calculation of

some of the above indicators, as shown below:

Where

These simple numerical examples are used to

illustrate the application of reliability indicators. It is

recommended to refer to, [4], for a more detailed

introduction and application.

The energy system is evolving from centralized

production to distributed production with the

development of power electronics technologies and

the popularization of renewable energy sources, [4].

Besides, micro networks are also gradually

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

206

Volume 18, 2023

developing. However, the above indicators of the

distribution network still apply. In, [8], [9], the

authors used SAIDI and ASAI in AC and DC

distribution networks to create reliability models for

transformer stations, DC transformers, and DC

circuit breakers. In, [5], the authors used SAIFI,

SAIDI and CAIDI to assess microgrids, which are

distribution networks that include various

distributed generators (DG) and energy storage

systems (ESS) with the ability to operate in an

autonomous mode.

3 Materials and Methods

Various indicators are used to assess the structural

reliability of the network, which to one degree or

another, indicate the stability of the network’s

functioning against the failure of its elements —

nodes or cable connections.

Three groups can represent all characteristics of the

communication network:

- morphological characteristics;

- operational characteristics;

- economic characteristics.

Characteristics describing these properties and

used to evaluate the construction and operation of a

telecommunications network can be classified in a

similar way. The indicators of any of the specified

groups are divided into exogenous and endogenous

when setting the task.

Exogenous indicators can be considered

primary, predetermining certain concepts,

requirements, and conditions. As a rule, these

indicators are set on the communication network

from the outside when formulating the task and

describing or evaluating its construction and

operation.

Endogenous indicators are formed from

exogenous ones due to the description of the process

of building and functioning the communication

network. In this sense, they can be considered

derived from the initial set of indicators.

In general, the division of indicators of the

communication network into exogenous and

endogenous may not have a clear limit, being

determined by the nature of the task to be fulfilled.

This division is, however, convenient when

performing operations that reveal the physical

essence of processes or phenomena taking place in a

network, especially because endogenous indicators

are interconnected with exogenous direct or reverse

links, and exogenous ones —do not have an inverse

effect within one task.

Morphological indicators characterize the

construction of any communication network.

Exogenous indicators of this type primarily include

the size of the network N, which is determined by

the number of switching centres included in it.

Switching centres are connected by branches. Their

number — M — describes the ability of the network

to ensure the establishment of connections, as well

as for the endogenous characteristics of the structure

and operation of this communication network.

The communication network's functional

morphological element is the communication

direction (information direction). The number of

communication directions is set by the control

system provided by this network. The number of

paths πi of the establishment of connections

characterizes each ith communication direction. It

determines the structural reliability, bandwidth and

other indicators of this network, together with

several performance indicators.

The length of each path can be expressed by

the total number of switching centres included in it

— , or by the number of transit switching centres

— , by the number of branch paths — .

Some exogenous probabilistic indicators

estimate communication reliability, the types of

communication network reliability are structural —

W(G) and functional — W(F). Structural and

functional reliability are the main exogenous

indicators of this type: W(G) and W(F),

respectively.

It is necessary to use methods that calculate the

probability W(G) of failure-free service of

applications in the specified communication

directions to assess the structural reliability of a

telecommunications network. Only independent

paths should be used as possible information

transfer paths to calculate this probability.

3.1 Adequacy of the

Production/Transmission System

Capacity margin (CM) is another important

adequacy indicator that is widely used in the energy

sector and by regulatory bodies. It was first defined

as the difference between a set of specific system

conditions and actual system limitations, [10].

Consequently, it can also describe network

adequacy. This definition is qualitative, mainly

because of the high network CM. Therefore,

practitioners have few incentives for quantitative

assessment (Formula 15).

(15)

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

207

Volume 18, 2023

We use the data in Table 1 and Table 2 to

briefly describe the use of some indicators.

Table 2. Load data

Daily peak load (MW)

Number of events

116

107

The second system, with a capacity of 100 MW

can be used to calculate LOLE and other indicators.

Table 2 provides the load data for 365 days.

These straightforward numerical examples are

used to demonstrate the practical implementation of

reliability indicators. Researchers often use EENS

and LOLE among the above indicators. In, [11], the

authors used a two-state model to simulate a wind

turbine generator (WTG) system and used EENS

and LOLE to estimate the WTG. In, [12], the

researchers considered distribution networks with

renewable generators and calculated LOLE and

EENS using the DG system.

3.2 Mathematical model for Calculating

Structural Reliability

The mathematical model for calculating the

structural reliability of a communication network

involves determining various parameters that

describe the requirements and characteristics of the

network. These parameters are used to assess the

network's ability to maintain reliable

communication. The model considers the following

parameters:

- Number of sources (consumers) of information

recognized by the PU (Processing Unit). This

parameter, denoted as N, represents the total number

of sources or consumers of information in the

network.

- Number of directed communications between the

PU. Denoted as I, this parameter represents the total

number of directed communications or connections

between the Processing Units in the network.

- Normalized values of structural reliability of

communication network elements. The structural

reliability of communication network elements, such

as PU and cable communication, is represented by

normalized values. These values provide a measure

of the reliability of individual network elements.

According to the definition of the structural

reliability of the ith path, its value can be

determined as (Formula 16):

(16)

– the probability of the existence of the jth

element sequentially included in the ith path.

Revealing this parameter and taking into account

that the CC and the branches forming the ith path are

connected sequentially, the probability of survival

of the entire path is equal to (Formula 17)

(17)

— the probability of the existence of the

branch jh, which makes up the ith path;

– the probability of the existence of transit

CCs entering the ith path.

The above expressions defining the structural

reliability parameters and the main relationships

between these parameters can be considered:

- at the stage of network synthesis as initial data and

requirements for its structural reliability;

- at the stage of network analysis as the result of

ensuring the set reliability indicators of this network

and its main components — directed

communication.

Optimizing morphological indicators (for

example, the structure of the network) makes it

possible to achieve compliance of the obtained

result with the requirements put forward with

minimal material and technical costs.

We will use graph theory, which is a convenient

way to formally represent the structure of

communication networks, for the morphological

description of TCM.

We will present the TCM's structure through an

undirected graph . The graph's structure

is presented in the form of the

connectivity matrix .

The set of paths from one graph vertex to other

forms the direction of communication, and

independent paths are defined in the model.

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

208

Volume 18, 2023

When describing the communication system, it is

necessary to consider not only the morphological

but also the functional features of the

communication network, determined, for example,

by the procedures for choosing and making

connections. Using Dijkstra’s algorithm when

solving the TCM research problem is appropriate.

Considering that when determining the

structural reliability of the ith path, the reliability of

the elements that make up this path is used, it is

convenient to describe the value of the reliability of

network elements with matrices of a certain size.

So, the main indicators of this model were

determined: the telecommunication network

connectivity and reliability matrices.

One or several communication lines can form

each network, for example, Figure 1(a). Each

communication line is characterized by its survival

probability . The communication channels of

each communication line on the branch are

connected in parallel, forming a multi-line (a single-

line with one communication line) edge of the graph

, as shown in Figure 1(b). In this regard,

assuming the independence of the failure events of

each type of connection, and consequently the

independence of the probabilities , the

reliability probability of the branch can be

determined as (Formula 18):

, (18)

where L – is the number of independent

communication lines, the channels of which are

included in the branch . Values of the

probability determined by the expression are matrix

elements.

Fig. 1: Example of a branch: a) with several communication lines; b) multiline edge of the graph

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

209

Volume 18, 2023

4 Results

Publications discussing the conventional concept of

network reliability can be categorized into three

primary groups. The first category encompasses

connectivity-based metrics, prioritizing network

node connectivity. The second category comprises

performance-based metrics, which evaluate network

reliability based on network attributes and

predefined thresholds. The third category involves

state-based metrics linked to specific network state

measurements. Table 3 contains a list of subgroups

and corresponding publications in each group.

Table 3. Consequences of interruption in the 2022 calendar year

Interruption

event

Probability

Affected

load point

Number of

interrupted

customers

Reduced

load (kW)

Break

duration

(time)

Reduced

customer

working hours

Unsupplied

energy

(kWh)

0.3

800

3,600

2,400

10,800

600

1,800

8,400

0.4

300

1,800

600

3,600

0.2

600

2,800

600

2,800

0.3

500

2,400

1.5

750

3,600

300

1,800

1.5

450

2,700

Total

3,100

14,200

6,600

31,900

In summary, the interruptions in the

communication network affected 3,100 customers,

reduced the total load by 14,200 kW and interrupted

working hours by 6,600. As a result of these

interruptions, 31,900 kWh of energy was not used.

These findings emphasize the importance of

communications network reliability and its impact

on load, customer hours, and energy consumption.

4.1 Connection-based Metrics

Two special cases are important for the probability

of connection to the K-terminal: 1) the reliability of

a connection with 2 terminals (ST), where K = 2,

one node in K is designated as the source node, and

the second is the receiver node; 2) general reliability

(full ultimate reliability).

4.2 Performance Indicators

We illustrate these metrics by using the network

example in Figure 2. In this context, every edge

within the network has a particular probability of

failure, and there are limitations on the amount of

data each edge can transmit. In this scenario,

reliability refers to the probability that a minimum

of d units of data can be successfully transmitted

from source point s to destination point t.

Fig. 2: Network example, [3]

In addition to service reliability based on

network traffic, performance-based reliability also

includes time-based reliability. In this case,

reliability represents the probability that the time

required to move from point s to point t is less than

τ. The signal/interference/noise ratio (SINR) is

another important factor in the telecommunication

network performance. Signal quality interruption

characteristics are crucial for meeting strict

reliability requirements. Performance indicators are

expressed as follows (Formula 19):

, (19)

where T — threshold for SINR. This indicator is

widely used in telecommunication networks, [13].

4.3 State-based Metrics

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

210

Volume 18, 2023

Besides, some new metrics classify the state of the

function in the network. Was proposed a new index

that combines efficiency and connectivity. The

telecommunication network reliability is expressed

as follows (Formula 20):

(20)

where represents the probability of state x

and represents the system's efficiency. In this

study, reliability is referred to the maximum number

of channels of state x. Similarly, the number of

nodes with which this node can interact is used as a

reliability measure, [14].

Similar to energy losses in the network —

EENS — traffic losses in telecommunication

networks are also considered. In [15], [16], the

authors defined reliability as the probability of the

network being operational at any given time. One

measure related to reliability is the expected lost

traffic, which refers to requests that cannot be

delivered due to failures in the network. The

researchers also consider routing and rerouting

strategies that come into play after a failure occurs.

Their model comprises two layers: the physical

network layer, which represents the physical

channels, and the logical network layer, which

captures the traffic requirements of source and

receiving nodes. This indicator can be expressed as

follows (Formula 21):

(21)

Here, represents the lost flow in state x.

These network reliability parameters make the

network reliability measurement more user-oriented

and intuitive, significantly contributing to the study

of network reliability.

4.4 Practical Reliability Indicators

There is, however, a certain gap between reliability

indicators in practical applications and academic

circles. The first commercial 1G network was

launched in Japan in 1979. This is an analogue

system that does not provide for data transmission.

In 1991, the first 2G network based on the new

GSM (Global System for Mobile Communications)

standard was launched in Finland. Until the 2000’s,

telecommunication networks were mainly limited to

wired networks with relatively simple network

functions, mainly for telephone or telegraph

services. The call setup success rate (CSSR) and call

drop rate (CDR) are the relevant indicators

(Formula 22-23).

, (22)

(23)

Sc is the number of successful call connections,

Tc is the total number of call requests, Nd is the

number of dropped calls, and Ns is the number of

successful connections. However, these indicators

were extrapolated from data records and user

complaints at an early stage.

After the 2000’s, telecommunication networks

began to play an increasingly important role in

everyday life with the rapid development of 3G and

4G networks. Telecommunication networks still use

the reliability of all terminals, [17]. The

telecommunication network gradually became a

heterogeneous wireless and wired network with the

development of wireless technology. In addition to

traditional indicators, social demand for the network

has grown, and such indicators as unified

connectivity can no longer satisfy demand. The QoS

concept was gradually applied to the

telecommunication network services, [18].

According to previous studies, [19], [20], [21],

in the context of network analysis, several

fundamental metrics are commonly utilized,

including delay, reliability, throughput, jitter, and

loss probability. It's important to note that while

reliability is one of these metrics, it is primarily a

low-level metric and should not be confused with

overall network-level reliability. In other words,

while reliability focuses on the performance and

dependability of individual network components or

links, overall network-level reliability considers the

collective behaviour and performance of the entire

network. Some researchers represent reliability by

the frequency of bit errors, [13].

More specific key performance indicators (KPI)

appeared based on QoS and became more

widespread. Besides, many companies, for example,

Huawei will use early warning indicators (EWI) in

real work. Key performance indicators of the

network are grouped into the following

subcategories: availability, retention, mobility,

integrity, and accessibility, [13], [14], [15]. In [20],

the author, [18], classified KPIs according to the

following categories in the METIS project, co-

financed by the European Commission: traffic

density, empirical user throughput, delay, reliability,

and energy consumption. Table 4 shows key

performance indicators.

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

211

Volume 18, 2023

Table 4. KPIs for the LTE (Long Term Evolution) network

Category

KPI

Description

Calculation Method

Measurement

Unit

Accessibility

Successful installation of

RRC (Radio Resource

Control)

Measures the availability of

successful RRC signaling

establishment

Number of successfully

established RRC divided by the

total number of RRC

establishment attempts

Percentage (%)

Successful installation of

E-RAB (E-UTRAN Radio

Access Bearer)

Measures the availability of

successful E-RAB

establishment

Number of successfully

established E-RAB divided by

the total number of E-RAB

establishment attempts

Percentage (%)

Retention

Call drop rate

Measures the proportion of

dropped calls

Number of dropped calls divided

by the total number of calls

Percentage (%)

Service call drop rate

Measures the proportion of

dropped service calls

Number of dropped service calls

divided by the total number of

service calls

Percentage (%)

Mobility

Inter RAT Handover

Outgoing Success Rate

(Radio Access

Technology)

Measures the success rate of

outgoing handovers between

different access technologies

Number of successful outgoing

handovers between different

access technologies divided by

the total number of handovers

Percentage (%)

Handover Success Rate

(from LTE to WCDMA)

Measures the success rate of

handovers from LTE to

WCDMA

Number of successful handovers

from LTE to WCDMA divided

by the total number of

handovers

Percentage (%)

Integrity

Average uplink/downlink

bandwidth

Measures the average

throughput in the

uplink/downlink direction

Average uplink/downlink

bandwidth

Bytes per second

(Bps)

Bit error ratio

Measures the ratio of

erroneous bits to the total

number of transmitted bits

Number of erroneous bits

divided by the total number of

transmitted bits

Ratio

SINR (Signal-to-

Interference plus Noise

Ratio)

Measures the ratio of signal

power to the combined

interference and noise power

Measured in

decibels (dB)

Packet error rate

Measures the proportion of

packets with errors in the total

number of transmitted packets

Number of packets with errors

divided by the total number of

transmitted packets

Percentage (%)

Latency

User-plane latency

Measures the delay

experienced by end users

during data transmission

Measured in

milliseconds (ms)

Control plane delay

Measures the delay at the

control plane of the network

Measured in

milliseconds (ms)

End-to-end delay

Measures the delay across the

entire data transmission path

from sender to receiver

Measured in

milliseconds (ms)

One-way delay

Measures the one-way delay

from sender to receiver

Measured in

milliseconds (ms)

Availability

Radio Network

Unavailability Rate

Measures the proportion of

time when the radio network is

unavailable

Time when the radio network is

unavailable divided by the total

time

Percentage (%)

Cell availability

Measures the proportion of

time when an individual cell is

available

Time when the cell is available

divided by the total time

Percentage (%)

Traffic

Radio carriers

Measures the number of active

radio carriers

Number of active radio carriers

Inbound/outbound traffic

volume

Measures the volume of

inbound/outbound traffic

passing through the network

Number of transmitted bytes of

inbound/outbound traffic

Energy

Efficiency

Spectral efficiency

Measures the number of

transmitted bits per unit of the

frequency spectrum

Number of transmitted bits

divided by the frequency

spectrum width

Bits per Hertz

(bps/Hz)

The energy efficiency of

E-UTRAN data

transmission

Measures the number of

transmitted bits per unit of

energy consumed for data

transmission

Number of transmitted bits

divided by the energy consumed

Bits per Joule

(bps/J)

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

212

Volume 18, 2023

As the 5G era approaches, the

telecommunications network carrying this advanced

technology has turned into a giant. It constantly

grows, from simple calls/communications to

increasingly complex messaging capabilities.

Enjoying the convenience of a telecommunications

network, every user wonders about the reliability

and stability of the network, the completeness of

data, and the timely transmission. Society is

becoming increasingly dependent on

telecommunication networks with the transition to

5G. The International Telecommunication Union

(ITU) identifies the following main use cases for the

5G era:

- Enhanced Mobile Broadband (eMBB),

which is mainly determined by the need for high

data transfer speed and high bandwidth;

- Massive Machine-Type Communications

(mMTC), which requires energy consumption,

provides efficient communication and wide

coverage;

- Ultra-Reliable Low Latency

Communications (URLLC), where high reliability

and low latency are critical.

The 3GPP Radio Access Network (RAN)

Working Group (3GPP, 2018) determined design

goals for next-generation requirements as shown in

Table 5.

The aforementioned Key Performance

Indicators (KPIs) serve as valuable tools for

network monitoring, where the metrics obtained

through monitoring are compared to target values to

assess whether the current network meets the

specified requirements. For instance, peak speed

assumes greater significance in enhanced Mobile

Broadband (eMBB) scenarios due to clear

downloading and uploading requirements. The peak

download speed should not fall below 20 Gbit/s,

while the peak upload speed should not exceed 10

Gbit/s. However, peak speed is not mandatory for

Massive Machine-Type Communications (mMTC)

and Ultra-Reliable Low-Latency Communications

(URLLC) scenarios. In the case of URLLC, latency

emerges as a critical factor, with user-level latency

needing to be below 0.5 ms. It should be noted that

these metrics, in practical applications, are

predominantly deterministic values acquired

through network testing, diverging from the

probabilistic indicators of reliability explored within

academic circles.

Table 5. The 3GPP Radio Access Network (RAN) Working Group (3GPP, 2018)

Usage

scenarios

KPI

Goal

Download

Upload

EMBB

Peak data transfer rate

20 Gbit/s

10 Gbit/s

Peak spectral efficiency

30 bit/s/Hz

15 bit/s/Hz

Control level delay (same as URLLC)

10 ms

User level delay

4 ms

Average spectral efficiency (bps/Hz)

Three times higher than IMT (International Mobile

Telecommunication)-advanced

Zone bandwidth

10 Mbit/s/m2

Data transfer speed for the user

100 Mbit/s

50 Mbit/s

The efficiency of using the 5% user spectrum

(bit/s/Hz/user)

Three times higher than IMT-advanced

Target maximum speed (same as in URLLC

and mMTC)

500 km/h

Mobility interruption time (same as URLLC

and mMTC)

0 ms

Network energy efficiency (similar to URLLC

and mMTC)

No quantitative requirements

User equipment energy efficiency (similar to

URLLC and mMTC)

No quantitative requirements

Bandwidth

Not less than 100 MHz; Up to 1 GHz for operation in

higher frequency ranges (for example, above 6 GHz)

mMTC

Coverage

Max. communication loss 140 dB

User equipment battery Service life

older than 10 years, preferably 15 years

Connection density

1,000,000 devices/km2

Infrequent small packets delay

10 s

URLLC

User plane delay

0.5 ms

Reliability

1*10–5 success probability for 32 bytes within 1 ms

with a user plane delay

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

213

Volume 18, 2023

Nevertheless, there are certain drawbacks to

many existing KPIs. For instance, numerous key

performance indicators generate an overwhelming

influx of early warning alerts. Consequently,

operators are inundated with a high volume of KPI

alerts daily, necessitating increased maintenance

resources and augmenting the burden of operations

and maintenance. As mentioned earlier, several

practical metrics are deterministic monitoring values

lacking statistical significance, making it

challenging to estimate the probability of network

satisfaction. Therefore, the advancement and

operation of communication networks encounter

obstacles. In contrast to electrical networks, there is

an urgent need to investigate unified reliability

indicators for telecommunication networks in

academic and industrial settings.

5 Discussion

The assessment and forecasting of the structural

reliability of complex multi-service communication

networks cause many difficulties in practice because

of their large size. This is why a method of

assessing a telecommunication network's structural

and topological indicators when normalizing its

structural reliability is proposed. Its main objective

is decomposing the network, that is the division into

information directions.

The proposed methodology was the basis for

analying the results of the assessed dependence of

the structural reliability indicators of information

channels on their structure (length and number of

paths included in the information channel).

In other words, the study determined how the

structure of information channels affects structural

reliability indicators. At the same time, such factors

as the length of the paths and the number of paths

included in each information direction were taken

into account. This analysis will help to understand

which aspects of the information direction structure

affect the telecommunications network's reliability.

The process of developing the reliability of

telecommunication networks began with industry.

Later, the researchers started working on it. The

problems studied in academic circles are closely

related to production practice, and the gap between

these two areas is relatively small. Academic studies

are also successfully applied in industrial practice.

However, the scientific community began to study

telecommunication network reliability in the early

days theoretically. It took the industry a long time to

pay attention to the telecommunication network

reliability and develop several indicators. The

indicators previously developed in academic studies

were not used in the industry. This missing

application has created a big gap between academia

and industry.

It is possible to single out universal methods

and models for assessing the network structural

reliability among their multitude, which would be

suitable for the analysis of arbitrary network

structures and specialized ones that take into

account certain network features and thus enable

obtaining more accurate estimates for them than in

[18]. Most of these methods are intended for

analysing networks with a certain topology. The

indicators for assessing the structural reliability of

complex systems and networks are much better than

in [22], [23]. The works, [20], [21], deal with the

problems of assessing the structural reliability of

TCM. In contrast to [19], a method of obtaining

structural reliability based on basic structural

characteristics is presented for networks with a

certain topology. The paper, [20], considered the

possibility of using structural characteristics to

assess the reliability of a network with an

undetermined topology, while this article presents a

method for obtaining the upper and lower limits of

structural reliability for a single link in the network.

The power system's reliability concept has

evolved into two distinct categories: distribution

reliability and generation/transmission adequacy.

The network reliability index, represented by the

Equivalent Energy Not Served (EENS), can be

applied to the entire power network and evaluate

end-to-end network reliability for multiple services.

In telecommunication networks, traditional

reliability indicators are typically categorized as

connection-based or performance-based.

Theoretically, these indicators provide a global

assessment of the network's reliability. However, in

practical terms, there is a lack of metrics capable of

assessing the end-to-end reliability of the entire

network for multiple services offered within the

telecommunications domain, [24], [25], [26], [27].

The number of network reliability indicators is

appropriate, as IEEE standards were established to

determine these indicators. Besides, many countries'

network operators and regulatory authorities often

use several reliability indicators. However, there are

many KPIs for telecommunications networks, as

different telecommunication operators use different

KPIs. A standard set of metrics for different

operators has not yet been established, [28].

The results of the analysis of the structural

parameters of TCM and, in particular, the influence

of the CC ranks of this network on the parameters of

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

214

Volume 18, 2023

its operation are also confirmed when evaluating the

morphology of the studied network from the

perspective of structural reliability. In other words,

it can be argued that when choosing an option of the

communication network structure (where the

general topology of this network corresponds to the

independence of the paths of establishing

connections with a given number of CCs and

branches), the option with the minimum number of

CC ranks of this network will be more reliable than

any other. It is characteristic that the tendency to

preferentially choose the option with a

certain (minimum) number of CC ranks for the

assessment of its morphological characteristics, both

in terms of throughput and reliability, coincide, [14],

[29], [30].

The influence of the number of CC ranks of the

communication network on its structural reliability

is actually the only “hidden” factor that determines

the network performance according to a given

parameter. Summarizing the influence of other

parameters on reliability indicators, it can be noted

that the reliability of the information direction is

higher:

- the shorter the path (this is determined by

kilometer length for cable and fiber-optic lines, by

the number of transmission and reception sections

for radio relay and tropospheric communication

lines);

- the more independent paths of established

connections;

- the more linear connections ("parallel") form

one branch of the network;

- the higher the reliability of each element of the

communication network included in this

information direction.

Therefore, the discussed issues and proposed

methods for investigating the structural reliability of

multiservice communication networks significantly

contribute to this field's advancement. The results of

these studies can be used both in production and the

academic environment, fostering the improvement

of telecommunications systems and their reliability.

6 Conclusions

This study provides a systematic assessment of

reliability indicators in telecommunications

networks. The findings reveal that the reliability of

telecommunications networks is still in the

developmental stage compared to other

infrastructure networks.

To address the challenges associated with

evaluating and predicting the structural reliability of

complex multiservice communication networks, a

methodology for assessing the structural-topological

indicators of a telecommunications network is

proposed. The main objective is to decompose the

network into informational directions, enabling a

more comprehensive evaluation of its structural

reliability.

Simultaneously, the importance of reliability in

telecommunications networks is growing. Based on

the current state of research in this field, several key

areas for future investigation are suggested.

Researchers should establish a unified

measurement system among the multitude of

reliability indicators. As different users have diverse

requirements in various network scenarios, it is

essential to determine the relationship between

network requirements and reliability indicators and

establish a common reliability standard.

Telecommunications networks are complex

systems composed of multiple subnetworks with

different properties, each processing different

components. Further research should focus on the

methods for evaluating the reliability of the entire

network based on the reliability of its subnetwork

components.

The scale of telecommunications networks is

constantly expanding, with node counts ranging

from millions to hundreds of millions. Further

studies should determine efficient and rapid

approaches to network reliability assessment.

The establishment of reliability indicators is

intended for management and supervision purposes.

Future research should investigate the widespread

application of standardized reliability indicators in

routine monitoring, operation, and maintenance.

In conclusion, this study emphasizes the need for

further research in telecommunications network

reliability. Developing a unified measurement

system, studying the reliability of network

components, and devising fast and efficient

evaluation methods are key areas for further studies

in this area.

References:

[1] GPP. TS 32.450: Key Performance Indicators

(KPI) for an Evolved Universal Terrestrial

Radio Access Network (E-UTRAN), 2021.

[2] GPP. TR 38.913: Exploring Scenarios and

Requirements for Next Generation Access

Technologies, 2018.

[3] Adefarati, T., Bansal, R. K. Reliability,

economic and environmental analysis of a

microgrid system in the presence of renewable

energy sources, Prikladnaya energiya, Vol.

236, 2019, pp. 1089–1114.

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

215

Volume 18, 2023

[4] Alakhmed, A. S., Taivo, S. U., Abido, M. A.,

Almukhaini, M. M. Intelligent Flexible

Priority List for Microgrid Requirements

Reconfiguration Using Deep Neural Network,

Innovatsionnaya intellektualnaya set IEEE

Tekhnologii–Aziya, Chengdu, 2019, pp. 3490–

3495.

[5] Altiparmak, F., Dengiz, B., Smit, A. E.

General neural network model for assessing

the reliability of a telecommunications

network, IEEE Transactions on Reliability,

Vol. 58, No. 1, 2019, pp. 2–9.

[6] Asokan R. Overview of Quality of Service

(QoS) routing protocols for mobile peer-to-

peer networks, Mezhdunarodnaya

konferentsiya po besprovodnoi svyazi i

sensornim vichisleniyam (ICWCSC), Chennai:

IEEE, 2020, pp. 1–6.

[7] Benjebbur, A., Kitao, K., Kakishima, Y., Na,

K. 3GPP has defined 5G requirements and

evaluation conditions, NTT DOKOMO

Tekhnicheskii Zhurnal, Vol. 19, No. 3, 2018,

pp. 13–23

[8] Chamandust, H., Derakhshan, G., Hakimi, S.

M., Bahramara, S. Three-criteria planning of

intelligent distribution electrical networks of

residential buildings with an optimal

combination of reactive loads with renewable

energy sources, Zhurnal Nakopleniya Energii,

Vol. 27, 2020, pp. 101112.

[9] Chittum, A., Relf, G. Estimation of distributed

energy resources: combined heat and power

and modern grid, Zhurnal Elektrichestva, Vol.

32, No. 1, 2019, pp. 52–57.

[10] Dekhganian, P., Fotukhi-Firuzabad, M.,

Aminifar, F., Billinton, R. Comprehensive

Scheme for Reliability-Oriented Maintenance

in Power Distribution Systems - Part I:

Methodology, IEEE Transactions on Power

Delivery, Vol. 28, No. 2, 2013, pp. 761–770.

[11] Fan, H., Sun, X. Multi-State Reliability Model

for P2P Networks, Inzheneriya nadezhnosti i

sistemnaya bezopasnost, Vol. 95, No. 4, 2010,

pp. 402–411.

[12] Hariri, A. M., Hashemi-Dezaki, H., Hejazi,

M. A. A new generalized method for the

analytical assessment of the reliability of

smart grids, including renewable and non-

renewable distributed generations and plug-in

hybrid electric vehicles, Inzheneriya

Nadezhnosti i Sistemnaya Bezopasnost, Vol.

196, 2020, pp. 106746.

[13] Syan, S., Yan, J. Reliability of the K-terminal

of Ad Hoc networks, taking into account the

influence of node failures and interference,

IEEE Transactions on Reliability, Vol. 69,

No. 2, 2020, pp. 725–739.

[14] Chzhao, Kh., Khu, E., Chzhan, Z., Shen, Z.,

Fu, M., Van, R. Assessing the Reliability of

an AC/DC Hybrid Distribution Network

Considering Renewable Energy Sources, 4-ya

Mezhdunarodnaya konferentsiya po

energetike i vozobnovlyaemim istochnikam

energii (ICPRE), Chengdu: IEEE, 2019, 46–

50.

[15] Chzhun, I., Kvek, T. Q. C., Ge, Kh.

Heterogeneous Cellular Networks with

Spatio-Temporal Traffic: Delay Analysis and

Planning, Zhurnal IEEE po otdelnim

oblastyam kommunikatsii, Vol. 35, No. 6,

2017, pp. 1373–1386.

[16] Chzhu, I. Kh., Yan, Dzh., Tan, I. F., Sun, I.,

Khe, Kh. B. Coordinated attacks on

substations and power lines in electrical

networks, Austin: Globalnaya

kommunikatsionnaya konferentsiya IEEE

(GLOBECOM), 2014, pp. 655–661.

[17] Krasniki, F., Maradzh, A., Blaka, E. 4G/LTE

mobile network performance analysis,

Konferentsiya po avtomatizatsii

proektirovaniya, kompyuternoi inzhenerii,

kompyuternikh setei i obshchestvennikh SMI v

Yugo-Vostochnoi Yevrope

(SEEDA_CECNSM), Kastoria: IEEE, 2018,

pp. 1–5.

[18] Lee, R. M., Assante, M. J., Conway, T.

Analysis of a cyberattack on Ukrainian energy

systems, Tsentr obmena i analiza informatsii

ob elektroenergetike, 2016.

[19] Lemeshko, O. V., Garkusha, S. V.,

Yeremenko, O. S., Khailan, A. M. Policy-

based QoS management model for

multiservice networks, Mezhdunarodnaya

Sibirskaya konferentsiya po upravleniyu i

svyazi (SIBKON), Omsk: IEEE, 2015, pp. 1–

[20] Li, R., Li, M., Liao, H., Huang, N. An

Efficient Method for Evaluating End-to-End

Switched Ethernet Transmission Time

Reliability, Zhurnal setevikh i kompyuternikh

prilozhenii, Vol. 88, 2017, pp. 124–133.

[21] Memari, M., Karimi, A., Khashemi-Dezaki,

Kh. Evaluation of the reliability of an

intelligent network using various classical and

metaheuristic clustering algorithms, taking

into account system uncertainties,

Mezhdunarodnie sdelki po

elektroenergeticheskim sistemam, Vol. 31,

No. 6, 2021, pp. e12902.

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

216

Volume 18, 2023

[22] Mezhdunarodnii soyuz elektrosvyazi.

Minimum performance requirements for IMT-

2020 radio interface(s), Otchet MSE-R

M.2410-0, Sektor radiosvyazi MSE, 2017.

[23] Krasniki, F., Gavrilovskaya, L., Marai, A.

Analysis of key performance indicators (KPI)

in 4G/LTE networks, Future opportunities for

access to ubiquitous and intelligent

infrastructures: Materiali 4-i

Mezhdunarodnoi konferentsii EAI, Springer

International Publishing, 2019, pp. 285–296.

[24] Ibragimov, B. G., Isaev, Ya. S. Study of the

operability of the subsystem for transmitting

multimedia messages in telecommunication

networks, Trudi Mezhdunarodnogo

simpoziuma Nadezhnost i kachestvo, Vol. 1,

2018, pp. 10–12.

[25] Ibragimov, B. G., Gumbatov, R. T.,

Ibragimov, R. F., Isaev, A. M. Method for

calculating the functional reliability indicators

of the IMS platform in telecommunication

networks, Trudi Mezhdunarodnogo

simpoziuma Nadezhnost i kachestvo, Vol. 1,

2018, pp. 108–109.

[26] Sheluhin, O. I. Modeling of information

systems, Moscow: Goryachaya liniya-

Telekom, 2018.

[27] Ibragimov, B. G. Analysis of multi-service

telecommunication networks of the future

generation based on the architectural concept

of SDN & NFV and IMS, Uchenie zapiski

AzTU, Vol. 3, 2018, pp. 34–38.

[28] Efimushkin, V. A., Kozachenko, Yu. M.,

Ledovskikh, T. V., Shcherbakova, Ye. N. The

future shape of the Unified

Telecommunication Network in the Russian

Federation, Elektrosvyaz, Vol. 10, 2018, pp.

18–27.

[29] Ibrcihimov, B. G., Humbatov, R. Т.,

Ibrahimov, R. F. Performance multiservice

telecommunication networks based on the

architectural concept future networks using

SDN technology, T-Comm,

Telekommunikatsii i Transport, Vol. 12, No.

12, 2018, pp. 84–88.

[30] Ying, Z. Network function virtualization.

Concepts and Applicability in 5G Networks,

Wiley-IEEE Press, 2017.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

The authors equally contributed in the present

research, at all stages from the formulation of the

problem to the final findings and solution.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

No funding was received for conducting this study.

Conflict of Interest

The authors have no conflicts of interest to declare.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on SYSTEMS and CONTROL

DOI: 10.37394/23203.2023.18.21

Nugzar Chedia, Marina Chkhartishvili

E-ISSN: 2224-2856

217

Volume 18, 2023