Development of Cyber-Physical System Security Management
Process Model on the Example of Smart Home
Vyacheslav G. Burlov
1
a
, Elena S. Grozmani
1
b
and Sergey V. Petrov
2
c
1
Peter the Great St. Petersburg Polytechnic University, Saint Petersburg, Russia
2
Joint Stock Company "Research Institute "Rubin", Saint Petersburg, Russia
Keywords: Cyberphysical System, Security, Management, Model, Management Decision.
Abstract: Cyber-Physical Systems (CPS) is a hierarchical information technology environment resulting from the
convergence of general-purpose telecommunication networks and specialized technological devices designed
to control and monitor physical processes. This type of systems is actively implemented in industrial, urban,
and residential infrastructure, allowing for more efficient use of resources, improved quality of management
and feedback. As complex heterogeneous distributed information management systems, CPSs are susceptible
to cyber threats. The peculiarities of construction and functioning of cyber-physical systems require the
division of threats and attacks into two classes. The first includes malicious actions aimed at violating the
confidentiality, availability, and integrity of data processed in the system. Ensuring the security of
conventional information systems is aimed at preserving these very properties. The second class includes
cyber threats, the successful implementation of which can disrupt or modify the system's management and
control processes. Depending on the CPS's objectives and the degree of criticality of the real-world processes
it controls, disturbance of flows within it can lead to serious consequences, such as great material damage or
threat to human life and health. With the rapid development and improvement of Programmable Logic
Controller (PLC) devices and their cost reduction, the smart home concept is becoming increasingly popular.
This approach involves integrating PLCs communicating with each other and the control center via
multiprotocol computational public access (Internet) networks into the residential infrastructure. Thus, a smart
home is built on the basis of CPSs. In this application environment, it is essential to ensure the information
security of the cyber-physical systems used. To ensure the required level of security of these objects, it is
necessary to effectively solve the problems of managing the process of ensuring their information security.
On this basis, the paper proposes an approach to develop a model of the CPS security management process
on the example of a smart home.
1 INTRODUCTION
Cyber-physical systems are information-
technological objects resulting from the fusion of
physical process with controlling the programmed
environment, created based on developed
heterogeneous multiprotocol computer networks, and
including productive information systems, which can
possess properties of artificial intelligence. On their
basis, the control of direct executors (electrical,
hydraulic, thermodynamic, climatic, robotic systems
and complexes, etc.) is implemented together with
a
https://orcid.org/0000-0000-0000-0000
b
https://orcid.org/0000-0000-0000-0000
c
https://orcid.org/0000-0000-0000-0000
monitoring and data collection necessary for the
organization of timely and quality feedback.
On the other hand, CPSs can be considered as a
conceptual paradigm of production and technological
systems representation in the form of a conglomerate
of means of transformation of matter and energy
different types and information and
telecommunication environment, which provides
both information exchange between components and
functioning of the entire complex under variable
external environment using automated control.
Burlov, V., Grozmani, E. and Petrov, S.
Development of Cyber-Physical System Security Management Process Model on the Example of Smar t Home.
DOI: 10.5220/0010617900003170
In Proceedings of the International Scientific and Practical Conference on Computer and Information Security (INFSEC 2021), pages 87-92
ISBN: 978-989-758-531-9; ISSN: 2184-9862
Copyright
c
2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
87
To date, the most prominent examples of CPSs are
systems that provide the implementation of the
concept of smart home and the Internet of Things
(IoT), Industrial Control Systems (ICS), as well as
Supervisory Control And Data Acquisition
(SCADA). Geographically distributed CPSs include
traffic control and management systems. Examples of
local cyber-physical systems are the modern
automobile and sophisticated medical equipment.
Cyber-physical systems, being information
systems, are susceptible to the digital environment's
destructive factors - threats to information security. In
contrast to conventional information systems
designed to provide data processing, the main task of
CPSs is to manage and control the physical processes
of the real world. This defines a change in the
attackers' primary targets. For traditional information
systems, it is unauthorized access to protected
information and violation of data and services
availability. The main targets of attacks on CPSs are
to disrupt information flows within the system and
intercept the control of the actors (actuators). As a
result of changing the attackers' targets, the
consequences of successfully executed attacks
change as well. For information systems, this is
usually a financial and reputational loss. In the case
of CPSs, the damage can be quite different, ranging
from damage and destruction of technical facilities
and objects to harming human health and even
endangering human life. The situation is complicated
by the fact that targeted computer attacks using zero-
day vulnerabilities, social engineering, and specially
designed tools can be carried out against cyber-
physical systems. The complexity and heterogeneity
of CPSs, the presence of multiple cyber threats and
attackers with special means and high qualifications,
and great potential damage require the
implementation of a complex multipurpose
information security system.
The protection subsystem is a fully or partially
independent metasystem concerning the protected
and consists of a set of processes aimed at identifying
and neutralizing threats. At the same time, the
information security subsystem management process
should be viewed as even more high-level. Most of
the research focuses specifically on security
processes rather than on security management.
Simultaneously, in the absence of command decisions
adequate to the real situation, even the most perfect
system will not be able to achieve its targets with the
required indicators.
Another important issue is the selection and
justification of the requirements for the CPS
protection system. In this situation, it is necessary to
build a mathematical model of the system, which
would evaluate the prototype's characteristics even
before its creation. Currently, the prevailing approach
is to build models based on analysis. In their simpler
implementation, these solutions do not take into
account all the regularities and conditions of the
existence of the target processes within the system.
Thus, the implemented security systems do not have
the required properties and/or indicators and do not
provide the required security level.
Given the above, ensuring the secure functioning
of cyber-physical systems is a priority in the context
of their rapid development and widespread
deployment. At the same time, current approaches to
the organization and provision of the CPS safety
management processes do not allow achieving
management objectives with a guaranteed result. This
poses threats of unacceptable damage.
The purpose of this study is to develop a model of
the CPS safety management process that allows a
guaranteed result. A smart home was used as the
object on which the proposed approach was tested.
2 PROPOSED METHODOLOGY
At the heart of the management process is a
management decision. Any solution is built on the
basis of a control process or subsystem model. Thus,
if this model is not "good", it becomes impossible to
work out the right solution. What criteria does the
model have to meet? Its main and most important
property is adequacy. A model must sufficiently
account for the patterns and attributes of the objects
being reflected. When we talk about the management
process, which by definition is a continuum, it is
crucial to define its conditions of existence. Only with
the right solution to this problem is it possible to
achieve the management objectives. Whether it is a
human or an information management system, which
is ultimately the realization of the intentions of its
developers, the decision-maker has a conceptual and
logical apparatus used in problem-solving (Andreev,
Burlov and Grachev, 2019; Burlov, 2020). This
process is not an atomic operation but consists of the
following steps: decomposition, abstraction or
formalization, and aggregation (three-component
cognition). At the first stage, the problem is broken
down into separate complete blocks, which can be
solved using the system's methods. The formalization
process moves to a certain level of abstraction by
highlighting the properties and their evaluation
criteria required to solve each problem. The final step
is to process and merge the individual results
INFSEC 2021 - International Scientific and Practical Conference on Computer and Information Security
88
obtained. If a system is intelligent, it can carry out the
accumulation of "experience", received as a result of
aggregation processes, and produce new laws on its
basis. The law of preservation of object integrity was
used to work out process existence conditions
(Burlov, 2017; Burlov, Andreev and Gomazov,
2018). It is expressed in the mutual transformation of
its action's object properties and properties under a
fixed purpose. In the context of the problem we are
solving, the object, action, and purpose are
transformed into a setting, an information-analytical
work, and a management decision. Let us introduce
the following definitions. A management decision is
a condition for the subject to ensure the conditions for
implementing the purpose of the object he manages
in an appropriate environment to achieve the purpose
of management. Environment is a set of factors and
conditions in which activities are carried out.
Information and analytical work is continuous
extraction, collection, study, display, and analysis of
situation data. For example, marketing, exploration,
and monitoring (Burlov and Grachev, 2020; Burlov,
Abramov, Istomin, Fokicheva and Sokolov, 2018).
Thus, based on the principle of three-component
cognition and the law of preserving the object's
integrity, it is necessary to proceed to the synthesis of
the management decision process model. Its graphical
representation is shown in Figure 1.
Figure 1: Structure diagram of the interpretation of the
synthesis process of the decision mathematical model.
At the first level, applying the decomposition
method, we decompose the solution exactly into three
elements "environment", "solution" and
"information-analytical work", which correspond to
"object", "purpose" and "action". Applying the
method of abstraction at the second level, we identify
the "object" ("environment") with the periodicity of
the problem manifestation for the system - Δt
PM
,
which requires the development of a solution.
"Intent" ("Solution") is identified with the periodicity
of neutralization of the problem (average time of
adequate response to the problem) - Δt
PN
. "Action"
("Information-analytical work") is identified with the
frequency of problem identification (average time to
recognize an adverse situation) - Δt
PI
(Burlov, Uzun,
Grachev, Faustov and Sipovich, 2021; Chumakov,
Zakharov and Tumanov, 2018). The temporal
characteristics are justified by the fact that only
temporal resources are irrecoverable. Also, the results
of research in the theory of functional systems of the
USSR Academy of Sciences Academician P.K.
Anokhin (Anokhin, 1979) showed that the decision-
maker's decision is formed in the scheme
"excitation", "recognition", "reaction to the
situation". Therefore, the following diagram of the
expression of the decision model formation basic
components is used in the paper:
Figure 2: Diagram of the manifestation of the basic
elements of decision model formation.
With the help of aggregation the concept of
"management decision" is transformed into an
aggregate - a mathematical model of management
decision of the following kind:
P = F (Δt
PM
, Δt
PI
, Δt
PN
)
(1)
3 PROPOSED METHOD
Due to the fact that the basic model of management
decision has three elements, let us present the
management structural scheme as follows:
Development of Cyber-Physical System Security Management Process Model on the Example of Smart Home
89
Decision maker
Information and
analysis work
Activity
Resources
Formation of the command
to use the resources
Figure 3: Structure diagram of the control.
- the value inverse of the average time of
manifestation of the problem;
v1 is the value inverse of the average problem
identification time;
v2 is the value inverse of the mean problem
neutralization time.
The decision-maker, expressed as an information
command system, should ensure the fulfillment of
existing conditions of the target processes under
adverse external conditions by performing the
following tasks:
threat/problem identification (recognition);
neutralizing (responding to) the
threat/problem.
In this regard, four basic states of the decision-
maker can be distinguished (Burlov and Popov, 2017;
Sokolov, Alimov, Golubeva, Burlov and Vikhrov,
2018):
A
00
- decision-maker does not identify or
neutralize;
A
10
- decision-maker identifies and does not
neutralize;
A
01
- decision-maker does not identify and
neutralize;
A
11
- decision-maker identifies and neutralizes.
P
00
, P
10
, P
01
, P
11
- - probabilities of being in these
states respectively.
To implement this approach, it is necessary to make a
system of Kolmogorov differential equations that relate the
probabilities of finding the system in different states, and
these equations do not work with absolute intervals (time),
but with relative - frequencies (inversely proportional to
time).
So, let's consider the graph of states of the information-
management system:
Figure 4: State graph of the management decision-making
process.
λ is the intensity of the problem manifestation
(1/Δt
PM
);
υ1 - intensity of problem identification (1/Δt
PI
);
υ2 - intensity of problem neutralization (1/Δt
PN
).
Let us form a system of Kolmogorov differential
equations:
(2)
The transition from differential equations to
algebraic equations is possible if we assume that there
are no transients, then the derivatives
,
by the condition of function constancy (derivatives
equal to zero), besides the sum of all probabilities is
equal to one: P
00
+P
01
+P
10
+P
11
=1.
(3)
The sought probabilities are no longer time-
dependent. The solution of the linear algebraic system
of equations (3) is the following relations:
(4)
Having received the relations defining the
probabilities of the system being in A
00
, A
10
, A
01
, A
11
states, we can work out the requirements to the
INFSEC 2021 - International Scientific and Practical Conference on Computer and Information Security
90
intensity of the processes of recognition of the
problem potentially arising in the system and the
processes of their neutralization taking into account
the supposed frequency of the destructive factors
manifestation.
(5)
P
INP -
probability of identifying and solving the
problem facing the decision-maker.
Three parameters are related in this relationship.
Thus, the analytical dependence of generalized
characteristics of environment (Δt
PM
), information-
analytical activity (Δt
PI
) and work on neutralization of
destructive factors (Δt
PN
), which can be used to assess
models of information-analytical systems of
information security process management and obtain
the estimated values of key properties of these
systems, was established.
4 EXPERIMENTAL RESULTS
This method has been tested on the example of the
Smart House cyber-physical system. Research shows
that securing the Smart Home system is one of the
biggest challenges facing implementing this system
(Robles and Kim, 2010; Yang, Mistretta, Chaychian
and Siau, 2017; Li, Gu, Chen, He, Wu and Zhang,
2018).
One possible approach to determine the values of
λ, υ
1
and υ
2
is the use of network models.
A network model is a graphical representation of
a set of interrelated activities performed in a certain
sequence. The schedule consists of elements -
activities and events (usually indicated by arrows and
circles). The event has no duration in time. It marks
the completion of one or more activities that
determine whether subsequent activities can begin.
According to its role in the network schedule, a
distinction is made between an initial (initial) event -
any work of the considered complex does not precede
it; a final (final) event - after which no work within
the considered complex is performed; an intermediate
event fixing the end of previous work and the
beginning of subsequent work.
According to the calculations that were made in
the network planning, the following results were
obtained:
1) the average time for the problem to appear is:
= 2 (days) =2880 (minutes).
λ =
= 0.5 (the inverse of the average problem
time).
2) the average time to identify the problem:
= 116 (minutes) = 0.08 (days).
=
= 12.41 (the value inverse of the average
problem identification time).
3) the average time to neutralize the problem:
= 281 (minutes) = 0.195 (days).
=
= 5.12 (the value inverse of the average
problem neutralization time).
Let us consider the conditions for the existence of
the process for given probabilities:
=
= 0.876
=
= 0.036
=
= 0.086
=
= 0.016
Thus, we obtain a solution model based on the
synthesis approach, which allows us to evaluate the
information-management system's properties before
its construction. This approach ensures that the
management objective is guaranteed to be achieved.
5 DISCUSSION
In this section, we present for consideration some
open issues and shortcomings of the proposed
approach.
Cyberphysical systems are dynamic objects. This
means that they change over time and space. To
ensure the required CPS security level, the
information management system must promptly
adapt to changes in the protected object. The
proposed methodology, which is based on the law of
preservation of the object's integrity, makes it
possible to provide the required property.
At the same time, the decision-maker is constantly
acting in the presence of uncertainty. In the case
considered in the article, threats and vulnerabilities
used in attacks on CPSs have the property of
uncertainty. The presence of unknown attack vectors
or zero-day threats significantly reduces information
management information security systems'
effectiveness. The approach proposed in the article
does not take this factor into account at the moment.
The decision-maker, as stated, operates in an
environment in which there is some uncertainty or
ambiguity. Together with this, the information
management system, being a complex object, is also
prone to errors and failures. Thus, there is a
possibility of misbehavior of the information and
control system in which it fails to meet its target task.
Refinement of the proposed method taking into
Development of Cyber-Physical System Security Management Process Model on the Example of Smart Home
91
account this factor will improve the accuracy of the
results obtained.
6 CONCLUSIONS
In this article, the types of attacks on cyber-physical
systems have been identified due to their
characteristics. It is established that disruption of the
processes taking place in the CPSs and interception
of control over them can lead to significant material
damage and threaten people's lives and health.
To guarantee the achievement of managing the
process of ensuring information security of the CPS,
it is required to have an adequate model of the
information and control system, which acts as a
decision-maker.
The paper proposes a methodology and method of
constructing this model using a synthesis-based
approach. It was tested on the example of the CPS
smart house.
This paper can act as a basic guide for developing
models of the CPS security subsystems.
The purpose of further research is to refine the
parameters of the control process model, in particular,
it is required to take into account the probability of
making an erroneous decision.
REFERENCES
Andreev, A., Burlov, V. and Grachev, M. (2019).
"Information technologies and synthesis of the
management process model in the enterprise", 2019
International Science and Technology Conference
"EastConf", pages 1–5.
Anokhin, P. (1979). System mechanisms of higher nervous
activity, M.: Science, p. 453.
Burlov, V. (2017). The law of preserving the integrity of the
object - the method of the basis for solving the problems
of information warfare and ensuring security,
Neurocomputers and their applications. Abstracts,
pages 261-263.
Burlov, V., (2020). "Mathematical model of human
decision: A methodological basis for the functioning of
the artificial intelligence system", Proceedings of the
European Conference on the Impact of Artificial
Intelligence and Robotics, pp. 38-48.
Burlov, V., Andreev, A. and Gomazov, F. (2018). Safety
management of a technosphere object based on the law
of preserving the integrity of the object, Federal State
Budgetary Educational Institution of Higher Education
"St. Petersburg State University of Economics
(SPbSEU)", pp. 56-60.
Burlov, V. and Grachev, M. (2020). Model of Management
Decision Making in Enterprises Implementing
Information and Measurement Technologies, 2020
International Multi-Conference on Industrial
Engineering and Modern Technologies (FarEastCon),
Vladivostok, Russia, pages 1-5.
Burlov, V., Abramov, V., Istomin, E., Fokicheva, A. and
Sokolov, A. (2018). The methodological basis for the
strategic management of territory development, 18th
International Multidisciplinary Scientific
GeoConferences SGEM 2018. Conference proceedings,
pages 483-490.
Burlov, V. and Popov, N., (2017). Space geoformation
application management system in the interest of
ensuring environmental security of the region, in the
collection: The successes of astronautics, pages 751-
760.
Burlov, V., Uzun, O., Grachev, M., Faustov, S. and
Sipovich, D. (2021). "Web-based power management
and use model". Advances in Intelligent Systems and
Computing, 1258 AISC, pages 629–641.
Chumakov, N., Zakharov, A. and Tumanov M., (2018). In
the collection: Materials of the XVII All-Russian
Scientific and Practical Conference on Planning and
Training of Engineering and Technical Personnel for
the Industrial and Economic Complex of the Region,
PTES 2018.17.2018, pp. 156-158.
Sokolov, S. S., Alimov, O. M., Golubeva, M. G., Burlov,
V. G. and Vikhrov, N. M. (2018). The automating
process of information security management, 2018
IEEE Conference of Russian Young Researchers in
Electrical and Electronic Engineering (EIConRus),
pages 124-127.
Li, M., Gu, W., Chen, W., He, Y., Wu, Y. and Zhang, Y.
(2018). Smart Home: Architecture, Technologies and
Systems, 8th International Congress of Information
and Communication Technology (ICICT-2018), v. 131,
pages 393-400.
Robles, J. and Kim, T. (2010). Application, systems and
methods in smart home technology, A review. Int. J.
Adv. Sci. Technol, pages 37–48.
Yang, C., Mistretta, E., Chaychian, S., and Siau, J. (2017).
Smart home system network architecture.
INFSEC 2021 - International Scientific and Practical Conference on Computer and Information Security
92
