Você está na página 1de 4

Bonfring International Journal of Research in Communication Engineering, Vol. 8, No.

1, February 2018 1

A Systematic Study on Cyber Physical System


M. Revathi, K. Prakash, R. Suguna 

Abstract--- Cyber-physical systems (CPSs) are physical need to be modernized in order to increase their efficiency and
and engineered systems whose operations are monitored, reliability. Existing infrastructure is aging and control systems
coordinated, controlled and integrated by a computing and can be highly improved. Cyber-physical systems (CPS) stand
communication core. The economic and societal potential of to have a significant impact on large-scale, computer-mediated
such systems is vastly greater than what has been realized, physical distributed systems such as the electric grid and water
and major investments are being made worldwide to develop distribution systems. The cyber and physical components
the technology. There are considerable challenges, integrated in the system have many interactions the can affect
particularly because the physical components of such systems the entire operation, but are poorly understood. The electric
introduce safety and reliability requirements qualitatively grid and other utilities can use CPS technologies to help the
different from those in general purpose computing. Moreover, system become smarter and more efficient.
physical components are qualitatively different from object-
oriented software components. The vision for smart
manufacturing is for all aspects of manufacturing to be highly
integrated, from plant operation to supply chain. The entire
life cycle of a product would be enabled to be tracked,
including aspects such as processes and resources, leading to
manufacturing environments that are flexible and can optimize
performance and efficiency. Advances in CPS can help to
reduce the Time-to-market for products and systems, while
promoting innovation, competition, and resilience in supply
chains. These advances can allow a more flexible optimization
of cost and markets since Production could be located near
materials, technology skill centers, or consumers. The Future
workforce of manufacturing will have core knowledge and
skills that can be applied to innovation and research.
Keywords--- Computing, Optimize, Time-to-Market,
Resilience.

I. INTRODUCTION Figure 1: CPS Security Architecture

W HAT is a CPS? remains open, widely recognized and The infrastructure for both the electric grid and water
accepted attributes of a CPS include timeliness, distribution systems is aging and face technical and reliability
distributed, reliability, fault-tolerance, security, scalability and problems. The Environmental Protection Agency reports that
autonomous. Many aspects of our lives, including finances, there are 240,000 water main breaks per year in the United
transportation, and emergency services, depend on the reliable States, and as a system ages the number of main breaks
production, transfer, and distribution of energy, including increases. New technologies, including those in CPS regarding
electricity, oil, and natural gas. Today’s electric grid is a sensors and control, may help to increase the effectiveness of
complex system of systems, with many different stakeholders water utility distribution systems.
and customers. The National Academy of Engineers has CPS science and technology can provide many
named the development of electrification, which includes possibilities for innovation in the manufacturing industry
today‗s electric system, as the greatest engineering through smart products and production and lifecycle design
achievement of the twentieth century. The water supply and for product safety, security, and sustainability. The scope of
distribution system is also listed as a great engineering areas that benefit from cyber-physical engineering concepts,
achievement; the supply of safe, clean, and reliable water is infrastructure, and tools is broad and includes homes, military
important to health, quality of life, and emergency services. logistics and weapons systems, healthcare, clothing,
Yet even though these systems are great achievements, they agriculture, food processing and preparation, and automotive
and air vehicles. There is a trend towards the demand for
                                                                  products that have safe, dependable, and secure plug and play
M. Revathi, Assistant Professor, Department of CSE, Paavai Engineering
College, Namakkal. E-mail:revathimohanpec@paavai.edu.in integration of cyber and physical components thus leading
K. Prakash, Assistant Professor, Department of CSE, Government towards the possibility of being highly customizable. Mass
Engineering College, Thanjavur. E-mail:prakashgreen@gmail.com customization in manufacturing currently only involves
R. Suguna, HOD, Department of CSE, The Kavery Polytechnic College, minimal flexibility to alter fixed parameters.
Mecheri. E-mail:suguna.rathinam@gmail.com
DOI:10.9756/BIJRCE.8356

ISSN 2277 - 5080 | © 2018 Bonfring


Bonfring International Journal of Research in Communication Engineering, Vol. 8, No. 1, February 2018 2

 Quantified Confidence. This design goal is archived


by a Unified Event/Information model. Any event/
information in this architecture should contain the
following built-in properties. Global reference time–
records the event/information occurrence/detected
time. Life-span–specifies how long until that event/
information‘s confidence level drops to zero.
Confidence and confidence fading equation–specifies
the event/information confidence level and how it
fades over time. The confidence level and equation are
decided by a particular device and control logic to
provide a standard method for the subscriber to
calculate the confidence of the subscribed event/
information at any point in time. Digital signature and
Figure 2: Ecosystem authentication code–specifies who published and who
can access the event/information. Trustworthiness–
Manufacturing in the United States should move towards
specifies how much the subscriber trusts a particular
mass customization in order to enable tailoring of products
publisher. Dependability–specifies the subscriber‘s
based on real-time customer input while maintaining cost and
dependence on event/information provided by a
production efficiency. The increased innovation in materials
publisher in order to produce a particular outcome/
such as carbon-fiber composites, conductive polymers, metal-
information. Criticalness-specifies the critical urgency
ceramic compounds, among others, enables the creation, local
of each event/information so the subscriber can
sourcing, and production of smart products.
allocate its system resources based on different system
The economic impact of any of these applications would design goals.
be huge. Computing and networking technologies today,  Publish/Subscribe Scheme. Using this scheme, each
however, may unnecessarily impede progress towards these CPS Control Unit acts like a human being in that it
applications. For example, the lack of temporal semantics and only subscribes to interesting events/information
adequate concurrency models in computing and today‘s-best based on its system goal, and publishes event/
effort networking technologies make predictable and reliable information when necessary.
real-time performance difficult. Software component  Semantic Control Laws. Usually defined in an Event
technologies, including object-oriented design and service- Condition-Action like form, they form the core of
oriented architectures, are built on abstractions that match each CPS control unit. With the abstraction of the
software better than physical systems. Many applications will real-time physical world shown in Figure, we can
not be achievable without substantial changes in the core precisely control system behaviors related to the
abstractions. environment context according to user defined
CPS Definition and Architecture Cyber-Physical Systems conditions/scenarios. In addition to providing a global
are a next-generation network connected collection of loosely reference time, the next-generation network should
coupled distributed cyber systems and physical systems also provide new event/information routing and data
monitored/controlled by user defined semantic laws. Here, management schemes. Each network node should use
cyber systems are collections of control logic and sensor units, a-publish like scheme for passing event/information to
while physical systems are collections of actuator units. its neighbor nodes based on current confidence. Once
Prototype architecture of a CPS is shown in Figure. The the confidence of the event/information drops to zero
highlights of this architecture include: (or below a certain threshold), the value of letting this
event/information continue to-live in the network is
 Global Reference Time, which is provided by the next zero. The Secured Network Knowledge Database
generation network. This should be accepted by all Servers only accept the data when it expires, and
system components, including humans, physical serves as a knowledge backup.
devices, and cyber logic in this architecture.
 Event/Information Driven. Just as human society is II. CPS ARCHITECTURE
Event/Information Driven, future CPSs should also
CPS architecture had a two-layer structure inherently, the
use a similar communication mechanism. Moreover,
physical part and the cyber part. The physical part senses
we differentiate between Events and Information.
physical environment, collects data, performs commands from
Events are either-raw facts reported by sensor units/
the cyber part. The cyber part analyzes and processes the data
humans (called Sensor Events) or-actions made by
from the physical part and rise commands.
actuator units/humans (called Actuator Events).
Information is the abstraction of the physical world  Execution layer-Besides a target environment,
made either by CPS control units or humans through execution layer consists of various physical devices,
event processing. such as sensor, actuator, RFID (radio frequency
identification) tags, RFID readers, mobile intelligent

ISSN 2277 - 5080 | © 2018 Bonfring


Bonnfring Internationaal Journal of Reseaarch in Communiccation Engineeringg, Vol. 8, No. 1, Feebruary 2018 3

terminal and etc., which is in charge of collecting data


d leakagee or damage and
a security isssues during massive
m data
from physical
p envvironment and a perform
ming integrattion. Control layer attacks include the loss
l of user
commandds. privacyy, incorrect acccess control policies andd inadequate
 Transporrt layer-The next. securityy standards.
Chaallenging issu ues: The U.S S. National Academy
A of
Engineeering has com mpiled a list off grand challennges that are
t CPS. Amonng the many challenging
particullarly related to
issues related
r to socciety, economyy, and environnment, CPS
researchh is particularlly relevant safeety, stability, performance,
p
reliabiliity, robustnesss, and efficiiency, in areas such as
transporrtation, energgy, medical and healthcarre, defense,
manufaacturing, and agriculture. Some representtative grand
challengges for CPS research are as follows
fo
1. Near-zero auutomotive traaffic fatalitiees, minimal
injuries, and significantly reduced
r trafficc congestion
and delays.
2. Sustainable anda blackout-ffree electricityy generation
and distributioon.
3. Clean and eneergy-aware buiildings and cities.
4. Smart, reliablle, and flexiblle medical and healthcare
F
Figure 3: Systeem Architecturre systems.
 Control layer-Control layer is the key part of the 5. Abstraction annd Architecture.
interactioon between th he cyber part and the physiical 6. Computing annd Networkingg Foundations.
part Conntrol layer proccess the data which
w is colleccted 7. Hybrid System ms and Controll.
from sennsors and generate commaands which will w 8. Verification, Validation,
V andd Certification..
feedbackk to actuator in execute layer through transpport 9. Robustness, Safety,
S and Seccurity.
layer. Ass a part of con ntrol layer, appplication suppport 10. Stability, Perfformance, and Safety.
layer makkes use of mid ddle-ware. 11. Sensing, Com mputing, and Neetworking Systtems.
 Infotainmment platform m, cloud com mputing platforrm, 12. Modeling, Deesign, and Deveelopment.
service support
s platforrm and etc. Coombining conttrol Need foor CPS Researcch
layer withw variouss industries environmenntal
monitorinng smart grid, smart auto, and industtry, CPS S research is still in its infancy.
i Profeessional and
control come
c into beinng generation network
n suppoorts institutiional barrier have resulteed in narrow wly defined,
transportt layer, the maain technical methods
m of whhich disciplinne-specific ressearch and eduucation venues in academia
contain LAN,
L Internett, communicattion network and a for thee science andd engineering disciplines. Research is
private network.
n Transpport layer not only
o realizes reeal- partitionned into isolaated sub discciplines such as sensors,
time trannsmission, but also has the ability
a to proccess commuunications and networking,
n coontrol theory, mathematics,
m
and manaage amounts of data softwarre engineeringg, and computter science. For example,
systemss are designed and analyzed using
u a variety of modeling
formalisms and toolss. Each represeentation highliights certain
featuress and disregarrds others to make analyssis tractable.
Typicallly, a particularr formalism reppresents either the cyber or
the phyysical process well, but not both. Whereass differential
equationns are usedd for modelling physical processes,
framew works such as Petri nets annd automata are used to
represennt discrete beehavior and control
c flows.. Workforce
expertisse is similarrly partitionedd, to the detriment
d of
a efficiency. Although this approach to
producttivity, safety, and
modelinng and formalisms may sufficce to support a component-
based-ddivide and connquerǁ approacch to CPS devvelopment, it
poses a serious probllem for verifyiing the overalll correctness
and saffety of designss at the system m level and coomponent-to-
componnent physical and a behaviorall interactions. The broader
Figuure 4: Attack an
nd Defenses Syystem aeronauutics communnity has ideentified verifi fication and
Based on thee CPS architeccture attacks are
a classified into validation methodologies and conccepts as a critiical research
exeecution layer,, transport laayer and conntrol layer. The T area. Thhe goals of reesearch in veriification and validation
v of
exeecution layer attacks
a include security attackks for nodes suuch aviationn flight-criticall systems incluude providing methods for
as sensors and actuators. Transsport layer attaacks include data
d rigorous and systemattic high-level validation
v of syystem safety
propertiies and requiirements, from m initial desiign through

ISSN 2277 - 5080 | © 2018 Boonfring


Bonfring International Journal of Research in Communication Engineering, Vol. 8, No. 1, February 2018 4

implementation, maintenance, and modification, as well as [7] L. De Alfaro and T.A. Henzinger, “Interface theories for component-
based design”, International Workshop on Embedded Software,
understanding tradeoffs between complexity and verification Pp. 148-165, 2001.
methods for supporting robustness and fault tolerance. Some [8] S.A. Edwards and E.A. Lee, “The case for the precision timed (PRET)
of the control engineering challenges include: machine”, Proceedings of the 44th Conference annual Design
Automation, Pp. 264-265, 2007.
 Large-scale, real-time, deterministic robust or [9] S. Johannessen, “Time synchronization in a local area network”, IEEE
stochastic optimization. control systems, Vol. 24, No. 2, Pp. 61-69, 2004.
 Algorithms [10] E.A. Lee, “Model-driven development-from object-oriented design to
actor-oriented design”, Workshop on Software Engineering for
 Multiple-objective, multiple-stakeholder optimization Embedded Systems: From Requirements to Implementation, 2003.
frameworks. [11] J.A. Stankovic, “Misconceptions about real-time computing: A serious
 Design of automation with graceful degradation problem for next-generation systems”, Computer, Vol. 21, No. 10,
Pp. 10-19, 1988.
modes.
 Safety diagnosis/health monitoring methods.
 System architectures that facilitate distributed decision
making.
 Data fusion from heterogeneous sensors and
assessment of the value of the Derived information.

III. CHALLENGES OF CPS


 Scientific Challenges: Multi-Paradigm–
Multidisciplinary; Socio-Technical Character–
Foundational Theories; Multi-domain modeling.
 Technological Challenges: Complexity–
Interoperability; Autonomy – Privacy; Dependability;
Uncertainty
 Economic Challenges: Disruption – Service vs.
Product; Disruptive dominance.
 Education Challenges: Competence– Knowledge
Excellence; – Balanced Education; Future Work force.
 Legal Challenges: Innovation Barriers – Relatory
environment – Certification, Harmonization.
 Societal Challenges: Change-Aversion – Public
awareness; Public Support.

IV. CONCLUSION
Cyber-physical systems are expected to play a major role
in the design and development of future engineering systems
with new capabilities that far exceed today‘s levels of
autonomy, functionality, usability, reliability, and cyber
security. Advances in CPS research can be accelerated by
close collaborations between academic disciplines in
computation, communication, control, and other engineering
and computer science disciplines, coupled with grand
challenge applications.

REFERENCES
[1] T. Henzinger, E.A. Lee, A. Sangiovanni-Vincentelli, S. Sastry, A.
Aiken, D. Auslander and M. Tomizuka, “Center for Hybrid and
Embedded Software Systems”, 2008.
[2] R.R. Rajkumar, I. Lee, L. Sha and J. Stankovic, “Cyber-physical
systems: the next computing revolution”, In conference on Proceedings
of the 47th design automation, Pp. 731-736, 2010.
[3] R. Baheti and H. Gill, “Cyber-physical systems”, The impact of control
technology, Pp. 161-166, 2011.
[4] J. Fitzgerald, C. Gamble, P.G. Larsen, K. Pierce and J. Woodcock,
“Cyber-physical systems design: formal foundations, methods and
integrated tool chains”, Proceedings of the Third FME Workshop on
Formal Methods in Software Engineering, Pp. 40-46, 2015.
[5] Martin Törngren, Cyber-Physical Systems: Characteristics, Trends,
Opportunities and Challenges, 2015.
[6] R. Kumar, Challenges of Cyberphysical systems, 2012.

ISSN 2277 - 5080 | © 2018 Bonfring