A metamodel for cyber-physical systems
Theresa Fitz, Michael Theiler, and Kay Smarsly
Chair of Computing in Civil Engineering
Bauhaus University Weimar, Germany
Abstract. With the advent of the Internet of Things and Industry 4.0 concepts, cyber-physical systems
in civil engineering experience an increasing impact on structural health monitoring (SHM) and
control applications. Designing, optimizing, and documenting cyber-physical system on a formal basis
require platform-independent and technology-independent metamodels. This study, with emphasis on
communication in cyber-physical systems, presents a metamodel for describing cyber-physical
systems. First, metamodeling concepts commonly used in computing in civil engineering are reviewed
and possibilities and limitations of describing communication-related information are discussed. Next,
communication-related properties and behavior of distributed cyber-physical systems applied for SHM
and control are explained, and system components relevant to communication are specified. Then, the
metamodel to formally describe cyber-physical systems is proposed and mapped into the Industry
Foundation Classes (IFC), an open international standard for building information modeling (BIM).
Finally, the IFC-based approach is verified using software of the official IFC certification program,
and it is validated by BIM-based example modeling of a prototype cyber-physical system, which is
physically implemented in the laboratory. As a result, cyber-physical systems applied for SHM and
control are described and the information is stored, documented, and exchanged on the formal basis of
IFC, facilitating design, optimization, and documentation of cyber-physical systems.
Keywords: Cyber-physical systems, structural health monitoring (SHM), wireless sensor networks,
metamodeling, semantic modeling, building information modeling (BIM), Industry Foundation
A cyber-physical system (CPS) is commonly referred to as a coupled system integrating computing,
networking, and physical processes . As such, cyber-physical systems are hybrid hardware/software
systems coupling heterogeneous subsystems able to sense, to act, and to communicate through
networks . Recent advancements in Industry 4.0-related research have paved the way for CPS
applications in civil engineering . Specifically, modern structural health monitoring (SHM) and
control systems exhibit all features listed in the above CPS definition . SHM and control systems
enable real-time monitoring and assessment of structural conditions by automated data acquisition,
data analysis, and appropriate actions performed by intelligent sensor networks and networked control
devices spatially distributed over the structures being monitored. In civil engineering, cyber-physical
systems consisting of SHM and control systems are used to improve structural control of civil
structures, such as bridges  or high-rise buildings in earthquake areas .
In recent years, cable-based SHM systems traditionally applied in structural health monitoring have
been progressively replaced by wireless SHM systems using “smart” sensor nodes with embedded
intelligence or on-board computing and sensing capabilities, respectively . The terms “smart” or
“intelligent”, unlike common definitions of intelligence in computer science, denote the embedment of
algorithms and models for on-board data analysis into sensor nodes . SHM systems, according to
, provide, on demand, reliable data about conditions of structures being monitored. Therefore,
wireless SHM systems need to integrate sensing devices into sensor networks distributed over a
structure that perform on-board data acquisition, data processing, and data communication . By
embedding engineering models, such as models of the structure or structural components, into SHM
systems, automated damage detection is supported .
Taking into account the rapid advancements in sensing technologies, formal semantic information
modeling concepts are needed to describe information about cyber-physical systems applied for SHM
and control. Information about SHM and control, referred to as “monitoring-related information”, and
not to be confused with sensor data, must be described independently of technical platforms and
programming languages [5, 10]. Technology-independent semantic descriptions of SHM and control
systems hold essential potentials for documenting and exchanging information about system
compositions and system states, including documentations of changes in setup and functioning of
cyber-physical systems . Using ontologies and description languages designed for describing
distinct subdomains of monitoring-related information, e.g. sensor characteristics or network
characteristics, the description of SHM and control systems is partially possible. However, limitations
are obvious, if communication-related information is to be described in the context of structural
systems of a CPS. Typically, referencing capabilities from communication-related information to
elements of the structural system of a CPS is not possible. In addition, distinct characteristics of SHM
and control systems are not covered by the ontologies and description languages, e.g. embedded
algorithms, overall SHM strategies, diagnosis levels, and history of sensor configurations. To
overcome the absence of referencing capabilities and distinct characteristics, this paper uses building
information modeling (BIM) concepts to describe both the structural system and communication-
related information in a holistic approach. In the architecture, engineering and construction industry,
technology-independent semantic descriptions of buildings using BIM based on the Industry
Foundation Classes (IFC) standard are already well-established and using IFC-based BIM approaches
is mandatory in many countries [12, 13, 14]. In Germany, following a phase plan published in 2015,
BIM is to be applied to all new projects in the area of the Federal Ministry of Transport and Digital
Infrastructure from 2020 onwards . The German Federal Ministry of Transport and Digital
Infrastructure states that IFC-based data exchange in structural engineering is well-established.
Additionally, the descriptiveness of civil infrastructure may be improved by extending the IFC
Although describing building information and infrastructure information using IFC is an active field,
recent research has shown that the current version of the IFC schema is not yet sufficient to describe
all aspects that enable the use of BIM models over the life cycle of buildings or infrastructure. For
example, for BIM models describing cyber-physical systems, monitoring-related information and
communication-related information need to be incorporated into the IFC schema [5, 16].
Communication-related information is a subset of monitoring-related information describing
communication technologies, such as communication protocols, routing of communication including
origin and destination of each communication process, transmission media, and technical devices
employed to realize sensor communication. Besides technological aspects, information about the data
exchanged between communicating sensor nodes is of interest for detailed descriptions of cyber-
To describe communication-related information on a well-defined basis, in this study, first an
overview of metamodeling approaches relevant to cyber-physical system modeling is provided
(Section 2). Then, theory and technical details on network communication are summarized in a
literature review (Section 3). Next, a BIM-based metamodel for cyber-physical systems applied for
SHM and control systems is presented using a technology-independent metamodel, with emphasis on
semantic descriptions of communication-related information (Section 4). Subsequently, preparing a
validation of the BIM-based description approach applied in this study, the metamodel is mapped into
the IFC schema that is extended for BIM-based descriptions of cyber-physical systems in form of an
IFC schema extension (Section 5). Upon verification of the extended IFC schema, the validation is
shown on a prototype CPS applied for SHM and control in the laboratory (Section 6). Finally, the
results are summarized and conclusions are drawn in Section 7.
2. Metamodeling approaches relevant to modeling cyber-physical systems in civil engineering
For modeling cyber-physical systems, a variety of metamodeling approaches can be applied. In this
section, basic concepts of metamodeling are illuminated and three common metamodeling approaches
are analyzed for describing information related to cyber-physical systems, (i) the Unified Modeling
language (UML) and additional modeling languages published by the Object Management Group
(OMG), (ii) the seven standards forming the Sensor Web Enablement (SWE) framework, and (iii) the
data modeling language EXPRESS, all following the object-oriented paradigm, are metamodeling
approaches frequently used in computing in civil engineering.
In software engineering and systems engineering, models are used to capture real-world aspects of
problem domains with different levels of abstraction . While models are abstractions of
phenomena in the real world, metamodels are further abstractions that specify the structure, the
semantics, and the constraints for a family of models that are situated in a certain domain. The term
“modeling” describes design techniques and development processes that require technical frameworks
for information integration and for tool interoperability. Software and systems engineering approaches
based on models are described in technical frameworks, referred to as model-driven development
(MDD). In 2000, the OMG, an international, open-membership, non-profit technology standards
consortium, has published the Model-Driven Architecture (MDA) standard, which is today a widely
used realization of MDD . The MDA framework has been developed to separate specifications of
system functionalities from platform-specific system implementations [17, 18]. Therefore, the MDA
design process starts with a platform-independent model (PIM) describing functionalities and behavior
of a system. In subsequent steps, a PIM is converted into a platform-specific model (PSM) and into a
working implementation. The purpose of a PIM, e.g. for describing cyber-physical systems, is to
remain stable as technology evolves and to enable mapping between different modeling languages,
such as between UML and other OMG standards, the SWE framework, and the EXPRESS data model
analyzed in the following subsections.
2.1 UML and other OMG standards
To develop models that remain stable as technology evolves, modeling languages, which are, e.g.,
based on OMG’s Meta Object Facility (MOF) are used to describe models in a platform-independent
manner . As can be seen from the schema of a MOF-based metamodeling approach in Figure 1,
the MOF is a so called “meta-metamodel”, situated on layer M3 providing a platform-independent
metadata management foundation for MDA and serves as a model of different modeling languages,
referred to as “metamodels” on layer M2 shown in the schema. The main goal of the MOF is to
provide a basis for defining and extending metamodels and models (layer M1) by generalizing the
core concepts of different modeling languages. The key modeling concept of MOF follows object-
oriented paradigms of classifiers and instances, or classes and objects, respectively. Thus, it is possible
to navigate from an instance (layer M0) to its class (layer M1), i.e. the metaobject, across any
metalevel or degree of abstraction, respectively .
Figure 1 MOF-based metamodeling approach.
The sequence of deriving metamodels from a meta-metamodel and models from metamodels forms
the basis of a well-defined modeling process. As a result, different models and metamodels can be
interchanged, because syntax and semantics of modeling languages are clearly defined, thus being
mappable to each other.
With the MOF, MDA of OMG aims to create, to store, and to transform machine-readable models
based on a rigorous underlying modeling infrastructure. Applying MOF-based modeling standards,
such as Unified Modeling Language (UML) and other OMG specifications, the meaning of diagram
elements and relationships between elements of systems can be captured in a machine-readable form,
enabling automated consistency control and generation of application code . MOF reuses a subset
of structural modeling symbols of UML 2, which contains key modeling concept of classifier and
instance (or classes and objects) for software development .
Within the UML specification, the semantics of UML, defining the meaning of statements made in
UML models, are subdivided into two semantic categories, structural semantics and behavioral
semantics . Structural semantics, forming the basis of behavioral semantics in UML, define the
meaning of UML model elements, such as classes, components, relationships, and data types.
Behavioral semantics describe the communication between structural elements influenced by methods
of associated model elements. Behavioral modeling can be used to express interaction sequences
necessary to describe communication processes. Regarding cyber-physical systems and, in particular,
communication-related information, UML offers a wide variety of notations and modeling constructs
to describe architecture and behavior of computational systems. An advantage of UML is the technical
capability of UML to derive MOF-compliant metamodels from the general UML specification by
creating profiles for different modeling purposes and domains, which may be used to develop
metamodels of cyber-physical systems.
2.2 The SWE framework
The Open Geospatial Consortium (OGC), an international non-profit organization, has published a
family of standards forming the SWE framework and providing metamodeling capabilities for
geospatial systems, such as sensor systems, with the key idea to make data of sensor systems online
accessible through interfaces and protocols following well-defined standards. The SWE framework is
composed of seven standards employing UML notations and XML notations to represent conceptual
schemas for describing sensor networks, sensors, sensor observations, and measurements . Within
the standards family of the SWE framework, the Sensor Model Language (SensorML) is used to
encode sensor descriptions. Sensor observations are described using the Observation & Management
(O&M) standard. To provide standardized access to sensor data and to sensor descriptions, the Sensor
Observation Service (SOS) has been introduced . In compliance with its concept of metamodeling,
the SWE framework aims to link several sensor-related technologies, while avoiding restrictions upon
specific products and approaches. Systems created using design principles of the SWE framework are
thus technology-independent and allow further extensions [28, 29].
The SWE framework also focuses on processes and processing components of sensor systems using
syntax and semantics of the SensorML metamodel. The core concept of SensorML is to describe
components of sensor systems (e.g. sensors or actuators) as processes. In general, processes may
receive inputs, generate outputs, and have parameters. Hence, SensorML is a process description
language supporting data of different formats exchanged between logical processes [30, 31].
SensorML is independent from communication protocols used to exchange data between system
components . In addition, SensorML can be used to describe interface characteristics, such as
communication protocols, baud rates (speed of communication over a data channel), and port settings
(e.g. port number, port type) . However, no graphical notation or exhaustive XML encoding has
been standardized for interface characteristics.
2.3 The EXPRESS metamodel
Another metamodel relevant to modeling cyber-physical systems in civil engineering is EXPRESS,
standardized in ISO 10303-11 . EXPRESS is a data modeling language and a metamodel that
provides computer-interpretable representations of product information, designed to enable product
data exchange . EXPRESS is used in civil engineering to formally define the IFC specifications as
a basis for open-source BIM . Recent research on integrating sensor data into building information
models has demonstrated the effectiveness of EXPRESS-based data modeling, therefore modeling
capacities of EXPRESS with respect to describing monitoring-related and communication-related
information are further discussed here . The key concept of ISO 10303 is to create models of
product data that may be used in software applications during the life cycle of products. The life cycle
of products includes, e.g., design, construction, production, marketing, application, and recycling.
To realize consistent exchange, storage, archiving, and transformation of product data defined in
EXPRESS-based data models, the “Standard for the Exchange of Product Model Data” (STEP, ISO
10303-21 ) is used. The metamodel EXPRESS includes a textual as well as a graphical
representation, EXPRESS-G, that comprises a subset of constructs of the textual modeling language.
Figure 2 shows main elements of EXPRESS-G describing the composition a node that can be either a
base station or a sensor node having a sensor or an optional actuator attached. Accordingly, the entities
describing nodes are shown in textual EXPRESS notation.
Figure 2 Example of EXPRESS-G and EXPRESS.
Although the metamodels presented in this section vary in (textual or visual) notation, models, i.e.
instances of metamodels representing real-world systems can be derived following metamodel-specific
syntaxes and semantics. UML and related OMG specifications, in comparison to EXPRESS, possess a
more comprehensive range of modeling capacities, because more visual notations are standardized and
can be represented. UML is characterized by a high flexibility and adaptability to various modeling
purposes. Using UML profiles, additional metamodels can be derived from UML that are as well in
compliance to MOF and the MOF-based metamodeling approach. The SWE framework also applies
UML notations, thus being comparable to special-purpose UML profiles restricting the variety of
UML notations for a special-purpose metamodel. The standards forming the SWE framework provide
UML models and XML schemas facilitating technology-independent descriptions of sensor networks
including physical system elements and constructs, such as observations and measurements
In summary, UML is the most general and most flexible metamodel reviewed herein. The wide scope
and notational variety reason the complexity of UML and make the metamodel applicable to many
modeling purposes, such as to describing cyber-physical systems. Thus, UML constructs for structural
modeling and behavioral modeling are exceptionally valuable for describing communication-related
information in cyber-physical systems. While components of cyber-physical systems, such as
communication units of sensor nodes, can be described using class diagrams, algorithms may be
modeled using state machine diagrams, and processes implemented in communication protocols can
be visualized in sequence diagrams. On the other hand, it should be emphasized that, due to the
growing importance of open BIM in civil engineering, the modeling approaches of EXPRESS along
with the IFC standard to describe structural systems is gaining attention in research and practice.
3. Communication in cyber-physical systems
To describe cyber-physical systems using metamodeling approaches, system components, including
attributes and methods, need to be defined. As a basis to characterize attributes and methods, in this
section, network topologies applied for communication in cyber-physical systems are studied. Also, a
survey of network communication is provided, and communication protocols suitable for cyber-
physical systems are reviewed. Developing a metamodel capturing communication-related
information, such as system components related to communication, network topologies, network
communication characteristics, and communication protocols, enables distinct advantages, such as
effective planning of SHM and control systems, improved data control and management in the
operation and maintenance phase of CPS, or linking of SHM and control system components with
As nodes of wireless sensor networks are spatially distributed, autonomous devices, power
consumption and resource management are important criteria in designing communication protocols
and network topologies . Topologies typically applied in wireless sensor networks are shown in
Figure 3 [9, 37]. Star topologies and mesh topologies are suitable for communication in cyber-physical
systems, therefore frequently applied in SHM and control applications. In star topologies, sensor nodes
are connected to a central node, e.g. to a base station. Communication is exclusively realized between
sensor nodes and the central node. As a consequence, failures of a central node result in failure of the
total sensor network. However, star topologies are tolerant to failure of single sensor nodes, rendering
star topologies suitable for SHM and control applications. Mesh topologies enable communication
across all nodes in a network. Partially connected and fully connected mesh networks can be
distinguished. In fully connected mesh networks, nodes directly communicate with other nodes,
whereas in partially connected mesh networks several nodes are connected indirectly. Mesh topologies
allow calculating optimal routing paths, which contributes to an efficient exploitation of energy
resources distributed over a wireless sensor network.
Figure 3 Visual representation of network topologies.
As data exchanged in sensor networks encompasses several layers of abstraction, a layered model of
networking has been developed by the International Organization for Standardization (ISO), published
in ISO/IEC 7498-1:1994 . The “Open Systems Interconnection (OSI) reference model” defines
seven layers of abstraction in network communication, shown in Figure 4, and the function of each
layer. Following the OSI reference model, nodes of networks communicate at equivalent layers of
abstraction by layer-specific protocols. Every layer of abstraction n-1 provides services to higher
abstraction layers n through service access points specified for each layer .
Figure 4 Node-to-node communication using the OSI reference model.
In the OSI reference model, layers 1 to 4 form the transport system of a network, while layers 5 to 7
represent application-oriented layers. On the physical layer (layer 1), bit transmission over a physical
medium, such as air or water, is specified. Protocols on the physical layer encode parameter
modulation schema, transmit power, and hop distance of data packets, which contribute to energy
consumption of wireless sensor networks. The modulation schema specifies the transformation of the
bit values 0 and 1 into electrical quantities. The term “transmit power” denotes the amount of energy
required to transmit data packets. Transmit power is in inverse proportion to errors in data
transmission and related to the distances between the transmitter and the receiver of communicating
nodes, which is called “hop distance” [37, 39]. The data link layer (layer 2) organizes bits of data
units into frames, referred to as multiplexing and demultiplexing, respectively. The data link layer is
responsible for error detection through check sums added to a data packet, and it provides flow control
to manage data transmission rates. Furthermore, medium access control (MAC) is performed on the
data link layer. The network layer (layer 3) manages routing, i.e. identifying optimal paths to forward
data from sensor nodes to base stations by passing intermediate nodes [37, 39]. The fourth layer, or
transport layer, establishes communication between nodes defined by the routing process of the
network layer. Transport layer protocols transform data units (multiplexing or demultiplexing) to
make data accessible to end-system processes. Moreover, transport layer protocols provide services,
such as error detection in data units, flow control and service reliability by the authorization of
retransmission of a data unit . The session layer (layer 5) enables and controls communication
processes for complete data exchange between nodes by managing multiple transport layer
connections. Following the session layer, the presentation layer (layer 6) performs data
transformations to make data processable by the application layer (layer 7) . The data format is
defined by a uniform syntax, referred to as transform syntax, enabling correct representations of data
with respect to encoding mechanisms, such as ASCII code or Unicode. Finally, the application layer
provides protocols to be used by software applications .
On each layer of the OSI reference model, several protocols exist that can be combined to a specific
protocol stack. In a protocol stack, data is transmitted from a node to another node by passing every
layer of abstraction (Figure 4). Starting from the application layer of a sending node, control
information is successively added to a data unit and removed in inverse direction at a receiving node.
Regarding the energy constraints of wireless sensor networks, which are typically composed of
battery-powered components, the deployment of cross-layer protocols improves communication
efficiency within a network. Cross-layer protocols integrate several functionalities of different OSI
communication layers into one protocol to enable highly reliable communication with minimal energy
consumption, adaptive communication decisions, and local flow control .
Due to the plenitude of communication protocols, suitable communication protocols for cyber-
physical systems are defined when designing a CPS for structural health monitoring and control,
following a requirements analysis applying specific criteria, such as the size of data packets to be
transmitted, the data transfer frequency, and the transfer range between nodes. Furthermore, limits on
the number of communicating nodes and security issues must be considered in selecting
communication protocols. Security is of growing importance in in cyber-physical systems specially in
those applied in Industrial Internet of Things (IIoT) applications . Security objectives in
communication between nodes of a network are (i) confidentiality and integrity of data, (ii)
authenticity of system elements, and (iii) data access authorization . Data confidentiality denotes
restrictions of readability of data to authorized network components, while data integrity describes the
verification of sources sending data as well as the detection of data modified and sent by unauthorized
sources. Therefore, system elements require unique identifiers (authenticity) and access authorizations
to read, to write, and to manipulate data. As security issues are of growing importance in CPS, IoT,
and IIoT (however not within the main scope of this study), information describing security-related
network properties is formalized along with communication-related information and monitoring-
Table 1 provides an example selection of communication protocols frequently applied for wireless
communication in cyber-physical systems, including Bluetooth, ZigBee, Wi-Fi, and the Message
Queue Telemetry Transport protocol (MQTT). ZigBee and MQTT are characterized by low power
consumption and provide embedded security strategies with a high degree of scalability with respect to
the number of nodes. High-level protocols, such as ZigBee and MQTT, are based on implementations
of the physical layer and the data link layer of the OSI model (Figure 4) [41,44]. A number of low-
level networking protocols, encompassing functionalities of the physical layer and the data link layer,
are standardized in the IEEE 802 standards , such as Wi-Fi (IEEE 802.11.a/b/g) and Bluetooth
(IEEE 802.15.1). A low-power and low-cost solution for wireless communication is given by the low-
rate wireless personal area network (LR-WPAN) standard IEEE 802.15.4 that allows setting up
networks with star and mesh topologies. An overview of frequency ranges and data rates in
compliance with local regulations is given in Table 2 . Two physical layers, i.e. the 2.4 GHz band
layer and the 868/915 MHz band layer, are defined in the IEEE 802.15.4 standard. The worldwide
unlicensed 2.4 GHz band layer is characterized by higher data rates compared to the 868/915 MHz
band layer, which is due to higher frequencies and the digital modulation schema applied. While for
binary phase-shift keying (BPSK) two phases separated by 180° are used for digital modulation, offset
quadrature phase-shift keying (O-QPSK) uses four phases to modulate data for high data rates. The
offset in the modulation schema is applied for limiting large amplitude fluctuations undesired in
Table 1 Comparison of wireless standards [41, 44]
Bluetooth ZigBee Wi-Fi MQTT
Specification IEEE 802.15.1 Based on
IEEE 802.11a/b/g ISO/IEC 20922:2016
machine to machine
(M2M) and Internet
of Things (IoT)
contexts, devices at
remote locations, low
Star, tree, mesh
Line, ring, star,
tree, mesh (partially
(through bridging of
10 m 10 m to 20 m
(100 m in
100 m Depending on the
choice of low level
and data link layer)
8 active devices,
255 in park
> 65,000 Unlimited Unlimited
Table 2 IEEE 802.15.4:2003 frequency bands and data rates
Modulation Number of
868 868-868.6 Europe BPSK 1 20
915 902-928 United States BPSK 10 40
2450 2400-2483.5 Worldwide O-QPSK 16 (11 to 26) 250
4. A metamodel for describing communication-related information
In this section, the metamodel for describing cyber-physical systems is presented based on the results
of analyzing metamodeling approaches frequently used in computing in civil engineering (Section 2)
and the topologies and communication protocols used in in cyber-physical systems (Section 3).
Developing the metamodel is based on a formal analysis of communication processes. A model of
communication processes that dates back to the work of Shannon  is shown in the schematic
diagram in Figure 5. The schematic diagram shows the main elements of a communication system and
relationships between the elements. In general, communication systems include (i) transmitters, (ii)
transmission media, (iii) receivers, and (iv) messages, which are transformed into (v) electrical signals.
Messages are initiated by (vi) information sources (i.e. sensor nodes) making observations (i.e.
temperature, acceleration) and are processed at (vii) destinations of communication systems (i.e. other
sensor nodes, base stations, or computer systems). In addition, the model of Shannon describes effects
of noise that perturbates signals transmitted between a sending and a receiving device. Due to noise in
communication systems, signals sent by transmitters may differ from signals received by receivers.
Figure 5 Schematic diagram of a communication system .
Two processes involved in communication are encoding and decoding. Encoding on the transmitter
end of communication systems specifies the translation of messages into languages (or code), in
compliance with defined syntaxes and semantics. As a result, messages are transformed into
transmittable signals that can be received and understood by receivers provided with suitable methods
to decode incoming signals. The decoding process on receiver side of communication systems denotes
the re-translation of signals into messages that can be processed by destinations, such as data
management systems or software applications controlling actuators . In cyber-physical systems,
semantically and syntactically correct encoding and decoding of messages is ensured by
communication protocols, as introduced earlier.
The main elements and processes involved in communication within cyber-physical systems are
assembled as communication-related information in the metamodel, illustrated in terms of a UML
class model based on previous research the authors [5, 16] (Figure 6). The UML classes shaded in
gray represent elements relevant to communication-related information in cyber-physical systems
proposed in this study, while the total UML class model represents a metamodel for describing cyber-
physical systems overall.
Figure 6 Metamodel of cyber-physical systems applied for structural health monitoring and control
including communication-related information (shaded in gray).
The metamodel shown in Figure 6 distinguishes two main components of cyber-physical systems,
ComputerSystem and SensorNetwork. Computer systems provide resources and processes for data
management and data analysis. Sensor networks are composed of nodes realizing distinct tasks
(Node). Therefore, two types of nodes are distinguished, SensorNode and BaseStation. Sensor nodes
are responsible for data collection and data processing (Sensor). In addition, sensor nodes can control
actuators in response to events measured by cyber-physical systems (Actuator). Base stations realize
communication between sensor nodes and on-site computer systems. Both node types, sensor nodes
and base stations, are composed of PowerSupply, Resource, Process, and CommunicationUnit and
share common attributes, e.g. for specifying node location and node geometry (NodeSpecification).
According to communication capabilities of nodes, different network topologies are defined by
characteristic communication paths. A network topology is adapted with respect to the structural
system being monitored. To relate nodes to a physical component of a structural system, on-board
computation capacities of the sensor nodes may be used, for example, to integrate models describing
the structural response of a structure being monitored.
In the metamodel, communication units are described as aggregations, termed CommunicationUnit,
which are composed of several aggregates. The class DataUnit describes raw or preprocessed data to
be transmitted or received by a communication unit. The cardinalities in Figure 6 denote that a single
communication unit can send or receive a range of different data units. Data units are characterized by
attributes, such as type, to define, e.g., acceleration data or temperature data. The attribute domain
further specifies data either of time domain or frequency domain.
The classes Transmitter and Receiver are aggregates of the class CommunicationUnit and implement
methods to encode outgoing data units and to decode incoming signals into data units that can be
processed. Encoding and decoding is performed following the regulations from the communication
protocols applied. For prescribing regulations for encoding of messages and for decoding of signals,
the class Protocol is defined as an aggregate of the class CommunicationUnit. Following the OSI
reference model presented in Figure 4, a protocol stack may contain different communication
protocols. Due to the diversity of communication protocols, the class Protocol is defined as an abstract
class that inherits attributes and methods defined by specific protocols used in a cyber-physical
The transmission medium, through which the communication is realized, puts further restrictions on
the implementation of communication in cyber-physical systems and is thus be part of the metamodel.
The class TransmissionMedium inherits attributes and methods from specific media, such as cable,
radio, or Internet, i.e. hardware and communication protocols are chosen in dependence of the
With the semantic model presented, it is capable to describe monitoring-related information,
containing the subset of communication-related information, in the context of structural models of
cyber-physical systems. For preparing the validation of the BIM-based description approach towards
describing SHM and control systems, the semantic model is mapped into the IFC schema in the
5. BIM-based description of communication-related information using the metamodel
In this section, for describing cyber-physical systems on the basis of open BIM, an extension of the
IFC schema, standardized in ISO 16739:2013  is proposed enabling documentation and
optimization of cyber-physical systems applied for SHM and control. The focus is put on describing
communication-related information in cyber-physical systems. The IFC schema extension, building
upon the “IFC Monitor extension” proposed by the authors in , comprises two property sets,
presented in Table 3 and 4, for describing communication-related properties of the IFC entities
IfcDistributionSystem and IfcDistributionPort that are already standardized in IFC. In the remainder of
this section, upon introducing the extended IFC schema the proposed property set
Pset_DistributionSystemTypeCommunication to describe IfcDistributionSystem entities used for
communication systems is explained, followed by an illumination of the property set
Pset_DistributionPortTypeRadio. The radio enumerator described by Pset_DistributionPortTypeRadio
is added to the EXPRESS schema and specifies IfcDistributionPort entities. In Figure 7, existing IFC
entities (colored in white) and the proposed IFC extension used for describing communication-related
information (colored in gray) are shown. The semantics of the IFC schema extension originate from
the most abstract fundamental entity IfcRoot. IfcRoot is the common supertype of all IFC entities and
is described by the obligatory attribute GlobalId assigning a globally unique identifier to entities and
optional attributes, such as names and textual specifications. As can be seen from Figure 7, the entities
IfcObjectDefinition and the IfcRelationship inherit attributes of the entity IfcRoot and supplement
attributes in the IFC inheritance tree.
Figure 7 Extract of the extended IFC schema showing entities relevant to describing communication-
The IFC entities IfcCommunicationsAppliance and IfcDistributionPort shown in Figure 7 are of major
importance for describing communication-related information and are related to each other by two
objectified relationships, IfcRelConnectsPorts and IfcRelNests. Entities of type
IfcCommunicationsAppliance describe communication appliances for transmission and reception of
digital information, such as sensor data. Therefore, IfcCommunicationsAppliance entities are used to
describe communication units that are components of sensor nodes, base stations, and computer
systems. Inherited from IfcDistributionElement entities, IfcCommunicationsAppliance entities feature
an attribute termed HasPorts indicating the connection between IFC elements and ports being
components of a communication appliance. A port, in general, provides means for connecting an
element to other elements . To semantically describe IfcPort entities as parts of IfcElement
entities, such as IfcCommunicationsAppliance entities, the objectified relationship IfcNests is used.
The connection of exactly one port of an element to exactly one port of another element, e.g. for
describing communication between two sensor nodes or between two IfcCommunicationsAppliance
entities, is realized through the objectified relationship IfcRelConnectsPorts. In the IFC schema, one
subtype of IfcPort exists, which is termed “IfcDistributionPort”. The IfcDistributionPort, as shown in
detail in Figure 8, has inherited the attributes (i) ContainedIn, (ii) ConnectedFrom, and (iii)
ConnectedTo from IfcPort and is additionally specified by the attributes (i) SystemType, (ii)
FlowDirection, and (iii) PredefinedType. The attribute ContainedIn describes a port as a component of
a communication appliance, while the attributes ConnectedFrom and ConnectedTo define the
connection between ports of Ifc entities connected by the media exchanged between two ports in a
distribution system. To describe distribution systems interconnecting different IfcDistributionPort
entities of the same type in detail, the property set Pset_DistributionSystemTypeCommunication is
Figure 8 EXPRESS-G diagram of IFC entities and attributes for describing communication-related
Property set Pset_DistributionSystemTypeCommunication
In IFC, distribution systems are assigned to a specific function by assigning a value to the attribute
SystemType. The connectivity of ports within systems is restricted to ports of the same system type.
The enumeration of system types available in IFC (IfcDistributionSystemEnum) provides the
enumerators relevant to cyber-physical systems, such as DATA, SIGNAL, and COMMUNICATION.
Following the specifications of the current IFC schema , the DATA enumerator is applied for
networks of general-purpose usage and the enumerator SIGNAL describes systems for distributing
analog signals, such as modulated sensor data. The COMMUNICATION enumerator is listed without
any definition or specification. For specifying distributions systems, such as cyber-physical systems,
with a COMMUNICATION enumerator assigned, in Table 3 the property set
Pset_DistributionSystemTypeCommunication is shown. As can be seen from Table 3, using the
attributes CommunicationSystemType and CommSystemDescription of the property set proposed,
communication systems in cyber-physical systems are specified by types and can be described
Table 3 Property set Pset_DistributionSystemTypeCommunication for describing
IfcDistributionSystem entities of type COMMUNICATION
Name Type Description
CommunicationSystemType P_ENUMERATEDVALUE /
IfcLabel / PEnum_Distribution
Property enumerators are, e.g.,
Wi-Fi, ZigBee, MQTT and other
CommSystemDescription P_SINGLEVALUE / IfcLabel Qualitative description of the
Property set Pset_DistributionPortTypeRadio
For describing communication in cyber-physical systems using IfcDistributionSystem entities,
IfcDistributionPort entities of different system components are semantically connected. To define
transmitting ports and receiving ports of signals, the attribute FlowDirection, representing flow
directions at distribution ports, is used. Enumerators of the IfcFlowDirectionEnum enumeration are
SOURCE, SINK, SOURCEANDSINK, and NOTDEFINED. Ports, where communication signals are
modulated, have SOURCE enumerators assigned that correspond to transmitters shown in the
metamodel of communication-related information (Figure 6). Hence, receiving ports, denoted as
SINK, correspond to receivers in the metamodel and are responsible for demodulation of signals. The
enumerator SOURCEANDSINK can be applied to devices serving as transmitter and receiver, while
NOTDEFINED is used, when specific options are not applicable.
The IfcDistributionPortTypeEnum enumeration is applied for specifying port types, which limit the
compatibility of ports to distribution systems. In the current version of the IFC schema, distribution
port types are limited to the enumerators CABLE, CABLECARRIER, DUCT, PIPE, USERDEFINED,
and NOTDEFINED. To enhance the descriptive capacities of the enumeration with respect to cyber-
physical systems, the port type enumerator RADIO is added to the list of enumerators available. The
RADIO enumerator describes wireless connections between ports of different IFC elements for
distribution of modulated signals, or data, respectively. Figure 9 shows an extract of the extended IFC
schema, including the RADIO enumerator, written in EXPRESS.
Figure 9 Extract of the IFC schema extension showcasing entities relevant to describe communication
in cyber-physical systems.
To describe all types of distribution ports, the property set Pset_DistributionPortCommon containing
the properties PortNumber and ColorCode is used. Depending on the port type and on the type of the
distribution system, multiple property sets can be added to the IFC model of distribution systems
describing communication-related information in cyber-physical systems. In the current version of the
IFC schema, property sets for describing distribution ports of type cable, duct, and pipe exist, while
wireless ports, such as the RADIO port proposed in Figure 9, cannot be adequately described.
To enable IFC-based descriptions of wireless communication according to the metamodel introduced
in this study, the property set Pset_DistributionPortTypeRadio is shown in Table 4. The property set is
used for describing the wireless port type RADIO, taking into account communication protocols and
properties describing the protocols applied in cyber-physical systems.
Table 4 Property set Pset_DistributionPortTypeRadio for describing IfcDistributionPort entities of
Name Type Description
ConnectionType P_ENUMERATEDVALUE /
The physical port connection can be,
among others, Wi-Fi, Bluetooth,
ZigBee antennas defined in the
ConnectionSubtype P_SINGLEVALUE / IfcLabel Further specifications can be added
to connection types. For example:
Wi-Fi: IEEE 802.11a/b/g/n
Bluetooth: basic, enhanced data
rate, high speed, low energy
Protocols P_LISTVALUE / IfcIdentifier Listing of every communication
protocol in the protocol stack
ModulationSchema P_SINGLEVALUE / IfcLabel Signal modulation schema (e.g.
Frequency P_BOUNDEDVALUE /
Actual signal frequency and operable
frequency band (MHz) of the
physical layer (OSI reference model)
TransmissionMedium P_SINGLEVALUE / IfcLabel Medium (such as air) where
communication takes place
The extended IFC schema (Figure 9) is verified in a three-step verification procedure. Within the
verification procedure, which is illustrated in detail in , syntactic checks, semantic checks, and unit
tests are devised. The verification procedure is conducted using the test software of the official IFC
certification program [50, 51]. As a result of the verification procedure, the IFC schema extension is
positively verified, confirming the compliance with the current IFC schema. In the following section,
the IFC schema extension is validated by example modeling of a prototype cyber-physical system.
6. Validation of the metamodel
In the previous section, the metamodel has been mapped into the IFC schema, as materialized in terms
of a verified IFC schema extension. In this section, to validate the metamodeling approach, the IFC
schema extension is applied to describe communication-related information of a prototype cyber-
physical system installed on a laboratory test structure. The cyber-physical system is composed of (i)
two sensor nodes for structural health monitoring and control, (ii) an Internet-enabled computer
system, and (iii) a semi-active tuned liquid column damper (TLCD) to reduce the structural response
of the laboratory test structure. For test purposes, acceleration response is automatically recorded and
processed by the cyber-physical system to control the TLCD.
The laboratory test structure, a four-story shear frame, is modeled using a conventional BIM software
tool. As shown in Figure 10, the test structure is composed of five aluminum slabs of dimensions 300
mm × 200 mm × 15 mm (length × width × thickness) resting on four 20 mm × 2 mm aluminum
columns. The story height is 300 mm and the plate-to-column connections are fully fixed. The base
plate and columns are clamped on a solid block at the base of the structure. In the center of the top
story, the semi-active TLCD is installed, which is controlled by an actuator connected to a wireless
senor node. As shown in Figure 10, the second wireless sensor node has two acceleration sensors
attached via cable-based connections and is fixed in the middle of the third aluminum plate. The
acceleration sensors are fixed to the second and fourth story of the test structure. In laboratory tests,
the structure is subjected to free vibration induced by manual deflections of the top story of the test
Figure 10 Laboratory test structure and the cyber-physical system.
The wireless prototype cyber-physical system is composed of wireless sensor nodes of type Raspberry
Pi 3 Model B+ and connected to a computer system via Wi-Fi. The Raspberry Pi sensor nodes possess
a system on chip (SoC) of type Broadcom BCM2837B0 including a quad-core Cortex-A53 (ARMv8)
64-bit processor running at 1.4 GHz . The two Raspberry Pi nodes are connected to a computer
system in star topology. The node responsible for monitoring acceleration is connected to two 3-axis
ADXL345 acceleration sensors via two GPIO signal pins . The second Raspberry Pi node serves
as an actuator connected to the semi-active TLCD controlling an electrical valve. In terms of wireless
communication, two specifications are implemented into the Raspberry Pi 3 Model B+, (i) 2.4 GHz
and 5GHz IEEE 802.11.b/g/n/ac wireless LAN and (ii) Bluetooth Low Energy 4.2. In the prototype
cyber-physical system, sensor data is communicated between the sensor nodes and the computer
system using a Wi-Fi connection. The computer system provides remote access to the cyber-physical
system through an Internet connection.
Communication-related information of the prototype cyber-physical system is summarized in an UML
object diagram shown in Figure 11 serving as a digital representation of the real-world prototype
cyber-physical system. The UML object diagram is derived from the metamodel proposed in this
study and, more precisely, from the communication-related information described by the metamodel
Figure 11 Object diagram of the prototype cyber-physical system.
The cyber-physical system is shown in the top of Figure 11 (CPS), composed of a Raspberry Pi-based
sensor network (RPNetwork) and a computer system (Computer). The computer system provides
resources for data storage and data management. Both the computer system and the sensor network
have communication units equivalent to type CommunicationUnit described in the metamodel. The
sensor network is composed of two nodes RP#1 and RP#2, featuring communication units including
transmitters and receivers that take advantage of Wi-Fi communication protocols. The computer is an
instance of the class ComputerSystem featuring a communication unit with transmitters and receivers.
Sensor data exchanged between communicating system components is described by the instance
“SensorData” of class “DataUnit” and is part of every communication unit of the cyber-physical
While the structure of the prototype cyber-physical system is described in Figure 11, the behavior of
the system with regard to communication is visualized in the sequence diagram shown in Figure 12,
which exemplarily shows the communication between sensor node RP#1 and the computer system.
Sensor node RP#1 is applied for measuring acceleration response and RP#2, the topmost node of the
cyber-physical system, is connected to the semi-active TLCD. Communication unit CU#1, connected
to RP#1, encodes sensor data following the syntax of Wi-Fi communication protocols and transmits
data units to the communication unit CUC of the computer system. The computer decodes the data
units and generates control sequences to be forwarded to RP#2 for controlling the TLCD. Encoding of
structural control sequences is performed by communication unit CUC of the computer system
according to the Wi-Fi protocol. Structural control sequences are sent to communication unit CU#2 of
RP#2 that has the actuator attached.
Figure 12 Sequence diagram of communication processes between the Raspberry Pi nodes and the
The IFC-compliant representation of communication-related information in cyber-physical systems,
corresponding to the object-oriented description illustrated above, is shown in Figure 13 in form of a
BIM model. In Figure 13, an extract of the BIM model is listed using the Standard for the Exchange of
Product Model Data (STEP) . The listing exemplarily illustrates distinct components and
properties of the communication unit of sensor node RP#1 to demonstrate the descriptive capacities of
the IFC property sets representing critical elements to map the metamodel into the IFC schema
extension. In the listing, the communication unit of RP#1 is described by an
IfcCommunicationsAppliance entity featuring an IfcDistributionPort entity. The entities are related by
the objectified relationship IfcRelNests and specified by the property set
Pset_DistributionPortTypeRadio including the properties shown in lines #134 to #139.
Figure 13 Extract of the BIM model in STEP format describing communication-related information of
sensor node RP#1
While the structural components of the prototype cyber-physical system are modeled using a
conventional BIM tool (Figure 14), monitoring-related information and, in particular, communication-
related information is added to the BIM model by manual postprocessing. The manual postprocessing
is implemented using the APSTEX IFC framework for manipulating IFC-based BIM models
Figure 14 BIM model of the laboratory test structure and the cyber-physical system.
As elucidated by describing and implementing the laboratory test structure and the cyber-physical
system, the metamodel developed to describe cyber-physical systems is suitable for real-world SHM
and control applications, such as BIM-based documentation and maintenance of information about
cyber-physical systems. Using the metamodel, structural semantics as well as behavioral semantics of
communication-related information in cyber-physical systems can be described. With the metamodel,
planning of SHM and control systems can be optimized, data control and management can be
improved, and SHM and control system components can be linked to communication-related
information. As a result, the metamodeling concept applied to communication-related information in
the prototype cyber-physical system demonstrated in this study can be used to formally describe a
variety of systems and communication processes based on building information modeling in
compliance with the IFC standard.
7. Summary and conclusions
In this study, metamodeling approaches have been analyzed with respect to capabilities of describing
cyber-physical systems applied for SHM and control of structures. The focus has been emphasized on
the formal description of communication processes in cyber-physical systems, referred to as
“communication-related information”. A cyber-physical systems metamodel, capable of describing
communication-related information, has been proposed, based on the analysis of three metamodeling
approaches that are applied in a broad range of applications, (i) UML and other OMG standards, (ii)
the SWE framework, and (iii) the EXPRESS metamodel.
UML and other OMG specifications are adaptable to a wide scope of applications. Models derived
from MOF-compliant modeling languages are compliant to the principles of metamodeling, which are
particularly advanced by OMG’s MDA. OGC specifications forming the SWE framework, proposed
to make sensor data better accessible and usable via the Internet, are related to UML by reusing UML
notations. EXPRESS and EXPRESS-G are of importance in architecture, engineering and construction
for open-source building information modeling and, thus, have been under consideration in this study,
representing the third metamodeling approach. Limitations of the modeling capacities of EXPRESS,
of EXPRESS-G, and of the SWE framework have been found with respect to describing
communication-related information. In particular, models developed using EXPRESS-G are restricted
to descriptions of structural semantics. As for the SWE framework, the UML notations used to
describe sensor networks exclude communication-related information. Rather, the description of
communication-related information is restricted to a non-exhaustive XML encoding without graphical
representation. In consequence, UML has been found the most general metamodel, which is adaptable
to other modeling languages. Hence, UML has been taken as a basis to develop the metamodel for
cyber-physical systems on a well-defined, formal basis.
Based on background information on network topologies and on theory on network communication,
communication-related information in cyber-physical systems has been semantically described in the
technology-independent metamodel. As cyber-physical systems are gaining importance in SHM and
control of civil structures and advancements in describing building information using the BIM-
compliant IFC standard are still ongoing, the metamodel is mapped into the IFC schema. An IFC
schema extension has been proposed, because the current descriptive capacities of the IFC schema are
not sufficient to fully describe cyber-physical systems. For verification of the IFC schema extension,
test software used in the official IFC certification program has been emploid confirming the
compliance of the IFC schema extension with the current IFC schema. Subsequently, the metamodel
proposed in this study has been validated by implementing a prototype cyber-physical system based on
the metamodel and using the extended IFC schema to describe communication-related information in
the cyber-physical system for SHM and control.
The results have demonstrated that the metamodel proposed in this study is suitable for describing
cyber-physical systems and specifically communication-related information of cyber-physical
systems, thus providing a formal basis for documenting and optimizing cyber-physical systems. In
future work, both the metamodel and the IFC schema extension may be enhanced by formal semantic
representations of sensor data and control sequences, which cannot yet be adequately modeled using
the IFC standard. In addition, the documentation capacities of the metamodel may be extended from
cyber-physical systems for SHM and control towards other engineering domains, such as
environmental monitoring applications and other applications in the field of the Industrial Internet of
The authors would like express their sincere appreciation to theGerman Research Foundation (DFG)
for supporting this research through grant SM 281/7-1. Major parts of this work have been conducted
in the “Structural Health Monitoring Laboratory”, sponsored by the European Union through the
European Fund for Regional Development (EFRD) and the Thuringian Ministry for Economic Affairs,
Science and Digital Society (TMWWDG) under grant 2016 FGI 0009, whose support is gratefully
acknowledged. Any opinions, findings, conclusions, or recommendations expressed in this paper are
those of the authors and do not necessarily reflect the views of DFG, EFRD, or TMWWDG.
 Legatiuk, D., Theiler, M., Dragos, K. & Smarsly, K., 2017. A categorical approach towards
metamodeling cyber-physical systems. In: The 11th International Workshop on Structural Health
Monitoring (IWSHM). Stanford, CA, USA, 12.09.2017.
 Shi, J., Wan, J., Yan, H. & Suo, H., 2011. A survey of cyber-physical systems. In: Proceedings of
the 2011 International Conference on Wireless Communications and Signal Processing (WCSP).
Nanjing, China, 09.11.2011.
 Lee, J., Bagheri, B. & Kao, H., 2015. A cyber-physical systems architecture for industry 4.0-
based manufacturing systems. Manufacturing Letters, 3: 18–23.
 Abdo, M. A.-B., 2014. Structural health monitoring: History, Applications and future. A review
book, Open Science Publishers, NY, USA.
 Theiler, M. & Smarsly, K., 2018. IFC Monitor – An IFC extension for modeling structural health
monitoring systems. Advanced Engineering Informatics, 37: 54-65.
 Smarsly, K. & Tauscher, E., 2016. Monitoring information modeling for semantic mapping of
structural health monitoring systems. In: The 16th International Conference on Computing in
Civil and Building Engineering (ICCCBE). Osaka, Japan, 06.07.2016.
 The Canadian network of centres of excellence on intelligent sensing for innovative structures
(ISIS Canada), 2001. Guidelines for structural health monitoring. Design manual no. 2.,
University of Manitoba, Canada.
 Smarsly, K., Law, K. H. & König, M., 2011. Resource-efficient wireless monitoring based on
mobile agent migration. In: Kundu, T. (ed). Proceedings of the SPIE (Vol. 7984): Health
Monitoring of Structural and Biological Systems 2011. San Diego, CA, USA, 06.03.2011.
 Daum W., 2013. Principles, concepts and assessment of structural health monitoring. In: Czichos
H. (ed.) Handbook of technical diagnostics. Springer, Heidelberg, Germany.
 Clark, T., Sammut, P. & Willans, J., 2015. Applied metamodelling. A foundation for language
driven development. NY, USA.
 Legatiuk, D., Dragos, K. & Smarsly, K., 2017. Modeling and evaluation of cyber-physical
systems in civil engineering. Proceedings in Applied Mathematics and Mechanics, 17(1), S. 807-
 U.S. General Services Administration (GSA). 2007. BIM guide overview. BIM guide series 01.
In: The national 3D-4D-BIM program. Washington, DC, USA. Available at:
guides/bim-guide-01-bim-overview (Accessed on: 30.10.2018).
 State of Wisconsin, Department of Administration, 2012. Building information modeling (BIM).
Guidelines and standards for architects and engineers. Available at:
https://doa.wi.gov/Pages/DoingBusiness/BIM.aspx (Accessed on: 30.10.2018).
 Cabinet Office and Infrastructure and Projects Authority, 2016. Government Construction
Strategy 2016-20. London, UK. Available at:
(Accessed on: 30.10.2018).
 German Federal Ministry of Transport and Digital Infrastructure (BMVI), 2015. Stufenplan
digitales Planen und Bauen. Einführung moderner, IT-gestützter Prozesse und Technologien bei
Planung, Bau und Betrieb von Bauwerken. (Phase plan for digital planning and construction.
Implementation of modern, IT-based processes and technologies for planning, construction, and
maintenance of civil structures). Berlin, Germany. Available at: https://www.bmvi.de/SharedDocs
/DE/Publikationen/DG/stufenplan-digitales-bauen.pdf?__blob=publicationFile (Accessed on:
 Theiler, M., Dragos, K. & Smarsly, K., 2018. Semantic description of structural health
monitoring algorithms using building information modeling. In: The 25th International Workshop
on Intelligent Computing in Engineering (EG-ICE). Lausanne, Schweiz, 10.06.2018.
 Favre, L., 2010. Model driven architecture for reverse engineering technologies: Strategic
directions and system evolution. Engineering Science Reference, Hershey, NY, USA.
 Object Management Group, 2014. Model Driven Architecture (MDA). MDA Guide rev. 2.0,
OMG Document, Needham, MA, USA.
 Object Management Group, 2014. MDA specifications, Needham, MA, USA. Available at:
https://www.omg.org/cgi-bin/doc?ormsc/14-06-01 (Accessed on: 08.05.2018).
 Object Management Group, 2016. OMG Meta Object Facility (MOF) Core Specification. Version
2.5.1. Needham, MA, USA.
 Object Management Group, 2017. OMG® Unified Modeling Language® (OMG UML®).
Version 2.5.1, Needham, MA, USA. Available at. http://www.omg.org/spec/UML/2.5.1
(Accessed on. 30.05.2018).
 Open Geospatial Consortium, 2012. OGC® Sensor Observation Service Interface Standard.
Available at: http://www.opengis.net/doc/IS/SOS/2.0 (Accessed on: 10.05.2018).
 Open Geospatial Consortium, 2013. OGC Abstract Specification. Geographic information –
Observations and measurements. Available at: http://portal.opengeospatial.org/files /?artifact
_id=41579 (Accessed on: 03.07.2018).
 Open Geospatial Consortium, 2018. Sensor Web Enablement (SWE). Available at:
http://www.opengeospatial.org/ogc/markets-technologies/swe (Accessed on: 04.07.2018).
 Open Geospatial Consortium, 2011. OGC® SWE Common Data Model Encoding Standard.
Available at: http://portal.opengeospatial.org/files/?artifact_id=41157 (Accessed on: 03.07.2018).
 Open Geospatial Consortium, 2011. OpenGIS® SWE Service Model Implementation Standard.
Available at: http://portal.opengeospatial.org/files/?artifact_id=38476 (Accessed on: 03.07.2018).
 Open Geospatial Consortium, 2014. OGC® SensorML: Model and XML Encoding Standard.
Available at: https://portal.opengeospatial.org/files/?artifact_id=55939 (Accessed on:
 International Organization for Standardization (ISO), 2008. ISO 10303-11:2004. Industrial
automation systems and integration -- Product data representation and exchange -- Part 11:
Description methods: The EXPRESS language reference manual. Geneva, Switzerland.
 International Organization for Standardization (ISO). 2013. “ISO 16739:2013 – Industry
Foundation Classes (IFC) for data sharing in the construction and facility management
industries”. Geneva, Switzerland.
 Anderl, R. & Trippner, D. (eds.), 2000. Standard for the Exchange of Product Model Data. An
introduction to development, implementation, and industrial application of ISO 10303 (STEP).
(Eine Einführung in die Entwicklung, Implementierung und industrielle Nutzung der
Normenreihe ISO 10303 (STEP). B. G. Teubner, Stuttgart, Germany.
 Sharma, D., Verma, S. & Sharma, K., 2013. Network topologies in wireless sensor networks: A
review. International Journal of Electronics & Communication Technology (IJECT), 4(3):93-97.
 Serpanos, D. & Wolf, T., 2011. Network protocols and network systems. In: Architecture of
network systems. Elsevier, Burlington, MA, USA.
 International Organization for Standardization (ISO), 1994. ISO/IEC 7498-1:1994. Information
technology -- Open Systems Interconnection -- Basic Reference Model: The Basic Model,
 Mandl, P., 2018. TCP and UDP internals. Protocols and programming. (TCP und UDP Internals.
Protokolle und Programmierung.), Wiesbaden, Germany.
 Serpanos, D. & Wolf, T., 2011. The layered Internet architecture and network protocols. In:
Architecture of network systems. Elsevier, Burlington, MA, USA.
 Fahmy, H. M. A., 2016. Wireless sensor networks. Concepts, applications, experimentation and
analysis. Springer, Singapore.
 Sadeghi, A.-R., Wachsmann, C. & Waider, M., 2015. Security and privacy challenges in
industrial Internet of Things. In: The 52nd ACM/EDAC/IEEE Design Automation Conference
(DAC). San Francisco, CA, USA, 08.06.2015.
 Finster, S. & Haas, C., 2018. Best practices OPC UA security. In need of protection
(Schutzbedürftig). iX Magazin für professionelle Informationstechnik, iX Special 2018 –
Industrial Internet of Things: 118-121.
 Obermaier, D., 2018. Safe application of MQTT (MQTT sicher einsetzen). Rundum abgeriegelt.
iX Magazin für professionelle Informationstechnik, iX Special 2018 – Industrial Internet of
 Institute of Electrical and Electronics Engineers (IEEE). 2014. 802-2014 - IEEE standard for local
and metropolitan area networks: Overview and architecture.
 NXP laboratories UK, 2013. Co-existence of IEEE 802.15.4 at 2.4 GHz. Application note.
Available at: https://www.nxp.com/docs/en/application-note/JN-AN-1079.pdf (Accessed on:
 Shannon, C.E., 1948. A mathematical theory of communication. The Bell System Technical
Journal, 27(3): 379–423.
 McQuail, D. & Windahl, S., 1993. Communication models for the study of mass
communications. Second edition. Pearson Education Limited, Essex, England.
 buildingSMART International Limited. 2016. IFC4 specification. Industry Foundation Classes.
Version 4 – Addendum 2. Available at: http://www.buildingsmart-tech.org/ifc/IFC4/Add2/html/
(Accessed on: 24.10.2018).
 Tauscher, E. & Theiler, M.: APSTEX IFC Framework. Online: www.apstex.com, Weimar,
Germany (Accessed on: 15.11.2018).
 BuildingSMART International, iabi, AEC3 and APSTEX. 2017. B-Cert – The official platform
for the buildingSMART IFC4 certification, http://www-b-cert.org, Berlin, Munich, and Weimar,
Germany (Accessed on: 15.11.2018).
 Raspberry Pi Foundation, 2017. Raspberry Pi 3 Model B+. (Product data sheet). Available at:
(Accessed on: 29.10.2018).
 Analog Devices, Inc., 2009. Digital Accelerometer. ADXL345. (Product data sheet). Norwood,
MA, USA. Available at: file:///D:/3_Journal%20paper%20metamodeling/datasheet_sensor_
ADXL345.pdf (Accessed on: 29.10.2018).
 International Organization for Standardization (ISO), 2016. ISO 10303-21:2016. Industrial
automation systems and integration – Product data representation and exchange – Part 21:
Implementation methods: Clear text encoding of the exchange structure. Geneva, Switzerland.
 Martínez-Castro, R.E. & Jang, S., 2018: Structural Cyber-Physical Systems: A Confluence of
Structural Health Monitoring and Control Technologies (2018). CEE Articles, 2. Available at:
https://opencommons.uconn.edu/cee_articles/2 (Accessed on: 16.05.2019).
 Toshiba Elevator and Building Systems Corporation, 2019. Taipei 101 - 1,010 meters per minute
Ultra High Speed Elevator. Available at: https://www.toshiba-
elevator.co.jp/elv/infoeng/projects/taipei/02.html (Accessed on: 16.05.2019).