Attack Simulation on Data Distribution Service-based Infrastructure

System

Basem Al-Madani

, Hawazen Alzahrani

and Farouq Aliyu

Department of Computer Engineering, King Fahad University of Petroleum and Minerals, Dhahran, Saudi Arabia

Center for Excellence in Development of Non-Profit Organization, KFUPM, Dhahran, Saudi Arabia

Keywords: Data Distribution Service, Quality of Service Policy, QoS, Security, Middleware, Real-Time, Critical

Systems, Iiot.

Abstract: Data Distributed Service (DDS) is a widely used publish-subscribe-based middleware protocol for real-time

machine-to-machine communication. Many critical infrastructure systems employ DDS for Real-Time

applications. These DDS-based systems must operate effectively. This study examined the possibility of

manipulating or improperly configuring DDS to facilitate malicious activities. A client-side attack on a DDS-

based system and its consequences were the main topics of the research since DDS systems are isolated from

other networks and external users. We investigated two security flaws in DDS in an isolated environment to

show how they could be employed to compromise a DDS feature. The manipulation of QoS policy

configurations in the DDS system demonstrated that it has become more secure than the early versions.

1 INTRODUCTION

The Data Distribution Service (DDS) powers systems

like; autonomous vehicles, airports, robots, and

military tanks, among others. It has been around for

ten years, and its use is constantly growing. DDS is a

middleware application that supports the creation of

middleware layers for machine-to-machine

communication. It follows the publish-subscribe

model. This software is essential for embedded

devices or applications that need real-time

functionality (OMG, 2015).

DDS is a reliable communication layer between

sensors, controllers, and actuators. It is maintained by

the Object Management Group (OMG) and utilized in

many critical applications. DDS is for peer-to-peer

and publish-subscribe applications because most

services cannot tolerate a single point of failure. The

middleware works on multicast for discovery,

enabling everything to function without any initial

setups (DDS Foundation, 2020).

Generally, the DDS layer is (encapsulated) in the

Real-time publish-subscribe (RTPS) packets. In other

words, each DDS implementation delivers its own

https://orcid.org/0000-0002-5875-154X

https://orcid.org/0000-0000-0000-0000

https://orcid.org/0000-0003-4481-8851

RTPS implementation. The DDS relies on the RTPS,

which is a lower-layer standard protocol. DDS is a

data-centric communication protocol that enables

developers to create a flexible shared data domain for

any application requiring two or more nodes to

exchange typed data. DDS and its layers are not

available as an off-the-shelf product; it is a library.

Developers use to create custom middleware

protocols with advanced functionality like custom

data types, QoS policies, network partitioning, and

authentication (DDS Foundation, 2019; Object

Management Group, 2022).

DDS's mechanisms give developers greater

flexibility when building their systems (OMG, 2018).

However, it introduces potential security issues that

attackers may exploit. DDS is a prime target for

attackers since it lies at the earliest stage of the

software supply chain. Thus, it is simple to lose track

of it (ENISA, 2021). Demonstrating an exploit's

effects is not as straightforward. The requirements,

priorities, and operational conditions will vary

depending on the vertical, making it challenging to

develop representative attack scenarios. In addition,

accessing testbed devices for aggressive research is

162

Al-Madani, B., Alzahrani, H. and Aliyu, F.

Attack Simulation on Data Distribution Service-based Infrastructure System.

DOI: 10.5220/0011956200003482

In Proceedings of the 8th International Conference on Internet of Things, Big Data and Security (IoTBDS 2023), pages 162-169

ISBN: 978-989-758-643-9; ISSN: 2184-4976

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

restricted due to the significance of the systems where

DDS is employed (Maggi, 2022).

This research aims to examine and demonstrate

the effect of tampering or misuse attacks on the

software configuration of DDS systems. The rest of

the paper is organized as follows: Section 2 gives a

background of the DDS model - its architecture and

security implementation. Section 3 presents our

problem statement. Section 4 comprises of literature

review on M2M protocols and the implementation of

security in DDS. The proposed work and simulation

setup is in Section 5, followed by experiment works

conducted for this research in Section 6. Section 7

provides an analysis and discussion. Lastly, the

conclusion and future work are in Section 8.

2 BACKGROUNDS

2.1 Data Distribution Service

OMG developed the DDS (DDS Foundation, 2019)

as a publish-subscribe middleware standard for

distributed systems where participants exchange

information in a DDS Domain. A Publisher and a

Subscriber can write or read in Domain, respectively.

The DDS standard allows an efficient, high-

performance, and interoperable data share for

mission-critical real-time systems. DDS provides

QoS policies and protocols that enable applications to

establish communication without; specifying the

underlying network architecture, changing publisher

and subscriber membership, and intermittent

connectivity (DDS Foundation, 2020).

The OMG DDS specification comprises two

layers: the DDSI (DDS Interoperability Wire

Protocol) layer, which describes the layer above

network transport, and the DDS layer, which provides

Data-Centric Publish-Subscribe (DCPS) application

communication behavior with each other and defines

QoS policies (Object Management Group, 2022).

2.2 DDS in Real-Time Systems

Networking protocols and hardware increasingly

become heterogeneous, and distributed computer

systems become increasingly dynamic, which require

loosely connected infrastructures. Traditional

software systems do not address these concerns are

not adequately.

DDS for real-time systems enables timely and

reliable collection of data and transmission via a

publish-subscribe method that supports dynamic

node discovery, topic-based data distribution, and

data stream time-space decoupling (Object

Management Group, 2022). A real-time system

consists of many networked devices such as

platforms, sensors, actuators, and services to connect,

which requires a middleware to manage production

processes, communication, and decision-making.

One of the implementations of DDS is RTI

Connext DDS. It finds application in; mission-

critical, ATC, SCADA, C2 systems, and machinery

control.

2.3 DDS Security

DDS security design avoids Man-in-the-Middle

attacks and ensures only members with permissions

can publish, subscribe, or access the data. Security

standards include Authentication, Access Control,

Cryptography, Logging, and Data Tagging. These

functionalities are in the plugins (OMG, 2018). DDS

gives developers maximum flexibility in building

DDS-based systems. Therefore, the DSS makes a low

effort to protect the software and its configuration on

the host from tampering. Protecting the host

containing DDS applications depends on physical

security and network isolation from other systems and

external users.

Security issues are investigated by identifying

potential threats and associated attack vectors, then

examined DDS-based systems to see if they are

vulnerable to the attacks.

3 PROBLEM STATEMENT

DDS middleware gives developers flexibility when

designing and building their systems, for example,

specifying the communication details between

applications (OMG, 2015; OMG, 2018). A published

application does not need to know where the data is

going, while a subscribed application does not need

to know how or from where the data comes. This

model and the complexity of DDS exacerbate its

security issues (Abaimov, 2021). These security

issues include ease of access to all data, the loss of

control over data routing and communication paths,

and the opportunity for entities to join the network.

Thus, developing methods for identifying

malicious activities within a DDS-based system is

crucial. However, access to testbeds for offensive

cybersecurity research on DDS-based applications is

difficult because of the significance of where the

middleware is employed. The first step in defending

against a cyberattack on a DDS-based critical system

is understanding how DDS can be compromised.

Attack Simulation on Data Distribution Service-based Infrastructure System

163

4 LITERATURE REVIEW

This section introduced previous studies related to

this paper in two subsections. The first one is the

limitation in middleware for industrial

communications and IIoT. The second subsection

discusses the previous works that investigated DDS

security.

4.1 M2M Protocols

Machine-to-Machine M2M communication is the

interaction and data exchange between two or more

interconnected machines, which enables devices like

smartphones, laptops, factory equipment, robots, and

autonomous sensors to react and modify their internal

processes based on external feedback.

Cybersecurity in the automated production

industry mainly focuses on securing organizational

and operational boundaries, such as preventing illegal

access to the industrial network. Several messaging

communication protocols are available for industrial

communications and IIoT, including; MQTT, OPC

UA, CoAP, SECS/GEM, DDS, and others. Many of

these protocols only offer basic security measures.

The following subsections provide a brief explanation

of the security limitation in the mentioned M2M

protocols.

4.1.1 MQTT

Message Queuing Telemetry Transport (MQTT) is a

resource-constrained communications protocol

interface paradigm. It works effectively in networks

where bandwidth requirements must be minimum. It

is best suited for machine-to-machine (M2M)

communications.

But since MQTT is for lightweight applications, it

neither encrypts the header nor the payload. Instead,

it transfers data in plaintext, which is insecure. As a

result, encryption must be employed as a separate

function, such as by Transport Layer Security (TLS),

increasing the computational overhead for resource-

constrained devices (Chen, 2020; Patel, 2020).

4.1.2 OPC-UA

The Open Platform Communications Unified

Architecture (OPC-UA) is an M2M communication

protocol designed by OPC Foundation for Industrial

Automation. OPC-UA provides security. It includes a

variety of security characteristics, such as

authentication, integrity, and confidentiality.

Several protection levels include; no security,

integrity only, integrity and confidentiality, and

security protocols that define encryption and

signature cryptographic techniques for secured

communication. Generally, OPC-UA provides

industrial automation with robust security

characteristics (Mühlbauer, 2020).

4.1.3 CoAP

The Constrained Application Protocol (CoAP) is an

Internet Application Protocol developed for resource-

constrained devices and networks (Iglesias, 2019).

CoAP uses DTLS instead of TLS as a security

solution because it runs on the UDP transport layer

protocol. It also encodes messages in a simple binary

format to provide a lightweight reliability

mechanism.

Some of the features by which CoAP keeps the

connection and prevents link termination in the event

of missed or out-of-order packets. Also, it has an

additional feature that prevents DoS attacks by

requiring a server to issue a verification query to

check the source address's authenticity (Rathod,

2017).

4.1.4 SECS/GEM

The SECS/GEM Protocol is a group of connectivity

standards developed by the Semiconductor

Equipment Materials Initiative (SEMI) (Laghari,

2021). The SECS/GEM protocol is a widely used

global industry standard. The transport

communication protocol used by SECS/GEM is

called High-Speed SECS Message Services (HSMS);

it uses TCP.

The SECS/GEM protocol has a weak security

mechanism and provides no security features to

connect to other network entities. In addition, the host

computer and factory equipment exchange messages

with no integrity verification. Hence, it creates

potential for cyberattacks such as denial-of-service

attacks, data tampering attacks, and spoofing attacks.

4.1.5 DDS

The Data Distribution Service DDS is a publish-

subscribe, data-centric middleware for real-time

systems developed by OMG (DDS Foundation,

2019). The OMG's DDS security requirements

provide a robust security architecture suitable for IoT

devices. Also, it can work on both UDP and TCP.

Given that devices participate in DDS in a distributed

manner and that there are numerous brokers, there

isn't a single point of failure, making DDS more

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

164

resilient and reliable, ensuring system availability

(Friesen, 2020).

The main risk to DDS systems is insider attacks.

These are attacks perpetrated by individuals who have

already been compromised (Michaud, 2018).

4.2 DDS Security Research

Researchers have conducted several investigations on

DDS security. However, the research has mainly

focused on how to enhance DDS's capabilities to meet

customer requirements. OMG has created a new DDS

Security specification to address DDS security

challenges on a large scale and consistently (OMG,

2018). However, it has not addressed any security

issues that could result from a hacked host. As a

result, even DDS-based systems built with the new

DDS Security features are yet vulnerable to client-

side attacks.

White et al. (2019) proposed a method for

performing passive network reconnaissance on

systems that rely on Secure DDS. The authors execute

vulnerability excavation offline without actively

engaging the targeted system. It opens the system for

targeted attacks such as selective denial of service,

adversarial databus segmentation, or vendor

implementation vulnerability excavation.

Friesen et al. (2020) investigated the security of

the RTPS and DCPS, along with the TLS and DTLS

protocols. The DDS Security Plugins are as follows:

Authentication Service Plugin, Logging Service

Plugin, Access Control Service Plugin, and

Cryptographic Service Plugin.

Michaud et al. (2018) this study examined the

DDS standard and vendor implementations for

potential security flaws before demonstrating and

validating five malicious attacks that took advantage

of the DDS systems' weaknesses using RTI Connext

DDS software. The following are the five attacks:

• Anonymous Subscribe and Republish

Functionality: It required only knowledge of

the transmitted data type and the Ownership

QoS policy.

• OWNERSHIP STRENGTH QoS Policy and

EXCLUSIVE Data Ownership: It allows the

Data Writer with the highest OWNERSHIP

STRENGTH parameter to send data samples

when the Ownership QoS Policy is

EXCLUSIVE.

• OWNERSHIP KIND QoS Policy and

SHARED Data Ownership: For this attack, the

attacker updates the Publisher's Ownership

QoS policy to differ from the original

Subscriber. The malicious Subscriber then

infiltrates the network with the same

Ownership QoS policy.

• LIFESPAN QoS Policy Causing Immediate

Data Expiration: This attack prevents a

Subscriber from receiving data samples.

• LocatorList Environment Variable Causing

Participant Discovery Domain Misdirection:

This attack takes advantage of this

environment variable which determines the IP

address that belongs to the DDS entities.

Abaimov et al. (2021) presented an empirical

study on the simulation of three types of attacks,

Malicious Publisher, Malicious Subscribe, and Clone

on DDS. Then, the authors use Deep Learning to

detect them. The result showed that Deep Learning

detects all the simulated attacks using metadata

analysis. However, advanced attacks are more

difficult to detect.

Park et al. (2021) evaluated DDS. They used

Kerberos to authenticate all Publishers and

Subscribers by creating tickets. They utilize ROS 2 as

a testbed. The results showed that DDS-C improves

DDS security by blocking impersonation attacks.

Kim et al. (2021) proposed the ABAC-based

security model for DDS and its execution. The model

enhances the authorization of participation and nodes

in the RTPS discovery mechanism. The results reveal

that the approach meets high-tier time criteria in the

healthcare and electricity domains.

Maggi et al. (2022) launched an attack against an

autonomous-driving mobile robot simulation and

then a physical one in a controlled environment. They

tested the vulnerabilities in ROS 2 that affect the

design and implementations of DDS by DoS attacks.

The result shows that the RTI Connext DDS node

crashes and causes the ROS 2 node to crash too.

5 PROPOSED WORK AND

SIMULATION SETUP

DDS systems find application in areas physically

isolated from other networks or systems; and are

access-controlled (Michaud et al., 2018). The main

risk would most likely come from a host compromise

created by someone with a basic understanding of the

system's use, architecture, and implementation. This

research assumed that the attacker had accessed the

DDS system through a software environment or

physical system. Thus, we concentrate on the attack

action to exploit DDS flexibility to customize the

QoS.

Attack Simulation on Data Distribution Service-based Infrastructure System

165

5.1 Selection of Security Issues and

Attack Methods

The analysis of DDS security gaps (Michaud, 2017)

revealed 60 security flaws in four areas RTPS, DCPS,

configuration mechanism, and Transport mechanism.

We selected two of these security issues for modeling

and demonstration. The issues are in the DCPS QoS

Policy area: Misuse of RELIABILITY and

RESOURCE_LIMITS DCPS QoS Policies and

Misuse of DEADLINE and TIME_BASED_FILTER.

The selection of these issues covers Data Hijacking

attacks.

Table 1: DDS QoS Policies.

QoS Policy Purpose Conflicting

RELIABILITY

Indicates the level of

reliability offered or

requested

depth <=

max_samples_per

_instance

<= max_samples

RESOURCE

LIMITS

Specifies the

resources that the

Service can

consume.

HISTORY

Specifies how the

service will buffer

sen

data

DEADLINE

Specifies maximum

waiting time

deadline >=

time_based_filter

TIME BASED

FILTER

Specifies minimum

separation time

The DDS DCPS specification specifies Quality of

Service (QoS) policies, which outline the minimal

service levels necessary for communication between

DCPS entities and place restrictions on the shared

data. This work focuses on the following QoS policies

and their conflict: Ownership, Resource-limits,

Reliability, Deadline, and Time-based-Filter. Table 1

describes them in detail.

5.2 DDS Demonstration and Validation

Environment

The following setup was built in an isolated

environment to model each DDS security flaw and

attack method. The environment consists of; a Virtual

Box (to virtualize the network environment), an

Ubuntu operation system (as a host environment for

DDS software), RTI Connext DDS Shapes Demo

application v 6.1.1 (to demonstrate the attack

methods), and a Microsoft Windows operating

system.

6 EXPERIMENT WORK

The demonstration environment of DDS security

issues applied on The RTI Connext DDS Shapes

Demo application. The Shapes Demo shows how

various QoS policies impact data flow by providing

simple methods for identifying unique data types as

shapes with different features (RTI Shapes Demo,

2019). The Shapes Demo can play the role of both a

publisher and a subscriber to simulate DDS Entities

on the hosts.

Figure 1: DDS entities interaction.

The domain participants define the QoS for

entities: Topic, Publisher, DataWriter, Subscriber,

and DataReader. The Publisher specifies and offers

the QoS policies to all Subscribers; the Subscribers

request the set of QoS policies they require, while the

DCPS ensures the policies from both sides match.

Communications can only start between the Publisher

and Subscriber when their QoSs are consistent.

Figure 1 shows the initial state of the

environment. It depicts the interaction between DDS

entities and the hosts within the network. This

isolated demonstration allows us to analyze malicious

changes in the configuration of a DDS publisher. It

also allows us to determine whether the domain, the

system, or a subscribing application could be affected

negatively by the change. The intended DDS-based

system has three DDS Entities, with two hosts

running a Shapes Demo application as follows:

 Publisher #1: published Red Square with

SHARED ownership

 Publisher #2: published Green Triangles with

SHARED ownership

 Subscriber #1: subscribed to Red Square and

Green Triangles with SHARED ownership

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

166

6.1 Misuse of RELIABILITY and

RESOURCE_LIMITS DCPS QoS

Policies

This DDS security issue occurred using the DDS

RELIABILITY and RESOURCE_LIMITS QoS

Policy. The RELIABILITY Indicates the level of

reliability offered/requested by the Service that

determines whether data can be dropped or delayed

(OMG, 2015). The RESOURCE_LIMITS specifies

the resources the Service can consume to meet the

requested QoS. This demonstration involved

changing the values of a data topic for these QoSs on

the publisher side. The QoS Policies change by

inserting a DDS configuration file with higher

priority.

Figure 2: Initial state of DDS entities.

Figure 2 shows the initial state of data distribution

of the intended DDS-based system with two

publishers and one subscriber. The subscriber listens

to data from both publishers. To simulate the attack,

the attacker changes the DDS configuration file of the

target publisher (Publisher#1).

The Shapes Demo application has a hardcoded

file that contains one line - the default QoS policy

values to the RTI DDS library. We initially removed

the reference to the RTI DDS library's default QoS

policy values since the DDS offers flexibility to

modify and customize the QoS. Then, a new file

named "USER QOS PROFILES.xml" was created,

containing new and modified default QoS Policy

values. It must be in the same repository as the script

that ran the application (RTI Core Libraries, 2016).

Figure 3 shows Publisher#1 (in red) after

compromising as a malicious entity. The

configuration of other DDS entities (Publisher#2 and

Subscriber#1) was unaltered. The HISTORY and

RESOURCE_LIMITS QoS policies affect

RELIABILITY QoS. The configuration change

involved the depth value of HISTORY,

max_samples_per_instance, and max_samples values

of RESOURCE_LIMITS policy with RELIABLE.

Figure 3: Post exploit state of DDS entities.

Execution of the misuse of DCPS QoS policies

resulted in RTPS messages showing on the Publisher

side. The message is; Samples lost for Square (count

#). On the subscriber side, nothing appears regarding

lost samples. The most important observation was

that the domain runs using the default QoS file. After

stopping, there are two cases: if the QoS policies

values were reasonable, it runs by reading from the

modified file; otherwise, it will show an error

message. The error message says: this file is not going

to be used. Therefore, the attacker cannot modify the

QoS policy directory and files after the Entity enables

in the domain.

6.2 Misuse of DEADLINE and

TIME_BASED_FILTER QoS

Policies

Figure 4: Time constraint of Deadline and

Time_Based_Filter.

This DDS security issue occurred using the DDS

DEADLINE and TIME_BASED_FILTER QoS

Policies. The DEADLINE for a DataReader expects

a new sample updating the value of each instance at

least once every deadline period.

TIME_BASED_FILTER is a filter that allows a

DataReader to specify that it is interested only in a

subset of the values of the data. Note that it is

inconsistent for a DataReader to have a

minimum_separation value longer than its

DEADLINE period, as Figure 4 illustrates.

This demonstration involved changing the values

of a data topic for these QoS on the publisher side.

Attack Simulation on Data Distribution Service-based Infrastructure System

167

We modified the QoS Policies by inserting a DDS

configuration file with higher priority. Figure 2 shows

the initial state of data distribution of the intended

DDS-based system with two publishers and one

subscriber (see Figure 3). The reason for

implementing this misuse is the restriction of the

values for these QoS policies and the ability to modify

the values at any time while running. We used the

same methods in Subsection 6.1 to simulate the

alteration of TIME_BASED_FILTER and

DEADLINE QoS Policies on Publisher #1. The

altering of the configuration file involved altering the

minumm_separation value to be longer than the

deadline period value.

Execution of this misuse of DCPS QoS policies

resulted in the domain running and reading from the

default file from the beginning of the experiment.

Then, after stopping the middleware, an error

message will appear because of the inconsistency of

the values. Therefore, the QoS Policy directory and

files will not read from the modified file even when

starting a new domain with new entities.

7 ANALYSIS AND DISCUSSION

This study examined the manipulation of the software

and the DDS protocol to harm a DDS-based system.

It is significant because a DDS-based system's

security and resilience depend on the assumption that

the hosts on which DDS applications run are in a

trusted zone and thus secure from misuse or abuse.

This paper works on two potential client-side attack

methods against DDS-based systems. We selected

them from 60 DDS security threats in the literature

(Michaud, 2017). Also, we used the same techniques

to investigate, demonstrate, and model our

experiments.

Our research showed that the DDS in RTI

Connext v6.1 is secured compared to the previous

versions. Misusing the DEADLINE and

TIME_BASED_FILTER, or RELIABILITY and

RESOURCE_LIMITS QoS policies could cause

DoS. However, in our experiments, the DDS system

did not respond to the altering and modifying of the

QoS values. Although the TIME_BASED_FILTER

QoS policy is modifiable at any time (OMG, 2015),

the system would not read the new data while running

if it is inconsistent or incorrect.

The security in the current DDS system does not

allow reading policies from another file after it has

started running, albeit it allows data altering from the

high precedence. A client-side attack might expand to

produce more problems and more efficient ways to

compromise a DDS-based system.

8 CONCLUSION AND FUTURE

WORK

In conclusion, this study examines two client-side

attack techniques that might affect systems by

tampering with or changing their configuration file.

DDS structures provide developers with flexibility.

But they also raise the potential security issues of

malicious attacks on a DDS-based application. DDS

is a prime target for attackers because it is located at

the beginning of the software supply chain, making it

easy to lose track of it. The first step in detecting and

defending against a cyberattack on a critical

infrastructure system is understanding DDS security

issues and how an attacker might use them to

compromise a DDS-based system.

For future work, more research is necessary to

evaluate the security of various DDS vendor

implementations. Real-world examination and

evaluation are needed to assess the ease with which

client-side attacks can compromise DDS-based

systems. In addition, it is necessary to evaluate the

damages that could occur if a DDS-based system is

compromised.

Another topic that requires further study and

development is intrusion detection and prevention-

based DDS systems: anomaly-based intrusion

detection systems (IDS) could be helpful since it

identifies contextual dangers DDS faces, and

implementing protection rules for Intrusion

Prevention Systems (IPS) can recognize anomalous

XML files.

REFERENCES

DDS Foundation. (2019). What is DDS? DDS Portal – Data

Distribution Services. https://www.dds-foundation.org/

what-is-dds-3/

DDS Foundation. (2020). How Does DDS Work? DDS

Portal – Data Distribution Services. https://www.dds-

foundation.org/%20how-dds-works/

Object Management Group. (2022, April). The Real-time

Publish-Subscribe Protocol DDS Interoperability Wire

Protocol (DDSI-RTPSTM) Specification. DDS

Interoperability Wire Protocol. https://www.omg.org/

spec/DDSI-RTPS

OMG. (2018, July). About the DDS Security Specification

Version 1.1. OMG. https://www.omg.org/spec/DDS-

SECURITY/1.1

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

168

OMG. (2015). About the Data Distribution Service

Specification Version 1.4. Object Management Group.

https://www.omg.org/spec/DDS/1.4

Kulkarni, A. M., Nayak, V. S., & Rao, P. B. (2012, April).

Comparative study of middleware for C4I systems Web

Services vis-a-vis Data Distribution Service. In 2012

International Conference on Recent Advances in

Computing and Software Systems (pp. 305-310). IEEE.

Maggi, F., Vosseler, R., Cheng, M., Kuo, P., Toyama, C.,

Yen, T. L., & Robotics, A. (2022) A Security Analysis

of the Data Distribution Service (DDS) Protocol.

Chen, F., Huo, Y., Zhu, J., & Fan, D. (2020, November). A

review on the study on MQTT security challenge. In

2020 IEEE International Conference on Smart Cloud

(SmartCloud)(pp. 128-133). IEEE.

Patel, C., & Doshi, N. (2020). A novel MQTT security

framework in generic IoT model. Procedia Computer

Science, 171, 1399-1408.

Mühlbauer, N., Kirdan, E., Pahl, M. O., & Carle, G. (2020,

September). Open-source OPC UA security and

scalability. In 2020 25th IEEE International

Conference on Emerging Technologies and Factory

Automation (ETFA) (Vol. 1, pp. 262-269). IEEE.

Iglesias-Urkia, M., Orive, A., Urbieta, A., & Casado-

Mansilla, D. (2019). Analysis of CoAP

implementations for industrial Internet of Things: a

survey. Journal of Ambient Intelligence and Humanized

Computing, 10(7), 2505-2518.

Rathod, D., & Patil, S. (2017). Security analysis of

constrained application protocol (CoAP): IoT protocol.

International Journal of Advanced Studies in

Computers, Science and Engineering, 6(8), 37.

Laghari, S. A., Manickam, S., & Karuppayah, S. (2021). A

review on SECS/GEM: A machine-to-machine (M2M)

communication protocol for industry 4.0. International

Journal of Electrical and Electronic Engineering &

Telecommunications, 10(2), 105-114.

Michaud, M. J., Dean, T., & Leblanc, S. P. (2018, October).

Attacking omg data distribution service (DDS) based

real-time mission-critical distributed systems. In 2018

13th International Conference on Malicious and

Unwanted Software (MALWARE) (pp. 68-77). IEEE.

Friesen, M., Karthikeyan, G., Heiss, S., Wisniewski, L., &

Trsek, H. (2020). A comparative evaluation of security

mechanisms in DDS, TLS and DTLS. In

Kommunikation und Bildverarbeitung in der

Automation (pp. 201-216). Springer Vieweg, Berlin,

Heidelberg.

White, R., Caiazza, G., Jiang, C., Ou, X., Yang, Z., Cortesi,

A., & Christensen, H. (2019, June). Network

reconnaissance and vulnerability excavation of secure

DDS systems. In 2019 IEEE European symposium on

security and privacy workshops (EUROS&PW) (pp. 57-

66). IEEE.

Michaud, M. (2017). Malicious use of omg data

distribution service (dds) in real-time mission critical

distributed systems (Doctoral dissertation).

Kim, H., Kim, D. K., & Alaerjan, A. (2021). ABAC-based

security model for DDS. IEEE Transactions on

Dependable and Secure Computing.

Park, A. T., Dill, R., Hodson, D. D., & Henry, W. C. (2021,

December). DDS-Cerberus: Ticketing performance

experiments and analysis. In 2021 International

Conference on Computational Science and

Computational Intelligence (CSCI) (pp. 1465-1469).

IEEE.

Abaimov, S. (2021). Towards a Privacy-preserving Deep

Learning-based Network Intrusion Detection in Data

Distribution Services. arXiv preprint

arXiv:2106.06765.

RTI Shapes Demo. (2019). RTI Community. https://commu

nity.rti.com/static/documentation/connext-dds/6.0.0/

doc/manuals/shapes_demo/RTI_Shapes_UsersManual.

pdf

RTI Core Libraries. (2016). RTI Community. https://

community.rti.com/static/documentation/connext-dds/

5.2.3/doc/manuals/connext_dds/RTI_ConnextDDS_C

oreLibraries_UsersManual.pdf

ENISA Threat Landscape 2021. (2021). ENISA.

https://www.enisa.europa.eu/publications/enisa-threat-

landscape-2021

Attack Simulation on Data Distribution Service-based Infrastructure System

169