A Reference Architecture for Dynamic IoT Environments Using

Collaborative Computing Paradigms (CCP-IoT-RA)

Prashant G. Joshi and Bharat M. Deshpande

Department of Computer Science & Information Systems, BITS Pilani K. K. Birla Goa Campus, Goa, India

Keywords:

IoT Reference Architecture, Collaborative Computing Paradigms, Architecture Models, System Software

Architecture, Collaborative Computing, Collaborating Paradigms, Dynamic IoT Environments.

Abstract:

Collaborative Computing Paradigms (CCP) has shown the potential to overcome challenges in dynamic IOT

environments. CCP’s features are interconnection and interplay, dynamic distribution of data processing,

ﬂuidity of computing across paradigms, storage and data management across participating paradigms, and

scalability and extendability of the systems software architecture. Reference Architecture and Models are

known to provide a blueprint, that can be applied across applications domains, and thus can potentially accel-

erate the development and deployment of systems software. Using the features of CCP, this paper proposes

a RA for dynamic IOT environments using the collaborative computing paradigm (CCP-IOT-RA). Proposed

CCP-IOT-RA reference architecture has been applied to commercial and telematics applications like building

automation and vehicle and driver behaviour, demonstrating it’s versatility and effectiveness.

1 INTRODUCTION

The Internet of Things (IOT) is maturing and evolv-

ing, with a signiﬁcant number of solutions already

deployed across various application domains. The

challenges within dynamic IOT environments per-

sist, driven by the heterogeneity of smart sensors, de-

vices, data and computing. In addressing these chal-

lenges, (Joshi and Deshpande., 2024) have introduced

a Collaborative Computing Paradigm (CCP), and val-

idated its efﬁcacy using two applications - data center

cooling systems and driver and vehicle behaviour in

telematics. They have demonstrated that, by intercon-

necting and orchestrating various paradigms, it is fea-

sible to ensure the continuous functionality of critical

use cases at all times. Data storage, choice of comput-

ing paradigm, scalability and extendability have been

identiﬁed as additional characteristics for the CCP.

In the industrial landscape, systems software ar-

chitectures have been designed and deployed in di-

verse application domains. With the demand for sat-

isfying more and more use cases and scenarios, the

complexity of applications has increased, and vari-

ous solutions have also been proposed using a number

of architectural styles, models, and schemes. While

many architectures and architectural styles have been

proposed, and deployed for numerous solutions over

past decade, there is a lacuna of Reference Archi-

tectures for distributed IOT systems (Muccini and

Moghaddam, 2018).

As per (Cavalcante et al., 2015) two architectures

(a) IOT ARM and (b) WSO2 are discussed, and con-

cludes that they fulﬁl the requirements of the IOT ar-

chitecture, there is a need to go step further. A system-

atic mapping study of architectures, in (Garc

es et al.,

2021), shows that reference architectures, especially

in IOT, need to be developed further. The IEEE has

constituted a working group for IEEE 42042 (Stan-

dard for Enterprise, Systems and Software — Ref-

erence architectures) for reference architectures (RA)

which includes RA for IOT. This is clear evidence of

the need to have such reference architectures (IEEE,

2024).

This paper proposes a RA, based on the Collab-

orative Computing Paradigm, and discusses the ap-

plicability to dynamic IOT environments. Proposed

RA has been applied to two applications in the do-

main building automation, and vehicle, and driver be-

haviour.

The paper is organised as follows. Section 2 pro-

vides a brief on characteristics of a typical Reference

Architecture. Section 3 develops an evaluation crite-

ria for IOT RA. Section 4 describes the state-of-art,

and identiﬁes the Challenges and Issues for a RA in

the IoT environments, summarising them based on the

published research, Section 5 proposes and develops a

Joshi, P. and Deshpande, B.

A Reference Architecture for Dynamic IoT Environments Using Collaborative Computing Paradigms (CCP-IoT-RA).

DOI: 10.5220/0012863700003753

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 19th International Conference on Software Technologies (ICSOFT 2024), pages 193-202

ISBN: 978-989-758-706-1; ISSN: 2184-2833

193

Reference Architecture for using Collaborative Com-

puting Paradigms to overcome the challenges and is-

sues with a focus on dynamic IoT environments. Sec-

tion 6 describes and discusses how the applications

and use-cases can be realised from the CCP-IOT-RA.

Section 7 provides a conclusion and Section 8 pro-

vides the details of the future work.

In FIGURE 1, a progression of the content and the

development of the reference architecture is shown.

Reference

Architectures

Reference

Architectures

Evaluation

CCP IOT

Architectures

IOT Environment

Challenges

CCP-IOT-RA

Define CCP-IOT-RA

Application of CCP-

IOT-RA

Figure 1: Progression of systems software architecture and

IOT architectures to deﬁne a CCP-IOT-RA.

2 SYSTEMS SOFTWARE

REFERENCE ARCHITECTURE

As systems become increasingly complex, reliant on

software, and software intensive; a need for RA

arises, which provides a blueprint to design and de-

velop the applications. A RA is known to accelerate

the design and development. Additionally, it ensures

that the systems developed, and deployed demon-

strate shared attributes, and qualities such as security,

interoperability, scalability and adaptability.

Developing an architectural design is a non-trivial

task, and takes deep knowledge and large experience,

both in software and in the target application domain

(Palma et al., 2022). Adopting a relevant RA to build

an architectural design of software system can accel-

erate the development process and increase the prob-

ability of success of the software intensive systems.

Over past decades, software architectures and sys-

tems software architectures, in various forms, directly

for applications or as a RA have been proposed, de-

signed, developed and deployed in academia, and in-

dustry. A RA for building IOT systems have been

discussed in (Kaiser et al., 2023). Generic charac-

teristics, and requirements of RA for IOT have been

discussed in (Weyrich and Ebert, 2016).

Systems software requires tailoring, and this is

even more prominent when addressing the speciﬁc,

and unique demands of the IOT systems. Given its

dynamic nature, IOT systems necessitates a distinct,

and signiﬁcant emphasis on developing system soft-

ware architecture and reference architectures.

3 IoT REFERENCE

ARCHITECTURE EVALUATION

CRITERIA

Dynamic IOT environments are characterised by

unique issues, and challenges which have been sum-

marised in (Joshi and Deshpande., 2024). To over-

come these, an architecture based on collaborating

computing paradigms has been developed, and vali-

dated using two applications (a) Data Center Cooling

System (b) Vehicle & Driver behaviour.

A set with three dimensions (a) Context (b) De-

sign (c) Goals, as part of the framework for software

RA, have been described in (Angelov et al., 2009).

• Contexts: IOT systems are complex due to it’s

multidisciplinary nature. When evaluating the

RA, the context is determined by its use by all

technical personnel responsible for designing, de-

veloping, deploying, and maintaining the system.

• Designs: Target design (detailed - High and Low

level and formally described) to satisfy the re-

quirements of the system, and systems software.

• Goals: While a RA is generic, when adopted to

build a system, will have speciﬁc goals. Goals

may be technical, engineering, performance and

business. Scope of the goals, as considered for

this evaluation, is limited to the technical, engi-

neering and performance of the systems software.

Business goals are speciﬁcally not considered.

In this paper these three dimensions are added to

the scope, and enhanced for use, in evaluation of the

RA for a systems software architecture, and it’s ap-

plication for IOT systems. Thus, a criteria to evaluate

the RA is arrived at. It is used for the evaluation of

the RA and is summarised in TABLE 1.

Further, this paper proposes a set of quality at-

tributes for the RA. They have been summarised in

TABLE 2.

Thus, evaluation proposed and developed in this

section is used, in conjunction with the requirements

of IOT systems, for designing architectures for spe-

ciﬁc application and application domains. They are

discussed in further sections.

ICSOFT 2024 - 19th International Conference on Software Technologies

194

Table 1: Collaborating Computing Paradigm (CCP) Reference Architecture Characteristics.

Characteristics Description ID

Interconnection & Interplay Smart devices, computing paradigms are totally interconnected

enabling a complete interplay.

CCP-C1

Dynamic distribution of data

processing

Dynamic distribution across computing paradigms, provides in-

creased ﬂexibility and efﬁcient distribution.

CCP-C2

Fluidity of computing across

paradigms

Allocation of tasks of computing is dynamic and context driven

to ensure outcome.

CCP-C3

Storage and data manage-

ment across participating

paradigms

As per the context the data is processed and stored in a comput-

ing paradigm. Ultimately the storage of data (raw and processed)

is on the common designated data store.

CCP-C4

Scalability and extendability

of the architecture

Integrate new devices, computing paradigms and infrastruc-

ture, additional hardware and software components to effectively

monitor subsystems, and analyse collected data.

CCP-C5

Context(s) Speciﬁcally from point of view of all the personnel responsi-

ble for design, development, deployment and maintenance of the

system.

CCP-C6

Goal(s) Every RA and it’s application must satisfy the technical goals set

of the system.

CCP-C7

Design(s) Architectural Design (High Level) and Detailed Design (Low

level) to satisfy the requirements of the system.

CCP-C8

Table 2: Reference Architecture Attributes - Proposed.

Attribute Description ID

Modularity Separation of concerns. CCP-QA1

Variability Expandable in pre-planned ways. CCP-QA2

Subsetability Support production of a subset. CCP-QA3

Uniﬁed Architectural uniﬁcation CCP-QA4

Utility Useful and function well for people. CCP-QA5

Robustness Strong and not be vulnerable for changes. CCP-QA6

Feasibility Be practical, implementable and viable. CCP-QA7

Adaptability Able to be changed based on requirements, customization and personalization. CCP-QA8

4 A REF. ARCH. FOR IoT

ENVIRONMENTS

4.1 IoT Reference Architecture: State

of Art and Requirements

Every IOT system is a highly connected, and a com-

plex system (Weyrich and Ebert, 2016). The authors

have compared IOT reference architectures, and have

described the requirements for a IOT RA. This paper

uses them as requirements criteria to satisfy the nec-

essary, and sufﬁcient condition to be a reference ar-

chitecture for IOT. The requirements are summarised

in the TABLE 3.

4.2 IoT Reference Architectures:

Challenges and Issues

IOT environments are unique, and have posed vari-

ety of challenges and issues. These challenges and

issues are speciﬁcally focused on data volume, fre-

quency and processing, data fusion and dynamic com-

puting needs, seamless and multi-interface connectiv-

ity, and choice of computing paradigms. The authors

(Joshi and Deshpande., 2024), have catalogued the

challenges and issues and a summary of the ones rel-

evant to the RA are depicted in TABLE 5

A Reference Architecture for Dynamic IoT Environments Using Collaborative Computing Paradigms (CCP-IoT-RA)

195

Table 3: IOT Reference Architecture Requirements (Weyrich and Ebert, 2016).

Requirements Description ID

Connectivity & Commu-

nication

This will involve one-to-one connectivity (unicast), one-to-many

(multicast) or data collection, information dissemination to multi-

ple entities. Protocols for exchange of data and information.

CCP-R1

Device Management Services for managing the device from on-boarding, device conﬁg-

uration, device provisioning and change propagation, Upgrade of

devices.

CCP-R2

Data Collection, Analy-

sis and actuation

Ultimately to offer services based on knowledge and information

created by computing the data gathered.

CCP-R3

Scalability Ability to handle the increased number of devices, volume of data

for installation of different sizes.

CCP-R4

Security Trust and privacy across the entire system i.e. all aspects of IOT and

all participating entities.

CCP-R5

4.3 IoT Reference Architecture

Requirements and Quality

Attributes

Any software intensive system, and it’s characteris-

tics along with functions require a due consideration

of quality attributes. Both are a must to to deﬁne a

functional system as neither of them stand on their

own (Bass et al., 2003). A set of quality attributes

have been used to describe the RA.

While the quality attributes have been identiﬁed

in this paper for the CCP-IOT-RA, they have not been

further developed. They have been identiﬁed to bring

in completeness to the proposed IOT-RA. The set of

proposed quality attributes have been identiﬁed (Cav-

alcante et al., 2015) and have been summarised in the

TABLE 4.

5 A RA FOR DYNAMIC IoT

ENVIRONMENTS (CCP-IoT-RA)

In previous sections, the RA characteristics, and IOT

RA requirements, characteristics and attributes have

been established. In this section, the RA based on col-

laborative computing paradigm, for IOT is described.

Criteria for evaluation for satisfying the necessary

and sufﬁcient condition for any RA has been based on

the RA and IOT RA requirements.

The proposed CCP-IOT-RA must satisfy the fol-

lowing:

• Must have the characteristics of a RA depicted in

TABLE 1

• Meet the quality attributes of RA depicted in TA-

BLE 2

• Meet the requirements of IOT Reference Archi-

tectures depicted in TABLE 3 and TABLE 4

• Overcome the challenges, and issues for dynamic

IOT environments depicted in TABLE 5

Participating

Mobile Nodes

Participating

Mobile Nodes

Other Systems

Cloud Computing

Remote Mobile

Access

Monitoring

System

Mobile Gateway

Edge Gateway

IOT Devices / Edge Devices . Mist Nodes

Note: All connectivity on an appropriate wired or wireless interface

To Internet

Participating

Fog Nodes

Participating

Fog Nodes

Figure 2: A Reference Architecture for Dynamic IOT Envi-

ronments CCP-IOT-RA.

CCP-IOT-RA, a RA based on Collaborative Com-

puting Paradigms, for Dynamic IOT Environments

is depicted in FIGURE 2. It can be observed, ev-

ery device and node (fog, mobile, cloud or edge) is

interconnected, and a complete interplay is possible.

With this, the limitations of the layering do not exit,

and communication for data, actuation, data collec-

tion and processing can be distributed, by any node,

across the computing paradigms.

Devices connect with gateways, fog, and cloud us-

ing the wired or wireless interfaces, and choose to

communicate (data, upgrade or conﬁguration) based

on the context, and the requirement of the applica-

tions. Data collected can be transmitted for process-

ICSOFT 2024 - 19th International Conference on Software Technologies

196

Table 4: Reference Architecture Requirements and Quality Attributes (Cavalcante et al., 2015).

Characteristics Description ID

Interoperability Among several other devices, services and systems. CCP-QA1

Device Discovery and Management Able to discover newer devices and manage them. CCP-QA2

Context-awareness Gather information and knowledge of context. CCP-QA3

Scalability Manage the increase in devices, data and computing. CCP-QA4

Management of large volumes of data Data generated by smart devices. CCP-QA5

Security, Privacy, and integrity Related to the data. CCP-QA1

Dynamic adaptation High availability and quality of applications execu-

tion time.

CCP-QA6

Table 5: Challenges & Issues in IoT Reference Architectures using Computing Paradigms (Joshi and Deshpande., 2024).

Challenge Description ID

Data Volume, frequency

and Processing

Large volumes of data at high frequencies are collected by large

number of devices in the ﬁeld. Collected data will need to be pro-

cessed and stored.

CCP-IOT1

Software System Archi-

tecture

IOT environments are dynamic and the need is to ensure availabil-

ity of computing infrastructure and capacity for the data processing

task.

CCP-IOT2

Data Fusion and Dy-

namic Computing

Data fusion needs to build knowledge based on heterogeneous data

from variety of sensors.

CCP-IOT3

Seamless and Multi-

interface Connectivity

Connectivity with each device and participating node is required for

data collection, device upgrades, device management and provision-

ing.

CCP-IOT4

Fog/Edge Computing

and Interplay with Cloud

Computing paradigms are available which can be exploited to build

reliable and real-time data processing and actuation.

CCP-IOT5

Choice of Computing

Paradigms

Dynamic IOT environments can beneﬁt from the use of most apr

computing paradigm in the given context to achieve the goal.

CCP-IOT6

Uniﬁed Model and Re-

source Allocation

With interconnected systems, numerous possibilities, policies and

strategies can be employed to ensure efﬁcient utilisation of re-

sources.

CCP-IOT7

Integration and Collabo-

ration

Build synergy among all the participating computing paradigms to

ensure scalability.

CCP-IOT8

ing to the nodes (edge, fog, cloud or mobile), and

the raw data and processed data can be stored locally

on the same node. As part of the storage manage-

ment, periodically or opportunistically the data, raw

and processed, will be transmitted to the Central store.

Diagram shows the following use cases,

• Use Case 1: Device has the capability to connect

to internet, thus can communicate with the cloud

directly, it is also connected to the gateway, and

thus has multiple paradigms are available for com-

puting.

• Use Case 2: Multiple devices are connected to

a gateway, each device is also connected to the

Fog Nodes which in-turn can connect to the cloud.

Each device also has a direct connectivity to the

cloud, thus enabling multiple paradigms to enable

applications.

• Use Case 3: A device has capability to connect

with a mobile node, fog node and also with cloud;

which other nodes are interconnected. This en-

sures that data is collected, and processed on any

one of the paradigms.

• Use Case 4: A device can be connected to inter-

net, like in use case 3, and will also have access to

a gateway node.

Based on the diagram, and functionality descrip-

tion, the criteria for RA is applied, and evaluation re-

sults are mentioned. In subsequent sub-sections the

RA is developed further.

Classically, the CCP focused on ensuring the con-

nectivity, and data while in the proposed RA, device

management, and security have been added as at-

tributes. The way CCP is structured, and design both

these additional attributes can be designed using the

CCP.

A Reference Architecture for Dynamic IoT Environments Using Collaborative Computing Paradigms (CCP-IoT-RA)

197

Characteristics of the RA uses the CCP-Systems

Software Architecture; thus inherits the characteris-

tics of a CCP and additionally able to satisfy the re-

quirements of an IOT RA stated in Requirements Ta-

ble 3

Device Management, a function that can be im-

plemented using the CCP-IOT. Security - Trust and

privacy become the quality attributes of the CCP-IOT.

5.1 CCP-IoT-RA Requirements

As per the evaluation criteria from previous sections,

the CCP-IOT-RA is evaluated for coverage of the re-

quirements of the RA. This is detailed in the TABLE

6. The requirements are mapped to the CCP-IOT-RA

characteristics.

5.2 CCP-IoT-RA Characteristics

The CCP-IOT-RA identiﬁed characteristics are de-

scribed and summarised in TABLE 1. The ﬁrst ﬁve

characteristics which are from the CCP architecture

and the last three are from a functionality perspective

that comes in to action for a particular application.

5.3 CCP-IoT-RA Quality Attributes

As part of the evaluation criteria established in pre-

vious sections, the CCP-IOT-RA is evaluated against

the coverage of the quality attributes of the RA. This

is detailed in the TABLE 8. It can be observed that

the quality attributes are mapped to the CCP-IOT-RA

characteristics.

5.4 Characteristics of CCP-IOT-RA

With the challenges posed by the dynamic IOT en-

vironments, the characteristics of the CCP-IOT-RA,

the RA, are mapped to the IOT - challenges and is-

sues. The identiﬁed challenges and issues become the

requirements for the CCP-IOT-RA to fulﬁl. The map-

ping along with comments is detailed in TABLE 7.

6 APPLICATIONS OF

CCP-IoT-RA

In this section, large systems in commercial and au-

tomotive have been discussed. Each system is re-

quired to be built using the IOT environments. The

CCP-IOT-RA is applied to each of the systems, and

sub-systems, and the requirements are mapped to the

characteristics of the CCP-IOT-RA. Aim is to ascer-

tain that the CCP-IOT-RA is indeed a RA, which can

form a template for all systems, and provide distinct

and signiﬁcant enhancements over classical IOT.

6.1 Building Automation

A typical commercial building, like a commercial

complex, with ofﬁces and shops, has multiple sub-

systems, which are deployed to automate the day-to-

day operations and upkeep.

Each of the subsystem requirements are listed and

discussed brieﬂy, and then mapped to the CCP-IOT-

RA. It evident that the CCP-IOT-RA can be used a

blueprint to design the entire system to be versatile,

and enable the system to work efﬁciently.

6.1.1 Fire Safety System

Fire Safety equipment is required to be installed, and

will need to be monitored, and in particular cases spe-

ciﬁc actuation is a must.

• Fire Extinguishers: Typically weight, and the re-

ﬁll schedule is to be monitored for the ﬁre ex-

tinguishers. Based on the CCP-IOT-RA the Use

Case 4 may be considered for operations.

• Fire Sprinklers: Water pressure is monitored and

maintained to a level, such that when the sprin-

klers are activated water can be pumped for con-

trolling the ﬁre. Sensors are deployed to check

water pressure on the lines, and a jockey pump is

operated when the pressure drops. Water pressure

is monitored continuously and communicated to

the nodes for processing. Based on the CCP-IOT-

RA the Use Case 2 may be considered for opera-

tions.

Salient Features:

• Real-time notiﬁcations are a must in case of ﬁre

or fault in equipment

• Data collected, raw and processed data, from sys-

tems for predictive analytics

• CCP-IOT-RA ensures notiﬁcations at all times

with interconnection and interplay

The system will notify the designated facility staff

for any malfunction in real-time. With CCP system

is interconnected - messages and notiﬁcations will be

sent, using available routes, reliably.

6.1.2 Pubic Toilets Monitoring System

Large buildings have multiple public toilet facilities

of various types like family, differently abled, in ad-

dition to the standard ones for male/female. Some

establishments may have facilities for pets too.

ICSOFT 2024 - 19th International Conference on Software Technologies

198

Table 6: CCP-IOT-RA Requirements evaluation.

ID Requirements CCP-

IOT-RA

Mapping

RA Comments

CCP-R1 Connectivity &

Communication

CCP-C1 With interconnection and interplay the information dissemination

in any modes (unicast or multicast) is enabled; and the intercon-

nection ensures protocol comparibility.

CCP-R2 Device Manage-

ment

CCP-C5. With the features to on-board and de-board devices and connectiv-

ity with Fog, Edge and CLoud, the device management in addition

to on-board and de-board along with provisioning, upgrade of de-

vices and conﬁguration can be managed.

CCP-R3 Data Collection,

Analysis and ac-

tuation

CCP-C1,

CCP-C2

System are interconnected and computing is distributed; with this

the data collection, processing, analysis and actuation is possible.

CCP-R3 Scalability CCP-C5,

CCP-C4.

Each of the computing paradigms are scalable, which bring in

scalability to computing, handling of added devices as well as

computing of the increased volume of data.

CCP-R5 Security CCP-C1. Largely security will be a function of sofware application which

is already enabled with the characteristics of interconnection and

interplay.

• Monitoring the Trash Bins: Status of the bins

can be determined by weight and/or by level of

ﬁlling. Based on the level the action to clear the

bins can be taken. Use Case 4 of CCP-IOT-RA

may be considered for operations.

• Monitoring the Soap and Water Consumption:

Soap dispenser and water faucets will need to

have the sensors which can transmit the levels,

and amount of consumption periodically. Use

Case 2 of CCP-IOT-RA may be considered for op-

erations.

• Monitoring the Odour and Air Quality: Air

quality and gas sensors can be installed to monitor

the odour and air quality to enable actuation of air

puriﬁers or exhaust fans. Use Case 2 of CCP-IOT-

RA may be considered for operations.

Salient features:

• Real-time notiﬁcations in case of cleaning re-

quirements

• Collect Data from all systems for analytics

• CCP-IOT-RA enables notiﬁcations at all times

6.1.3 Parking Management System

Parking system is dynamic with the ingress and egress

of cars along with detection of presence of a car in a

particular parking space, to aggregate the numbers per

parking level. Such data can be shown on a dashboard

to ascertain the number of free and occupied parking

spaces.

• Car Entry/Exit Barrier Operation: Sensors are

installed to detect the presence of the car and

movement (entry or exit) and based on the pay-

ment system the barrier is lowered or cleared for

movement. All such sensors are required to be re-

motely monitored and can be automated as well.

Based on the CCP-IOT-RA the Use Case 3 may

be considered for operations.

• Parking Spot Car Detection: Presence or ab-

sence of a car at a spot and aggregation of such

data levelwise can indicate the parking capacity

situation. Sensors on each parking spot can pro-

vide that data. Data and location may be used to

guide the car entering for parking to the available

spot. Based on the CCP-IOT-RA the Use Case 2

and 4 may be considered for operations.

Salient Features:

• Real-time notiﬁcations in case of issues or assis-

tance need

• CCP-IOT-RA required to ensure required notiﬁ-

cations are sent in relevant use cases

6.1.4 A CCP-IOT-RA Based System

For the three systems, discussed brieﬂy above, a CCP-

IOT-RA can be adapted, tailored and deployed to en-

able all or more such sub-systems. In such a system,

CCP-IOT-RA can enable:

• Addition of sub-systems, more features, and func-

tionality like addition of monitoring of the air-

conditioning system or Addition of speciﬁc ﬁre

safety (additional equipment) for newly added

restaurants can be done easily with scalability and

extendability of CCP-IOT-RA

A Reference Architecture for Dynamic IoT Environments Using Collaborative Computing Paradigms (CCP-IoT-RA)

199

Table 7: CCP-IOT-RA IoT Reference Architectures Evaluation - Challenges and Issues.

ID Challenge/Issue CCP-IOT

RA map-

ping

RA Comments

CCP-IOT1 Data Volume, frequency and

Processing

CCP-C4,

CCP-C2,

CCP-C3

CCP enables high volume of data collection, and

processing as multiple paradigms and storage

management is made available.

CCP-IOT2 Software System Architec-

ture

CCP-C5,

CCP-C1

With scalability, and interconnection and interplay

enabled with ﬂuidity of computing dynamic envi-

ronments are easily served.

CCP-IOT3 Data Fusion and Dynamic

Computing

CCP-C2 Data handling in IOT is crucia, and CCP-IOT-RA

enables not only collection but processing of data

across multiple nodes.

CCP-IOT4 Seamless and Multi-

interface Connectivity

CCP-C5 CCP-IOT-RA is based on the multi-interface con-

nectivity and leverages all means of communica-

tion.

CCP-IOT5 Fog/Edge Computing and

Interplay with Cloud

CCP-C1. Inherent feature of the way CPP-IOT-RA is de-

signed.

CCP-IOT6 Choice of Computing

Paradigms

CCP-C1,

CCP-C3,

CCP-C6

With an interplay possible, availability of comput-

ing paradigm is increased and the a decision can

be made based on the policy and the context.

CCP-IOT7 Uniﬁed Model and Resource

Allocation

CCP-C1,

CCP-C3

With the way CCP-IOT-RA is designed, it func-

tions as a uniﬁed system along with ﬂexibility to

allocate resources for various functionality in an

IOT system.

CCP-IOT8 Integration and Collabora-

tion

CCP-C1,

CCP-C3

With the way CCP-IOT-RA is designed, entire

system work on collaboration and synergy of the

interconnected components.

• With interconnection and interplay, notiﬁcations,

acutuation, and operational data, and alerts can be

guaranteed for excellence in operations

6.2 Vehicle and Driver Behaviour

Figure 3: CCP-IOT-RA as applied to Telematics - Vehicle

& Driver behaviour.

Fleets of commercial and passenger vehicles need to

manage their drivers, vehicles, routes and schedules.

In most cases, there is a daily plan, for say city bus

ﬂeet, transportation companies face dynamism in the

way they operate. Monitoring vehicles, and drivers, is

required to ensure compliance and safety, say hours of

operation for drivers, and dash cameras for insurance

claims. Various use cases can be considered which re-

quire real-time to near real-time information example:

Accident or faults in vehicles will need to be reported

in real-time.

Driver behaviour is analyzed based on the vari-

ety of data like speed, braking habits, and accelera-

tion along with correlating the location of the vehi-

cle. This data is collected regularly, and processed

in batches to derive patterns, which provide a score

for driving. The CCP-IOT-RA facilitates data col-

lection, and ensures data storage, and processing on

various intermediate nodes, which enhances the over-

all storage and processing capacity. This data stored

on various nodes, can then to sent to central data

store for storage and comprehensive analysis of driver

behaviour. Alerts for issues such as speeding, lane

changes, acceleration, and braking can be generated

by processing the data either on the device itself or

through a gateway (mobile or ﬁxed equipment) within

the vehicle.

Vehicle data can be collected for each trip, and an-

ICSOFT 2024 - 19th International Conference on Software Technologies

200

Table 8: CCP-IOT-RA Quality Attributes Evaluation.

ID Characteristics CCP-IOT-

RA Comments

CCP-QA1 Interoperability CCP-C1 With interconnections and interplay, interoperability

forms the central feature of CCP-IOT-RA for all compo-

nents to work with each other, based on standards

CCP-QA2 Device Discovery

and Management

CCP-C5 As on-boarding and de-boarding of devices is enabled in

the CCP-IOT-RA enables device discovery and manage-

ment as a function

CCP-QA3 Context-awareness CCP-C1,

CCP-C6

Context as deﬁned for the topology of the connection or

the application can be arrived at with the interconnection

and interplay.

CCP-QA4 Scalability CCP-C5 A base feature of CCP-IOT-RA and enables addition of

devices and nodes dynamically.

CCP-QA5 Management of large

volumes of data

CCP-C4 With multiple participating paradigms the volume of data

can be easily managed.

CCP-QA6 Security, Privacy,

and integrity

CCP-C1 Though a software feature interconnection and interplay

is pivotal in enabling this.

CCP-QA7 Dynamic adaptation CCP-C5,

CCP-C8

With the way the system is designed for dynamic IOT

systems, overall dynamism of any complex system can be

handled, and the CCP-IOT-RA can adapt to the dynamics

alyzed over time to assess the need for maintenance or

repairs, along with fault codes for diagnostics. Some

use cases require the immediate transmission of fault

codes, while others involve collecting data over an ex-

tended period to analyze operational efﬁciency, and

performance in greater detail.

All these cases require the versatility and efﬁ-

ciency provided by CCP-IOT-RA. Its unique features

of interconnectedness enable real-time alerts and no-

tiﬁcations in the discussed use cases.

7 CONCLUSION

Using collaborative computing paradigms, a RA

(CCP-IOT-RA), for dynamic IOT environments has

been developed. The CCP-IOT-RA has been applied

to two IOT system - building automation, and vehi-

cle and driver behaviour. With this it is established

clearly that a RA using CCP-IOT-RA will form a

blueprint for design of solutions in dynamic IOT en-

vironments. Characteristics of CCP-IOT-RA has also

been mapped to the characteristics and quality at-

tributes of reference architectures, and to the chal-

lenges of IOT environments. With the available appli-

cation it is evident that the CCP-IOT-RA is versatile

and effective for IOT environments.

8 FUTURE WORK

Furthermore, there is a need to experiment, and ap-

ply the CCP-IOT-RA to applications in variety of

domains. It is planned to detail the RA functional

blocks and develop for target domains of building au-

tomation and telematics. By focusing on the qual-

ity attributes of the RA, which can enable studying

the characteristics, attributes and performance of the

CCP-IOT-RA further. In addition, compare and con-

trast with other reference architectures, based on the

proposed RA requirements in this paper.

REFERENCES

Angelov, S., Grefen, P., and Greefhorst, D. (2009). A clas-

siﬁcation of software reference architectures: Ana-

lyzing their success and effectiveness. In 2009 Joint

Working IEEE/IFIP Conference on Software Architec-

ture & European Conference on Software Architec-

ture, pages 141–150.

Bass, L., Clements, P., and Kazman, R. (2003). Software

Architecture in Practice. SEI series in software engi-

neering. Addison-Wesley.

Cavalcante, E., Alves, M. P., Batista, T., Delicato, F. C.,

and Pires, P. F. (2015). An analysis of reference ar-

chitectures for the internet of things. In 2015 1st In-

ternational Workshop on Exploring Component-based

Techniques for Constructing Reference Architectures

(CobRA), pages 1–4.

A Reference Architecture for Dynamic IoT Environments Using Collaborative Computing Paradigms (CCP-IoT-RA)

201

Garc

es, L., Mart

ınez-Fern

andez, S., Oliveira, L., Valle, P.,

Ayala, C., Franch, X., and Nakagawa, E. Y. (2021).

Three decades of software reference architectures: A

systematic mapping study. Journal of Systems and

Software, 179:111004.

IEEE (2024). Ieee standards association - 42042. In IEEE

Standards Association.

Joshi, P. and Deshpande., B. (2024). Collaborative comput-

ing paradigms: A software systems architecture for

dynamic iot environments. In Proceedings of the 12th

International Conference on Model-Based Software

and Systems Engineering - MODELSWARD, pages

297–306. INSTICC, SciTePress.

Kaiser, J., McFarlane, D., Hawkridge, G., Andr

e, P., and

Leit

ao, P. (2023). A review of reference architectures

for digital manufacturing: Classiﬁcation, applicability

and open issues. Computers in Industry, 149:103923.

Muccini, H. and Moghaddam, M. T. (2018). Iot architec-

tural styles. In Cuesta, C. E., Garlan, D., and P

erez,

J., editors, Software Architecture, pages 68–85, Cham.

Springer International Publishing.

Palma, E. S., Nakagawa, E. Y., Paiva, D. M. B., and Cagnin,

M. I. (2022). Evolving reference architecture descrip-

tion: Guidelines based on iso/iec/ieee 42010.

Weyrich, M. and Ebert, C. (2016). Reference architectures

for the internet of things. IEEE Software, 33(1):112–

116.

ICSOFT 2024 - 19th International Conference on Software Technologies

202