International Journal of Computer Science Trends and Technology (IJCST) – Volume 13 Issue 2, Mar - Apr 2025
RESEARCH ARTICLE OPEN ACCESS
Edge Computing for IoT: Challenges and Future Directions
Suhail Sahdev
Master of Computer Application Master of Computer Application
R.D. Engineering,Ghaziabad,India R.D. Engineering,Ghaziabad,India
Abhinay Kumar Shivam Saini
Master of Computer Application Master of Computer Application
R.D. Engineering,Ghaziabad,India R.D. Engineering,India
Ashutosh Pradhan
Head of Department (MCA)R.D. Engineering,Ghaziabad,India
Abstract
The explosive proliferation of IoT devices brings forth enormous challenges in data processing and network
management, as mostly attributed to drawbacks associated to centralized cloud-based systems. While cloud
computing has been the underpinning for the management of IoT data, it is fraught with problems due to latency,
bandwidth consumption, and scalability. Computing data on the edge of the proposed systems is perceived as a
possible solution to these. Edge computing helps in real-time decision-making, optimizes bandwidth, and enhances
system performance with less reliance on remote cloud infrastructures. This paper focuses on defining and analyzing
edge computing in an IoT environment, emphasizing its advantages and identifying its challenges. It also proposes
directions for future research in this area.
Keywords: Edge Computing, IoT, Low Latency, Real-Time Processing, Privacy, Network Infrastructure, Scalability.
1. Introduction paper attempts to elaborate the perspective of
integration of edge computing in IoT networks,
IoT completely transform the world through the
recognize the multiple advantages brought by it, list
connection of billions of devices that collect and
its main barriers, and direct possible work that exists
communicate desired information for better decision-
in the future.
making in industries. Such areas such as industrial
automation to health monitoring systems and smart
homes are growing in a rapid pace IoT infrastructure.
2. Background and Related Work
The number of data generated by these devices
increases alongside the number of devices; hence, the Edge computing took on steam with the great
problem of data storage, transmission, and processing expansion of the Internet of Things and stream of
will need a solution. Conventional data management data. Unlike the traditional cloud computing schemes
techniques have mostly relied on cloud computing as where the core data centers are used for storing and
a predominant means of data handling, but they also processing data, edge computing extends the
have major drawbacks, especially concerning latency, processing of data from the edge of the networks,
bandwidth limits, and high costs of processing nearer the devices. This makes data real-time and
massive amounts of data into central data centers. reduces the volume of data that has to be transferred
through the network, thus logically reducing the
Edge computing is a distributed computing paradigm
delay and maximizing the overall efficiency of the
under which computing resources are brought close
system.
to the devices that produce data. Furthermore, edge
computing moves out workloads outside the cloud, Many research works focus on edge computing with
enabling faster processing of data, reduces IoT systems. Shi et al. (2016) mentioned the concept
dependency on the cloud, even more, enhances the called "fog computing," which brings cloud
responsiveness of IoT systems. Hence the present computing to the network edge to improve latency
ISSN: 2347-8578 www.ijcstjournal.org Page 110
International Journal of Computer Science Trends and Technology (IJCST) – Volume 13 Issue 2, Mar - Apr 2025
and bandwidth efficiency. Zhang et al. (2017) shown privacy of data is governed, such as health
that edge computing indeed boosts performance for care.
IoT applications needing real-time processing, such • Enhanced Reliability: The distributed
as smart city and autonomous vehicle applications. computing load renders the IoT system
Yet challenges stay in realization of seamless expensive in terms of repair for out-of-
integration of edge computing in IoT networks network failures. Even local devices will
particularly in the areas of security scalability and work despite an interrupted connection
energy efficiency. between them and the cloud, ensuring that
mission-critical operations continue
3. Edge Computing in IoT: A Conceptual
unaffected by disconnections.
Framework
3.2. Applications of Edge Computing in IoT
The aspect of edge computing in IoT is concerned
with distributed processing of data close to where • Smart Cities: Smart cities take an edge by
creation of data occurs. Edge devices, such as utilizing edge computing for more efficient
gateways, routers, or embedded systems, are applications of traffic, waste, and
distributed at strategic locations within the IoT environmental monitoring through localized
network so that data can be processed, analyzed, and processing of the data acquired from various
reacted in real-time. This greatly influences the sensors in the city. City authorities will
performance and efficiency of the entire IoT therefore be empowered to make timelier
application. decisions that are beneficial to urban life
through local data analysis.
3.1. Key Advantages of Edge Computing
• Edge Computing: In patients' healthcare,
• Reduced Latency: Latency minimization is edge computing offers real-time monitoring
one of the major strengths of edge by employing portable devices. The devices
computing. By processing information can be able to recognize vital signs and send
locally, edge computing achieves a fast an alarm during emergency cases without
response time which is most desirable for having to rely on the cloud to make on-time
real-time applications, for instance in cases decisions.
of autonomous driving, industrial • Industrial Internet of Things (IIoT): Edge
automation, and emergency response computing makes real-time monitoring of
systems. industrial equipment and systems. It can
• Effective Bandwidth Usage: Huge amounts provide a predictive preventive maintenance
of data are generated by IoT sensors, of where machines can be supervised in terms
which a considerable part does not need to of wear and tear assessed, and repairs done
be uploaded onto the cloud for processing. before failure takes place, thus reducing
Edge computing allows local filtering and downtime and operational costs.
aggregation of data before uploading it to
4. Challenges of Edge Computing for IoT
the cloud, thereby meaningfully reducing the
amount of data uploaded, with the added While there are many benefits to edge computing,
advantage of conserving bandwidth and some challenges hinder its widespread adoption and
decongesting the networks. integration in IoT networks.
• Improved Privacy and Security: Edge
computing allows private information to be 4.1. Scalability
processed locally without it having to travel A great challenge regarding scalability is the ever-
to a main cloud server, thereby reducing increasing number of IoT devices. The more
data breaches and increasing privacy. This is connected devices, more edge nodes are needed to
especially important in applications where process data locally at the risk of running out of
ISSN: 2347-8578 www.ijcstjournal.org Page 111
International Journal of Computer Science Trends and Technology (IJCST) – Volume 13 Issue 2, Mar - Apr 2025
resources such as low processing power or storing While edge cloud computing aims to process data
capacity at the edge. Then managing large-scale locall, it still needs to access the cloud for long-term
distances from these nodes becomes an issue. storage and specialized analytics. Further, it calls for
the development of hybrid edge-cloud architectures
4.2. Security and Privacy
that will enable seamless interaction between edge
Since edge nodes are generally much used in public nodes and cloud variables. This will allow one to
or untrusted places, they become perfectly vulnerable process real-time data on the edge while reaching for
to cyberattacks. The protection of the data in transit the cloud for high-volt processing tasks.
and that of the devices is utmost. The security of data
5.3. Security Frameworks for Edge Devices
while being processed locally becomes primary,
especially if it involves sensitive data, e.g. case with Edge computing environments most commonly
personal health information. include less-controlled atmospheres, so building
strong security frameworks becomes increasingly
4.3. Data Synchronization
compelling. Research requires being done in ensuring
It becomes necessary quite often to synchronize the secure data transmission and the authentication of
cloud with IoT infrastructure devices and keep them such devices, as well as establishing trust between
consistent while maintaining real-time data accuracy devices and users.
across the edge-cloud environments, especially when
5.4. Standardization and Interoperability
data is in constant updating and changing status.
The heterogeneity of different vendors in IoT devices
4.4. Energy Efficiency
and edge devices will be a major problem in
Edge devices tend to run in power-constrained interoperability, because interoperability is a
settings, using battery power or small energy sources. challenge to get across this diversity. Thus, standards
Energy efficiency thus becomes an important and protocols should be in place for a common
concern. Energy-efficient algorithms for processing interaction between edge nodes and cloud systems if
and transmission of data need to be developed so that edge computing is going to be accepted generalize
edge nodes are not drained too quickly, especially in into adaptable IoT networking environments.
case of large-scale deployment.
6. Conclusion
5. Future Directions for Edge Computing in IoT
Edge computing is expected to become a
The most interesting thing about edge-computing fundamental technology for IoT, allowing data
future in IoT is the many areas that require further processing in real-time, enhancing latency, and
study: minimizing reliance on cloud-centric infrastructures.
As much as it has several merits, several challenges
5.1. AI and Machine Learning at the Edge still persist, such as scalability, security, and power
efficiency. Overcoming these challenges will be
At the edge, users can enhanced real-time processing
paramount in the increased adoption of edge
of information with the use of AI and machine
computing in IoT systems. Future research should
learning. Thanks to AI-enabled edge devices, data
focus on combining AI and ML at the edge,
processing and decision-making take place locally
optimizing edge-cloud hybrid architectures,
rather than relying on external cloud infrastructure.
strengthening security frameworks, and defining
Such high real-time analysis and decision-making
interoperability standards.
capabilities are important for applications such as
autonomous vehicles. 7. References
5.2. Edge and Cloud Collaboration • Shi, W., et al. (2016). "Edge Computing:
Vision and Challenges." IEEE Internet of
Things Journal.
ISSN: 2347-8578 www.ijcstjournal.org Page 112
International Journal of Computer Science Trends and Technology (IJCST) – Volume 13 Issue 2, Mar - Apr 2025
• Zhang, Y., et al. (2017). "Fog Computing • Zhang, K., et al. (2018). "Machine Learning
and Its Applications in IoT." IEEE for Edge Computing in IoT: Challenges and
Communications Magazine. Opportunities." IEEE Transactions on
Industrial Informatics.
ISSN: 2347-8578 www.ijcstjournal.org Page 113
