Scribd
Upload
4 views

BLOCKCHAIN Basepaper

The document presents a Blockchain-Enabled Federated Learning (BFL) architecture aimed at improving collaborative intrusion detection in Vehicular Edge Computing (VEC) environments. It addresses the limitations of traditional centralized Intrusion Detection Systems (IDS) by enhancing privacy, scalability, and security through decentralized collaboration among vehicles and edge nodes. The proposed system leverages the advantages of both blockchain and federated learning to ensure efficient threat detection while maintaining data privacy.

Uploaded by

deekshu560103
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
4 views

BLOCKCHAIN Basepaper

The document presents a Blockchain-Enabled Federated Learning (BFL) architecture aimed at improving collaborative intrusion detection in Vehicular Edge Computing (VEC) environments. It addresses the limitations of traditional centralized Intrusion Detection Systems (IDS) by enhancing privacy, scalability, and security through decentralized collaboration among vehicles and edge nodes. The proposed system leverages the advantages of both blockchain and federated learning to ensure efficient threat detection while maintaining data privacy.

Uploaded by

deekshu560103
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

BLOCKCHAIN-ENABLED FEDERATED LEARNING FOR ENHANCED

COLLABORATIVE INTRUSION DETECTION IN VEC

Anjali.K.P, V Sem Student, CSE(TOT&CSBT), SEACET, Bangalore-560049


Deekshitha.V, V Sem Student, CSE(TOT&CSBT), SEACET, Bangalore-560049
Monisha.P, V Sem Student, CSE(TOT&CSBT), SEACET, Bangalore-560049

ABSTRACT

Provisioning of CIDs in real time process and decision making had overcome the
Drawback of (ITS) in VEC at the Network Edge.The federated learning in
Blockchain have significally proven to be challenging for various network
attackers.The VEC method have proven to be most reliable to minimize any
network edge attacks.This study advances the field by introducing a lightweight
consensus protocol tailored for resource – constrained VEC scenarios,ensuring
high efficiency.

INTRODUCTION
The emergence of autonomous vehicles and the quick growth of Intelligent
Transportation Systems (ITS) have made Vehicular Edge Computing (VEC) a
crucial component in enabling real-time services and applications. In safety-
critical automotive settings, where quick data processing is essential for
decision-making in applications like traffic control, accident avoidance, and
autonomous driving, this decentralised design is very advantageous.Federated
learning and blockchain are being combined because of their complementing
advantages. Blockchain offers a tamper-resistant framework for safe
cooperation, while federated learning makes it possible to train models in a
scalable, privacy-preserving manner.

LITERATURE SURVEY
Title: A Survey of Federated Learning for Vehicular Networks
The authors of this study are Kai Yang, Tuo Xiang, Longfei Wei, and Qiang Yang.
It offers a thorough analysis of the use of federated learning in automotive
networks.
Xiaosong Zhang, Zhen Yang, and Shuo Wang are the authors of the second
article, "Blockchain for Secure and Efficient Data Sharing in Vehicular Edge
Computing." It explores how blockchain technology can be used to secure data
sharing amongst vehicular edge computing (VEC) devices.

EXISTING SYSTEM
Traditional Intrusion Detection Systems (IDS) in Vehicular Edge Computing (VEC)
environments mostly use centralised systems to monitor and detect cyber
threats. Usually, these centralised systems gather information from different
cars and edge nodes, send it to a central server or cloud, and then examine it
for potential security breaches. High communication latency, single points of
failure, limited scalability, and substantial privacy threats are the main
disadvantages of centralised intrusion detection systems (IDS), particularly
when sensitive vehicle data is sent to a central site for processing.

DISADVANTAGES OF EXISTING SYSTEM


▪ CENTRALIZED ARCHITECTURE VULNERABILITY

▪ HIGH COMMUNICATION LATENCY

▪ PRIVACY CONCERNS

PROPOSED SYSTEM
We provide a Blockchain-Enabled Federated Learning (BFL) architecture that
improves collaborative intrusion detection while preserving privacy, scalability,
and security in order to overcome the issues with the current intrusion
detection systems (IDS) in vehicular edge computing (VEC). This system
develops a decentralised, effective, and safe method for identifying
cyberthreats in automotive networks by utilising the advantages of both
blockchain and federated learning.Federated learning, the mainstay of the
suggested approach, enables cars and edge nodes to work together to train
machine learning models without exchanging raw data. Each node uses its own
data to train a local model rather than sending private information to a central
server. After being shared with a central aggregator, the locally trained models
are combined to produce a worldwide.

ADVANTAGES OF PROPOSED SYSTEM


 ENHANCED PRIVACY PRESERVATION

 DECENTRALIZED TRUST AND SECURITY

 SCALABILITY AND FLEXIBILITY

SYSTEM ARCHITECTURE

Figure 1: Flowchart for System Architecture


IMPLEMENTATION
Dataset Collection
Dataset collection refers to the process of gathering and organizing data from
various sources to create a structured set of information for analysis, training
machine learning models, or conducting research. The dataset can include raw
data or processwd information,depending on the purpose.

FIGURE 2 : CODE IMPLEMENTATION


RESULT

FIGURE 3: Webpage of Blockchain

FIGURE 4:Final Output CONCLUSION AND


FUTURE WORK
To sum up, the Blockchain-Enabled Federated Learning (BFL) framework offers a
revolutionary way to improve collaborative intrusion detection in scenarios
involving vehicular edge computing (VEC). The BFL framework's architecture
efficiently permits collaboration between edge nodes and cars while protecting
sensitive data privacy, guaranteeing that raw data stays on local devices.
Incorporating blockchain technology not only strengthens the system's security
by offering a decentralised trust mechanism, but it also makes participant
interactions more transparent and verifiable.Because of this transparency, a
cooperative atmosphere is created in which cars can freely exchange model
updates and insights without worrying about data manipulation. Additionally,
the blockchain's lightweight consensus mechanism enables the system to grow
effectively, taking into account the dynamic nature of vehicle networks while
maintaining.

REFERENCES

 Zhang, Y., et al. (2020). Federated Learning for Intelligent Transportation


Systems: A Survey.

 This paper provides a comprehensive overview of federated learning


applications in intelligent transportation systems, highlighting its benefits
and challenges in data privacy and collaborative learning. The authors
discuss various use cases and potential frameworks that can be
employed to enhance security in vehicular networks, laying the
groundwork for integrating federated learning with blockchain
technology.

 Mohanta, A. K., & Tripathy, A. (2021). Blockchain-Based Secure and


Efficient Vehicle-to-Everything (V2X) Communication.

 This research explores the role of blockchain technology in securing V2X


communications, emphasizing its potential to improve trust and security
in vehicular networks. The authors propose a blockchain framework that
can enhance the security of data exchanged between vehicles and other
entities, setting the stage for incorporating federated learning in this
context.

You might also like