Early Detection of Cyber Threats in EVCS Using Machine Learning: A Focus on Reconnaissance Attacks

Title

Early Detection of Cyber Threats in EVCS Using Machine Learning: A Focus on Reconnaissance Attacks

Creator

Wagh, Rupali Sunil; Dokaniya, Tina

Description

There is a significant rise in electric vehicle adoption and robust and secure electric vehicle charging station infrastructure to meet this increasing demand. However, advanced technology is vulnerable to several cyber threats. Primarily starting with reconnaissance attacks, attackers gather information about the system to plan greater attacks. This can further lead to several kinds of attacks such as Denial of Service and Host Attacks where the attacker can bypass firewalls, create false traffic and disrupt service for the users. Thus, it is important to detect and prevent these attacks at an early stage. This paper presents a robust machine learning model in order to detect reconnaissance attacks. To the best of our knowledge, there have not been enough studies that focus on specific attack categories for early detection of cyber threats. The ensemble model used in the study demonstrates an impressive accuracy of 97.71% with a good balance between precision and recall. Moreover, variables related to power consumption which are harder to manipulate are used as features. This approach contributes towards more secure EVCS, fosters user trust and promotes adoption of electric vehicles at large. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

Source

Lecture Notes in Networks and Systems;Volume;1612 LNNS;pp.143-154

Date

01-01-2026

Publisher

Springer Science and Business Media Deutschland GmbH

Subject

Cyber security; Early detection; Machine learning; Reconnaissance attacks

Coverage

Wagh R.S., CHRIST (Deemed to be University), Bengaluru, India; Dokaniya T., CHRIST (Deemed to be University), Bengaluru, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISSN: 23673370; ISBN: 978-981952871-4;

Format

online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1007/978-981-95-2872-1_11

https://www.scopus.com/pages/publications/105028287055?origin=resultslist