Skip to main content
logo

NFSU Journal of Forensic Science

Volume 3, Issue 4 - 2026 (July - December 2026)

Volume 3 Issue 4 Cover

Issue Details:

Volume 3 Issue 4
Published:Sep 24, 2026

Editorial: July - December 2026

Welcome to the 2026 issue of NFSU Journal of Forensic Science. This issue showcases the remarkable breadth and depth of contemporary research across multiple disciplines. From cutting-edge applications of machine learning in climate science to the revolutionary potential of quantum computing in drug discovery, our featured articles demonstrate the power of interdisciplinary collaboration in addressing global challenges.

We are particularly excited to present research that bridges traditional academic boundaries, reflecting our journal's commitment to fostering innovation through cross-disciplinary dialogue. The integration of artificial intelligence with environmental science, the application of blockchain technology to supply chain management, and the convergence of urban planning with smart city technologies exemplify the transformative potential of collaborative research.

As we continue to navigate an era of rapid technological advancement and global challenges, the research presented in this issue offers both insights and solutions that will shape our future. We thank our authors, reviewers, and editorial board members for their continued dedication to advancing knowledge and promoting scientific excellence.

Prof. (Dr.) Naveen Kumar Chaudhary
Editor in Chief
NFSU Journal of Forensic Science

Articles in This Issue

Showing 5 of 5 articles
Research PaperID: nfsujfs330055Pages 1-12Open AccessOpen Access

SIM Swapping Attacks: Mechanisms, Identity Takeover Pathways, and Forensic Countermeasures in Modern Telecom Networks

Sandeep Munjal, Hiba A N, Steve Jerson K, Shreya Naik, Naveen Kumar Chaudhary

Due to the ubiquity of mobile devices for digital identity verification, SIM swapping has emerged as an unintended high-stakes vector for cybercrimes. With the advancement of telecommunications technology from GSM to 5G, reliance on the Subscriber Identity Module (SIM) for two-factor authentication (2FA) has become a critical vulnerability. In this paper, fraudsters employ a variety of techniques, ranging from social engineering to technological exploitation, across both traditional and contemporary network infrastructures. We investigate identity takeover methods that circumvent traditional security mechanisms and cause significant financial and personal losses. In addition, this study assesses telecom infrastructure weaknesses where unique vulnerabilities of GSM, LTE, and UMTS interworking exist, and covers new attacks, including AdaptOver. Eventually, we propose a model for forensic countermeasures and mitigation by focusing on location-based verification, devicespecific authentication binding, and increased ceremony modelling to protect subscriber identities.

telecommunication fraudSIM swappingidentity theftauthenticationnetwork securityforensic analysis
74 views
56 downloads

Contributors:

 Sandeep Munjal
,
 Hiba A N
,
 Steve Jerson K
,
 Shreya Naik
,
 Naveen Kumar Chaudhary
Research PaperID: nfsujfs330056Pages 12-28Open AccessOpen Access

Detecting Adversarial Agents in Telecom Financial Networks Through Inter-Agent (A2A) Communication Behaviour)

Aditya Mohan Srivastava, Panem Charanarur, Naveen Kumar Chaudhary, Dr. Lokesh Chouhan

As financial ecosystems migrate toward an architecture composed of autonomous software agents representing banks, telecom operators, payment processors, and end users. The communication channels between these agents become high-value targets for adversarial exploitation. Conventional fraud detectors operate on transaction-level attributes such as amount and type, and are blind to the behavioural signatures embedded in inter-agent message exchange. This paper proposes a detection framework that extracts twelve communication behavioral features (response time, message length, retry count, session anomaly score, authentication failure rate, interaction frequency, and others) from agentto-agent (A2A) traffic and feeds them through a two stage pipeline: (i) an autoencoder trained exclusively on legitimate interactions to produce an eight dimensional latent embedding and a scalar reconstruction error anomaly score, and (ii) an XGBoost classifier that operates on the 21 dimensional fusion of raw and autoencoder derived features. We evaluate the system on PaySim, a widely used financial dataset containing 6.36 million records, under a deliberately balanced protocol (50/50 fraudlegitimate split) that avoids the metric inflation common in imbalanced evaluations. Across six distinct attack types : (i) impersonation (ii) prompt injection (iii) collusion (iv) trust-chain exploitation (v) data exfiltration, and (vi) transaction manipulation. The proposed framework attains an F1-score of 0.933, AUC-ROC of 0.983, and per-attack F1 ranging from 0.928 (collusion, the stealthiest category) to 1.000 (transaction manipulation). A complementary four-tier graduated mitigation strategy (Monitor/Throttle/Quarantine/Block) achieves 93.52% fraud prevention, marginally exceeding binary blocking (93.27%) while reducing friction for legitimate users by 30%. We discuss the implications for emerging agent-to-agent (A2A) ecosystems expected in 2028 and beyond, in which autonomous AI agents will negotiate, transact, and settle on behalf of human principals without direct human oversight.

multi-agent systemsagent-to-agent communicationfraud detectionautoencoderXGBoostadversarial attacks+3 more
62 views
54 downloads

Contributors:

 Aditya Mohan Srivastava
,
 Panem Charanarur
,
 Naveen Kumar Chaudhary
,
 Dr. Lokesh Chouhan
Research PaperID: nfsujfs330057Pages 29-42Open AccessOpen Access

Reconstructing Telecom Fraud Networks: Operational Insights from CID

Sundaram Vishal, Vinay K A

One of the most critical emerging threat to India's digital economy is the Telecom-enabled fraud with estimated losses exceeding ₹50,000 crore over the past six years. The paper provides a functional model of forensic network reconstruction, which is based on field level investigations at Criminal Investigation Department (CID). The paper not only explores the technical anatomy of SMS interception malware, but also investigates the dynamics of fraud in the real world, specifically utilizing the cryptocurrency vectors, and offers a specialized forensic tool, CDRAnalyz, to analyze Call Detail Records (CDRs). These technical findings are contextualized in the research with references to the legal provisions on Section 63 of the Bharatiya Sakshya Adhiniyam (BSA) 2023 [1], which pertains to the admissibility of electronic evidence. The proposed system incorporates malware reverse engineering, network topology analysis, and legal compliance systems to improve investigation systems in fighting high-tech telecom based fraud ecosystems.

Telecom-enabled fraudSMS malwareCDR analysisNetwork mappingBSA 2023Cryptocurrency fraud.
69 views
65 downloads

Contributors:

 Sundaram Vishal
,
 Vinay K A
Research PaperID: nfsujfs330058Pages 43-52Open AccessOpen Access

Detection and Mitigation of Bot-in -theMiddle Attacks Targeting OTP-Based Authentication in Telecom Frauds

Anouska Dutta, Hrishikesh Khedekar, T.K. Gundoor

The one-time passwords (OTPs) that are delivered by the use of SMSs have become a very attractive target of telecom fraud since it can be used as a second line of authentication. OTP Bots are automation-based interception services that run through Telegram-based Fraud-as-a-Service (FaaS) platforms that have become a systemic tool of real-time avoidance of multi-factor authentication, caller-ID spoofing, and abusive number-porting. In this paper, we systematically investigate the OTP bot threat space and make three main contributions, namely: (i): four-tier taxonomic framework distinguishing between Vishing Bots, SMS Interception Bots, SIM-Swap Bots, and AI-enhanced hybrid bots; (ii): five-step operational life cycle model superimposed on the MITRE ATTACK framework; and (iii): a hybrid detection architecture that uses rule-based analysis in combination with a Random Forest classifier that we tested on 50 thousand artificial authentication sessions. It has an accuracy of 96.84% and recall of 94.67% as well as a falsepositive rate of 2.9% and a detection latency of 42 milliseconds, which is better than both rule-based and machine learning-only baselines. Threat-intelligence reports suggest that during the next year, 2023 to 2024, the activity of dark-web OTP bots is expected to increase by 31 per cent each year, and the cost of services is USD 10-50 per session, and success rates can be up to 90 per cent. It is estimated that artificially inflated traffic (AIT) contributes to 20% of OTP SMS volume, which implies USD 1.15 billion of fraudulent charges per year. We offer a tiered mitigation structure that operates on the level of telecom operators (implementing STIR/SHAKEN, SS7 firewalling, and porting KYC verification), financial institutions (migration to FIDO2 and reducing the OTP validity window), regulators (attestation requirements and coordinated takedown), and end users (improved social-engineering awareness). The evidence indicates that, without reform at the infrastructure level, OTPs, which are based on SMS, do not currently offer a high enough level of second-factor authentication to prevent the high-assurance authentication

OTP BotOne-Time Password InterceptionTelecom FraudFraud-as-a-Service (FaaS)Multi-Factor Authentication BypassBot-in-the-Middle Attack+10 more
83 views
62 downloads

Contributors:

 Anouska Dutta
,
 Hrishikesh Khedekar
,
 T.K. Gundoor
Research PaperID: nfsujfs330059Pages 53-59Open AccessOpen Access

Self-Adapting Enterprise Threat Intelligence tool with Automated Deception and Real-Time Cyber Defense

Deepkumar Wagh, Dr. Panem Charanarur

Cyber threats have increased in sophistication and complexity, and traditional perimeter security defenses are failing to stop advanced persistent threats, zero-day attacks, and targeted intrusion attacks. The project aims to design, implement, and test an AI-powered Cyber Deception System. It is a proactive cybersecurity system that includes honeypot emulation, threat classification with machine learning, unsupervised anomaly detection and reinforcement learning based deception strategy adaptation in a single deployment that requires limited resources. The system will deploy 4 honeypot services: SSH, HTTP, FTP, Telnet. These services monitor the attacker's actions without revealing any production resources. Each source IP is given a behavioural fingerprint and passed on to a Random Forest classifier which classifies attack traffic into five categories: brute-force, reconnaissance, exploitation, lateral movement, and data exfiltration. It is accompanied by another anomaly detector called Isolation Forest for the detection of unknown attack patterns (zero-day and out-of-distribution) that are not present in the training data of the classifier. A composite threat score will add the confidence levels of the classifiers, the magnitude of the anomaly, how often it occurs in the behaviour and the recurrence signals for each attacker IP and can help prioritise responses according to the risk. A Q-Learning reinforcement learning agent selects a strategy from a catalogue of seven deception strategies. The agent learns the strategies that most help the attacker to spend more time during simulated engagement episodes. It shows consistent improvement in rewards as training continues.

Cyber DeceptionHoneypotMachine LearningRandom ForestIsolation ForestReinforcement Learning+1 more
57 views
50 downloads

Contributors:

 Deepkumar Wagh
,
 Dr. Panem Charanarur

Related Issues

Volume 3, Issue 3

2026, 2 articles