arXiv subjects

Shantanu Pal

Publications and source records attributed to Shantanu Pal.

At least 19 recordsLinked to original sources

arXiv2026

AGSP-DSA: An Adaptive Graph Signal Processing Framework for Robust Multimodal Fusion with Dynamic Semantic Alignment

In this paper, we introduce an Adaptive Graph Signal Processing with Dynamic Semantic Alignment (AGSP DSA) framework to perform robust multimodal data fusion over heterogeneous sources, including text, audio, and images. The requested approach uses a dual-graph construction to learn both intra-modal and inter-modal relations, spectral graph filtering to boost the informative signals, and effective node embedding with Multi-scale Graph Convolutional Networks (GCNs). Semantic aware attention mechanism: each modality may dynamically contribute to the context with respect to contextual relevance. The experimental outcomes on three benchmark datasets, including CMU-MOSEI, AVE, and MM-IMDB, show that AGSP-DSA performs as the state of the art. More precisely, it achieves 95.3% accuracy, 0.936 F1-score, and 0.924 mAP on CMU-MOSEI, improving MM-GNN by 2.6 percent in accuracy. It gets 93.4% accuracy and 0.911 F1-score on AVE and 91.8% accuracy and 0.886 F1-score on MM-IMDB, which demonstrate good generalization and robustness in the missing modality setting. These findings verify the efficiency of AGSP-DSA in promoting multimodal learning in sentiment analysis, event recognition and multimedia classification.

cs.CV↗

arXiv2025

Context-Aware Intelligent Chatbot Framework Leveraging Mobile Sensing

With the rapid advancement of large language models (LLMs), intelligent conversational assistants have demonstrated remarkable capabilities across various domains. However, they still mainly rely on explicit textual input and do not know the real world behaviors of users. This paper proposes a context-sensitive conversational assistant framework grounded in mobile sensing data. By collecting user behavior and environmental data through smartphones, we abstract these signals into 16 contextual scenarios and translate them into natural language prompts, thus improving the model's understanding of the user's state. We design a structured prompting system to guide the LLM in generating a more personalized and contextually relevant dialogue. This approach integrates mobile sensing with large language models, demonstrating the potential of passive behavioral data in intelligent conversation and offering a viable path toward digital health and personalized interaction.

cs.HC↗

arXiv2025

AegisAgent: An Autonomous Defense Agent Against Prompt Injection Attacks in LLM-HARs

The integration of Large Language Models (LLMs) into wearable sensing is creating a new class of mobile applications capable of nuanced human activity understanding. However, the reliability of these systems is critically undermined by their vulnerability to prompt injection attacks, where attackers deliberately input deceptive instructions into LLMs. Traditional defenses, based on static filters and rigid rules, are insufficient to address the semantic complexity of these new attacks. We argue that a paradigm shift is needed -- from passive filtering to active protection and autonomous reasoning. We introduce AegisAgent, an autonomous agent system designed to ensure the security of LLM-driven HAR systems. Instead of merely blocking threats, AegisAgent functions as a cognitive guardian. It autonomously perceives potential semantic inconsistencies, reasons about the user's true intent by consulting a dynamic memory of past interactions, and acts by generating and executing a multi-step verification and repair plan. We implement AegisAgent as a lightweight, full-stack prototype and conduct a systematic evaluation on 15 common attacks with five state-of-the-art LLM-based HAR systems on three public datasets. Results show it reduces attack success rate by 30\% on average while incurring only 78.6 ms of latency overhead on a GPU workstation. Our work makes the first step towards building secure and trustworthy LLM-driven HAR systems.

cs.CR↗

arXiv2025

TrajSyn: Privacy-Preserving Dataset Distillation from Federated Model Trajectories for Server-Side Adversarial Training

Deep learning models deployed on edge devices are increasingly used in safety-critical applications. However, their vulnerability to adversarial perturbations poses significant risks, especially in Federated Learning (FL) settings where identical models are distributed across thousands of clients. While adversarial training is a strong defense, it is difficult to apply in FL due to strict client-data privacy constraints and the limited compute available on edge devices. In this work, we introduce TrajSyn, a privacy-preserving framework that enables effective server-side adversarial training by synthesizing a proxy dataset from the trajectories of client model updates, without accessing raw client data. We show that TrajSyn consistently improves adversarial robustness on image classification benchmarks with no extra compute burden on the client device.

cs.LG↗

arXiv2023

Blockchain-Based and Fuzzy Logic-Enabled False Data Discovery for the Intelligent Autonomous Vehicular System

Since the beginning of this decade, several incidents report that false data injection attacks targeting intelligent connected vehicles cause huge industrial damage and loss of lives. Data Theft, Flooding, Fuzzing, Hijacking, Malware Spoofing and Advanced Persistent Threats have been immensely growing attack that leads to end-user conflict by abolishing trust on autonomous vehicle. Looking after those sensitive data that contributes to measure the localisation factors of the vehicle, conventional centralised techniques can be misused to update the legitimate vehicular status maliciously. As investigated, the existing centralized false data detection approach based on state and likelihood estimation has a reprehensible trade-off in terms of accuracy, trust, cost, and efficiency. Blockchain with Fuzzy-logic Intelligence has shown its potential to solve localisation issues, trust and false data detection challenges encountered by today's autonomous vehicular system. The proposed Blockchain-based fuzzy solution demonstrates a novel false data detection and reputation preservation technique. The illustrated proposed model filters false and anomalous data based on the vehicles' rules and behaviours. Besides improving the detection accuracy and eliminating the single point of failure, the contributions include appropriating fuzzy AI functions within the Road-side Unit node before authorizing status data by a Blockchain network. Finally, thorough experimental evaluation validates the effectiveness of the proposed model.

cs.CR↗

arXiv2022

A Fog-Based Security Framework for Large-Scale Industrial Internet of Things Environments

The Industrial Internet of Things (IIoT) is a developing research area with potential global Internet connectivity, turning everyday objects into intelligent devices with more autonomous activities. IIoT services and applications are not only being used in smart homes and smart cities, but they have also become an essential element of the Industry 4.0 concept. The emergence of the IIoT helps traditional industries simplify production processes, reduce production costs, and improve industrial efficiency. However, the involvement of many heterogeneous devices, the use of third-party software, and the resource-constrained nature of the IoT devices bring new security risks to the production chain and expose vulnerabilities to the systems. The Distributed Denial of Service (DDoS) attacks are significant, among others. This article analyzes the threats and attacks in the IIoT and discusses how DDoS attacks impact the production process and communication dysfunctions with IIoT services and applications. This article also proposes a reference security framework that enhances the advantages of fog computing to demonstrate countermeasures against DDoS attacks and possible strategies to mitigate such attacks at scale.

cs.DC↗

arXiv2022

A Trusted, Verifiable and Differential Cyber Threat Intelligence Sharing Framework using Blockchain

Cyber Threat Intelligence (CTI) is the knowledge of cyber and physical threats that help mitigate potential cyber attacks. The rapid evolution of the current threat landscape has seen many organisations share CTI to strengthen their security posture for mutual benefit. However, in many cases, CTI data contains attributes (e.g., software versions) that have the potential to leak sensitive information or cause reputational damage to the sharing organisation. While current approaches allow restricting CTI sharing to trusted organisations, they lack solutions where the shared data can be verified and disseminated `differentially' (i.e., selective information sharing) with policies and metrics flexibly defined by an organisation. In this paper, we propose a blockchain-based CTI sharing framework that allows organisations to share sensitive CTI data in a trusted, verifiable and differential manner. We discuss the limitations associated with existing approaches and highlight the advantages of the proposed CTI sharing framework. We further present a detailed proof of concept using the Ethereum blockchain network. Our experimental results show that the proposed framework can facilitate the exchange of CTI without creating significant additional overheads.

cs.CR↗

arXiv2022

VeriBlock: A Blockchain-Based Verifiable Trust Management Architecture with Provable Interactions

There has been considerable advancement in the use of blockchain for trust management in large-scale dynamic systems. In such systems, blockchain is mainly used to store the trust score or trust-related information of interactions among the various entities. However, present trust management architectures using blockchain lack verifiable interactions among the entities on which the trust score is calculated. In this paper, we propose a blockchain-based trust management framework that allows independent trust providers to implement different trust metrics on a common set of trust evidence and provide individual trust value. We employ geo-location as proof of interaction. Some of the existing proposals rely upon geo-location data, but they do not support trust calculation by multiple trust providers. Instead, they can only support a centralised system. Our proposed architecture does not depend upon a single centralised third-party entity to ensure trusted interactions. Our architecture is supported by provable interactions that can easily be verified using blockchain. Therefore, it allows a high degree of confidence in trust management by ensuring the actual interactions between the entities. We provide a detailed design and development of the architecture using real-world use case examples. The proof of prototype was implemented on the Ethereum blockchain platform. Experimental results demonstrate that the employment of independent trust providers adequately provides a high degree of trust scores and that the proposed architecture can be used in a real-world environment.

cs.DC↗

arXiv2022

Security of Machine Learning-Based Anomaly Detection in Cyber Physical Systems

In this study, we focus on the impact of adversarial attacks on deep learning-based anomaly detection in CPS networks and implement a mitigation approach against the attack by retraining models using adversarial samples. We use the Bot-IoT and Modbus IoT datasets to represent the two CPS networks. We train deep learning models and generate adversarial samples using these datasets. These datasets are captured from IoT and Industrial IoT (IIoT) networks. They both provide samples of normal and attack activities. The deep learning model trained with these datasets showed high accuracy in detecting attacks. An Artificial Neural Network (ANN) is adopted with one input layer, four intermediate layers, and one output layer. The output layer has two nodes representing the binary classification results. To generate adversarial samples for the experiment, we used a function called the `fast_gradient_method' from the Cleverhans library. The experimental result demonstrates the influence of FGSM adversarial samples on the accuracy of the predictions and proves the effectiveness of using the retrained model to defend against adversarial attacks.

cs.DC↗

arXiv2022

Challenges and Opportunities of Blockchain for Cyber Threat Intelligence Sharing

The emergence of the Internet of Things (IoT) technology has caused a powerful transition in the cyber threat landscape. As a result, organisations have had to find new ways to better manage the risks associated with their infrastructure. In response, a significant amount of research has focused on developing efficient Cyber Threat Intelligence (CTI) sharing platforms. However, most existing solutions are highly centralised and do not provide a way to exchange information in a distributed way. In this chapter, we subsequently seek to evaluate how blockchain technology can be used to address a number of limitations present in existing CTI sharing platforms. To determine the role of blockchain-based sharing moving forward, we present a number of general CTI sharing challenges, and discuss how blockchain can bring opportunities to address these challenges in a secure and efficient manner. Finally, we discuss a list of relevant works and note some unique future research questions.

cs.CR↗

arXiv2022

A Conceptual Trust Management Framework under Uncertainty for Smart Vehicular Networks

Trust is a fundamental concept in large-scale distributed systems like the Internet of Things (IoT). Trust helps to resolve choices into a decision. However, the trust calculation depends on the amount of uncertainty present in data sources. Trust in an IoT network is proportional to the amount of uncertainty generated by such sources as hardware malfunctions, network stability, adversarial issues, and the nature of data exchanged between the entities. The relationship between trust and uncertainty warrants approaches designed to maximize the former quality whilst minimizing the latter. Unfortunately, there is no consensus on an approach to ensure the trustworthiness of IoT networks, in particular, addressing the uncertainty issues in a fine-grained way. This paper aims to explore a generalized framework designed to manage trust in IoT networks of varying scales. In the proposed framework, several sources of uncertainty are expressed as quantities, trust ratings are calculated for individual entities in an IoT network, and a network model capable of effectively distributing workloads to trustworthy nodes is proposed. We consider a practical use case of smart vehicular networks. By realizing this paper, a standardized approach to building trustworthy IoT networks can be established, which can further guide subsequent works in the field of trust management under uncertainty.

cs.DC↗

arXiv2022

Design and Implementation of EEG-Mechatronic System Interface for Computational Intelligence

The devices that can read Electroencephalography (EEG) signals have been widely used for Brain-Computer Interfaces (BCIs). Popularity in the field of BCIs has increased in recent years with the development of several consumer-grade EEG devices that can detect human cognitive states in real-time and deliver feedback to enhance human performance. Several studies are conducted to understand the fundamentals and essential aspects of EEG in BCIs. However, the significant issue of how can consumer-grade EEG devices be used to control mechatronic systems effectively has been given less attention. In this paper, we have designed and implemented an EEG BCI system using the OpenBCI Cyton headset and a user interface running a game. We employ real-world participants to play a game to gather training data that was later put into multiple machine learning models, including a linear discriminant analysis (LDA), k-nearest neighbours (KNN), and a convolutional neural network (CNN). After training the machine learning models, a validation phase of the experiment took place where participants tried to play the same game but without direct control, utilising the outputs of the machine learning models to determine how the game moved. We find that a CNN trained to the specific user playing the game performed with the highest activation accuracy from the machine learning models tested, allowing for future implementation with a mechatronic system.

cs.HC↗

arXiv2022

Device Identification in Blockchain-Based Internet of Things

In recent years blockchain technology has received tremendous attention. Blockchain users are known by a changeable Public Key (PK) that introduces a level of anonymity, however, studies have shown that anonymized transactions can be linked to deanonymize the users. Most of the existing studies on user de-anonymization focus on monetary applications, however, blockchain has received extensive attention in non-monetary applications like IoT. In this paper we study the impact of de-anonymization on IoT-based blockchain. We populate a blockchain with data of smart home devices and then apply machine learning algorithms in an attempt to classify transactions to a particular device that in turn risks the privacy of the users. Two types of attack models are defined: (i) informed attacks: where attackers know the type of devices installed in a smart home, and (ii) blind attacks: where attackers do not have this information. We show that machine learning algorithms can successfully classify the transactions with 90% accuracy. To enhance the anonymity of the users, we introduce multiple obfuscation methods which include combining multiple packets into a transaction, merging ledgers of multiple devices, and delaying transactions. The implementation results show that these obfuscation methods significantly reduce the attack success rates to 20% to 30% and thus enhance user privacy.

cs.CR↗

arXiv2022

Design and Development of Automated Threat Hunting in Industrial Control Systems

Traditional industrial systems, e.g., power plants, water treatment plants, etc., were built to operate highly isolated and controlled capacity. Recently, Industrial Control Systems (ICSs) have been exposed to the Internet for ease of access and adaptation to advanced technologies. However, it creates security vulnerabilities. Attackers often exploit these vulnerabilities to launch an attack on ICSs. Towards this, threat hunting is performed to proactively monitor the security of ICS networks and protect them against threats that could make the systems malfunction. A threat hunter manually identifies threats and provides a hypothesis based on the available threat intelligence. In this paper, we motivate the gap in lacking research in the automation of threat hunting in ICS networks. We propose an automated extraction of threat intelligence and the generation and validation of a hypothesis. We present an automated threat hunting framework based on threat intelligence provided by the ICS MITRE ATT&CK framework to automate the tasks. Unlike the existing hunting solutions which are cloud-based, costly and prone to human errors, our solution is a central and open-source implemented using different open-source technologies, e.g., Elasticsearch, Conpot, Metasploit, Web Single Page Application (SPA), and a machine learning analyser. Our results demonstrate that the proposed threat hunting solution can identify the network's attacks and alert a threat hunter with a hypothesis generated based on the techniques, tactics, and procedures (TTPs) from ICS MITRE ATT&CK. Then, a machine learning classifier automatically predicts the future actions of the attack.

cs.CR↗

arXiv2021

Addressing Adversarial Machine Learning Attacks in Smart Healthcare Perspectives

Smart healthcare systems are gaining popularity with the rapid development of intelligent sensors, the Internet of Things (IoT) applications and services, and wireless communications. However, at the same time, several vulnerabilities and adversarial attacks make it challenging for a safe and secure smart healthcare system from a security point of view. Machine learning has been used widely to develop suitable models to predict and mitigate attacks. Still, the attacks could trick the machine learning models and misclassify outputs generated by the model. As a result, it leads to incorrect decisions, for example, false disease detection and wrong treatment plans for patients. In this paper, we address the type of adversarial attacks and their impact on smart healthcare systems. We propose a model to examine how adversarial attacks impact machine learning classifiers. To test the model, we use a medical image dataset. Our model can classify medical images with high accuracy. We then attacked the model with a Fast Gradient Sign Method attack (FGSM) to cause the model to predict the images and misclassify them inaccurately. Using transfer learning, we train a VGG-19 model with the medical dataset and later implement the FGSM to the Convolutional Neural Network (CNN) to examine the significant impact it causes on the performance and accuracy of the machine learning model. Our results demonstrate that the adversarial attack misclassifies the images, causing the model's accuracy rate to drop from 88% to 11%.

cs.DC↗

arXiv2021

A Blockchain-Enabled Incentivised Framework for Cyber Threat Intelligence Sharing in ICS

In recent years Industrial Control Systems (ICS) have been targeted increasingly by sophisticated cyberattacks. Improving ICS security has drawn significant attention in the literature that emphasises the importance of Cyber Threat Intelligence (CTI) sharing in accelerating detection, mitigation, and prevention of cyberattacks. However, organisations are reluctant to exchange CTI due to fear of exposure, reputational damage, and lack of incentives. Furthermore, there has been limited discussion about the factors influencing participation in sharing CTI about ICS. The existing CTI-sharing platforms rely on centralised trusted architectures that suffer from a single point of failure and risk companies' privacy as the central node maintains CTI details. In this paper, we address the needs of organisations involved in the management and protection of ICS and present a novel framework that facilitates secure, private, and incentivised exchange of CTI related to ICS using blockchain. We propose a new blockchain-enabled framework that facilitates the secure dissemination of CTI data among multiple stakeholders in ICS. We provide the framework design, technical development and evaluate the framework's feasibility in a real-world application environment using practical use-case scenarios. Our proposed design shows a more practical and efficient framework for a CTI sharing network for ICS, including the bestowal and acknowledgment of data privacy, trust barriers, and security issues ingrained in this domain.

cs.CR↗

arXiv2021

Controlling Resource Allocation using Blockchain-Based Delegation

Allocation of resources and their control over multiple organisations is challenging. This is especially true for a large-scale and dynamic system like the Internet of Things (IoT). One of the core issues in such a system is the provision of secure access control. In particular, transfer of access rights from one entity to another in a secure, flexible and fine-grained manner. In this paper, we present a multi-organisational delegation framework using blockchain. Our framework takes advantage of blockchain smart contracts to define the interactions and resource allocation between the consortium of organisations. We show the feasibility of our solution in a real-world scenario using the allocation of transportation credits in a multi-level organisational setting as a use-case. We provide proof of implementation of the proposed framework using the Hyperledger Fabric blockchain platform. Our results indicate that the proposed framework is efficient and can be used for city-wide transport, potentially even scale country-wide with a shared blockchain with complex access control rules. It also bestows better transparency to the delegation of access rights and control over the employees' transportation access for the organisations.

cs.DC↗

arXiv2021

Blockchain in Supply Chain: Opportunities and Design Considerations

Supply chain applications operate in a multi-stakeholder setting, demanding trust, provenance, and transparency. Blockchain technology provides mechanisms to establish a decentralized infrastructure involving multiple stakeholders. Such mechanisms make the blockchain technology ideal for multi-stakeholder supply chain applications. This chapter introduces the characteristics and requirements of the supply chain and explains how blockchain technology can meet the demands of supply chain applications. In particular, this chapter discusses how data and trust management can be established using blockchain technology. The importance of scalability and interoperability in a blockchain-based supply chain is highlighted to help the stakeholders make an informed decision. The chapter concludes by underscoring the design challenges and open opportunities in the blockchain-based supply chain domain.

cs.CR↗