Context-Aware Self-Healing Security Framework for WBAN Against Sybil Attacks

Team

Supervisors

Table of Content

  1. Abstract
  2. Related Works
  3. Methodology
  4. Experiment Setup and Implementation
  5. Results and Analysis
  6. Conclusion
  7. Publications
  8. Links

Abstract

Wireless Body Area Networks (WBANs) are a critical component of the Internet of Medical Things (IoMT), enabling continuous health monitoring through wearable and implanted medical sensors. Due to their resource-constrained nature and reliance on open wireless communication, WBANs are highly vulnerable to security threats. Among these threats, Sybil attacks pose a significant risk by allowing a single malicious node to impersonate multiple identities, leading to false medical data injection, network disruption, and compromised patient safety.

This project proposes a lightweight, context-aware, self-healing security framework for detecting and mitigating Sybil attacks in IoMT-based WBANs. The framework integrates cryptographic identity verification with context-aware machine learning techniques to improve detection accuracy while minimizing energy consumption and latency. Additionally, a self-healing mechanism is introduced to autonomously isolate compromised nodes, revoke security credentials, and restore network reliability. Experimental evaluation using a simulated WBAN environment demonstrates the effectiveness and practicality of the proposed approach.

Existing Sybil attack mitigation techniques in IoMT and WBAN environments can be broadly categorized into trust-based, context-aware, cryptographic-based, and machine learning-based approaches. Trust-based methods rely on node behavior and reputation scores but often suffer from bad-mouthing and ballot-stuffing attacks. Context-aware approaches utilize network and physical context information; however, many rely on unrealistic assumptions and lack robust validation.

Cryptographic solutions provide strong identity verification but introduce high computational and energy overhead, making them unsuitable for WBAN devices. Machine learning-based approaches improve adaptability but frequently lack real-time response and self-healing capabilities. The absence of lightweight, adaptive, and autonomous recovery mechanisms highlights a clear research gap, which this project aims to address.

Methodology

The proposed methodology follows a multi-phase approach:

  1. Literature Review – Comprehensive analysis of existing Sybil detection techniques in WBAN and IoMT systems.
  2. Dataset Preparation – Generation of a simulated WBAN dataset containing both normal and Sybil attack scenarios.
  3. ML Model Development – Design of a lightweight, context-aware machine learning model using network and physiological features.
  4. Self-Healing Mechanism Design – Development of autonomous recovery strategies including node isolation and credential revocation.
  5. Evaluation – Performance evaluation based on detection accuracy, false positive rate, energy consumption, and latency.

Experiment Setup and Implementation

Results and Analysis

Conclusion

Publications