Computer Sciences and Information Technology
Title : | Robust Forensic Analysis of Unmanned Aerial Vehicle using Federated Learning and Deobfuscation Techniques |
Area of research : | Computer Sciences and Information Technology |
Focus area : | Cybersecurity, Artificial Intelligence |
Principal Investigator : | Mr. Vinesh Kumar Jain, Engineering College Ajmer, Rajasthan |
Timeline Start Year : | 2023 |
Timeline End Year : | 2026 |
Contact info : | vineshjain@ecajmer.ac.in |
Details
Executive Summary : | Unmanned aerial vehicles (UAVs) have gained popularity due to their flexibility and low cost deployment, but they have also been used for criminal activities such as invasions of sensitive places, smuggling illegal goods, and spying on individuals or countries. UAV forensics is a relatively unstudied area, with some researchers proposing frameworks to gather information from seized UAVs to create a trail of hints. However, these frameworks struggle against anti-forensic techniques like code obfuscations. The absence of standards for UAV file formats makes it difficult to examine any UAV. Flight logs for a single UAV's individual flights are stored in DAT files, which can lead to missing and corrupted values, making it challenging to accurately estimate flight trajectories. Therefore, attempts to build reliable procedures, specific forensic frameworks, and UAV-based forensic technologies are crucial. This project aims to design and develop a robust framework for forensic analysis of UAVs that can recover data from obfuscated applications, files of different formats, and corrupted flight trajectories to ensure the accuracy of forensic investigations. The design objectives include designing deobfuscation techniques based on Satisfiability Modulo Theories (SMT) based program analysis for obfuscated parts of applications, developing improved federated learning-based forensic techniques to extract useful information from different file formats/components of UAVs, and designing a probabilistic solution to recover flight trajectories and logs to find the source and path of UAVs. This research will help establish a scientific research framework for using current digital forensics methods against consumer- and enterprise-level UAVs. |
Co-PI: | Dr. Chandra Prakash Jain, Engineering College Ajmer, Rajasthan-305025, Dr. Rekha Mehra, Engineering College Ajmer, Rajasthan-305025, Dr. Jyoti Gajrani, Engineering College Ajmer, Rajasthan-305025 |
Total Budget (INR): | 28,90,250 |
Organizations involved