Today's smartphones and tablets use different applications and software for all sorts of purposes: communication, entertainment, fitness, to share images with each other, to keep up to date with the news and lots of different daily tasks. With the heavy usage of all these apps, it is no wonder that it comes with a few issues. Private data is stored in large quantities both on the local device and on the app-creators' servers. It is no wonder that applications advertising user secrecy and transient storage of user data. One of these applications is Snapchat, with over 500 million downloads on Google Play store, at the time of writing. Snapchat is a communication application with the niched feature that the images and messages sent, disappear once opened or after 24 hours have passed. With the illusion of privacy behind Snapchats niche it has become a breeding ground for criminal activity. The niche itself translates to a troublesome hurdle for law enforcement trying to retrieve evidence from devices of Snapchat users. This paper is aimed to investigate these issues and perform a methodology to retrieve potential evidence on a device using Snapchat to send images and messages. By performing a physical acquisition on a test device and analyzing to find artifacts pertaining to Snapchat and the test-data that was created. The method is performed on a Samsung Galaxy S4 with Android 5.0.1 running Snapchat version 10.52.3.0. Test data such as different images and messages were created and attempted to be retrieved at three points in time. First one being right after data creation. Second one after a restart and 24 hours after the data was created. And the third with 48 hours passed and the Snapchat user logged out at the time of acquisition. The acquisition resulted in the extraction of several sent images and a full text conversation between the experimental device and another party. A full video which was uploaded by the receiving user was able to be extracted even though the experimental device never actually viewed the video. The second acquisition which was made when 24h had passed gave the same results as the first one. This meant that time at least up to a day after the initial creation of the data did not have any effect on the evidence. However, when the Snapchat user was logged out from the application, the data was then unobtainable and had disappeared. Presumably Snapchat has a function which deletes personal data about the user when logged out from the application. This function might become a hurdle in law enforcement investigations where the application Snapchat is involved.

This work deals with the design of distributed repository aimed at storing digital forensic data. The theoretical part of the thesis describes digital forensics and what is its purpose. There are also explained Big data, suitable storages, their properties, advantages and disadvantages, in this part. The main part of the thesis deals with the design and implementation of distributed storage for digital forensic data. The design is also focused in suitable indexing of stored data, and supporting new types of digital forensic data. The performance of implemented system was evaluated for chosen type of digital forensic data PCAP files.

This thesis explores the evolving landscape of digital forensics in the context of quantum computing advancements, which challenge the foundational integrity of digital evidence. The focus is on the globally recognized digital forensic guidelines, NIST SP 800-86 and ISO/IEC 27037:2012, and their capacity to safeguard evidence against the unique capabilities of quantum systems. This thesis identifies vulnerabilities within existing forensic models through a comprehensive document analysis and expert interviews and proposes strategic modifications to enhance their robustness. Key findings suggest that traditional digital forensic methodologies, while robust under current technological standards, must address quantum data’s multi-state, entanglement, and no-cloning properties, which can fundamentally alter digital evidence. The thesis advocates for a paradigm shift in forensic processes to incorporate quantum-resistant techniques that ensure the integrity and admissibility of evidence. Additionally, it highlights the necessity for ongoing education and collaborative research to effectively adapt digital forensics to this new technological era. This research contributes to the theoretical framework and practical applications of digital forensics, aiming to future-proof forensic practices against the disruptive nature of quantum computing.

<p>Synthesized speech may be used for nefarious purposes, such as fraud, spoofing, and misinformation campaigns. We present several speech forensics methods based on deep learning to protect against such attacks. First, we use a convolutional neural network (CNN) and transformers to detect synthesized speech. Then, we investigate closed set and open set speech synthesizer attribution. We use a transformer to attribute a speech signal to its source (i.e., to identify the speech synthesizer that created it). Additionally, we show that our approach separates different known and unknown speech synthesizers in its latent space, even though it has not seen any of the unknown speech synthesizers during training. Next, we explore machine learning for an objective in the aerospace domain.</p> <p><br></p> <p>Compared to conventional ballistic vehicles and cruise vehicles, hypersonic glide vehicles (HGVs) exhibit unprecedented abilities. They travel faster than Mach 5 and maneuver to evade defense systems and hinder prediction of their final destinations. We investigate machine learning for identifying different HGVs and a conic reentry vehicle (CRV) based on their aerodynamic state estimates. We also propose a HGV flight phase prediction method. Inspired by natural language processing (NLP), we model flight phases as “words” and HGV trajectories as “sentences.” Next, we learn a “grammar” from the HGV trajectories that describes their flight phase transition patterns. Given “words” from the initial part of a HGV trajectory and the “grammar”, we predict future “words” in the “sentence” (i.e., future HGV flight phases in the trajectory). We demonstrate that this approach successfully predicts future flight phases for HGV trajectories, especially in scenarios with limited training data. We also show that it can be used in a transfer learning scenario to predict flight phases of HGV trajectories that exhibit new maneuvers and behaviors never seen before during training.</p>