Dissection of Modern Malicious Software

• Main Author: Rodrigues, Fábio José Sousa
• Publication Date: 2014
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.6/6169
• topic: Análise Dinâmica; Análise Estática; Antivírus; Comportamento de Malware; Dissecação; Famílias de Malware; Malware; Software Malicioso; Stuxnet; Tinba/Zusy; Trojano; Vírus Informático; Worm

The exponential growth of the number of malicious software samples, known by malware in the specialized literature, constitutes nowadays one of the major concerns of cyber-security professionals. The objectives of the creators of this type of malware are varied, and the means used to achieve them are getting increasingly sophisticated. The increase of the computation and storage resources, as well as the globalization have been contributing to this growth, and fueling an entire industry dedicated to developing, selling and improving systems or solutions for securing, recovering, mitigating and preventing malware related incidents. The success of these systems typically depends of detailed analysis, often performed by humans, of malware samples captured in the wild. This analysis includes the search for patterns or anomalous behaviors that may be used as signatures to identify or counter-attack these threats. This Master of Science (Ms.C.) dissertation addresses problems related with dissecting and analyzing malware. The main objectives of the underlying work were to study and understand the techniques used by this type of software nowadays, as well as the methods that are used by specialists on that analysis, so as to conduct a detailed investigation and produce structured documentation for at least one modern malware sample. The work was mostly focused in malware developed for the Operating Systems (OSs) of the Microsoft Windows family for desktops. After a brief study of the state of the art, the dissertation presents the classifications applied to malware, which can be found in the technical literature on the area, elaborated mainly by an industry community or seller of a security product. The structuring of the categories is nonetheless the result of an effort to unify or complete different classifications. The families of some of the most popular or detected malware samples are also presented herein, initially in a tabular form and, subsequently, via a genealogical tree, with some of the variants of each previously described family. This tree provides an interesting perspective over malware and is one of the contributions of this programme. Within the context of the description of functionalities and behavior of malware, some advanced techniques, with which modern specimens of this type of software are equipped to ease their propagation and execution, while hindering their detection, are then discussed with more detail. The discussion evolves to the presentation of the concepts related to the detection and defense against modern malware, along with a small introduction to the main subject of this work. The analysis and dissection of two samples of malware is then the subject of the final chapters of the dissertation. A basic static analysis is performed to the malware known as Stuxnet, while the Trojan Banker known as Tinba/zuzy is subdued to both basic and advanced dynamic analysis. The results of this part of the work emphasize difficulties associated with these tasks and the sophistication and dangerous level of samples under investigation.