Vol 7 Issue 2 October 2020-March 2021
Coulibaly Fatoumata Siga, Zhang Jie Fu
Abstract: Stegomalware is analyzed using a steganographic algorithm which helps the programmers of malware with avenues for concealing malicious payloads. These algorithms have proven to be very efficient because most of the stegomalware deployed in the Android markets has yet to be identified, detected, or taken down. Encoding schemes such as Steganography provide traces of the text's calculation as an outcome of changes needed to embed the text information. This is as a result of digital assailants rising rapidly, and they need to curb them, i.e., recognizing whether the malicious object has or has not been embedded into the data. To do so, a Steganalysis approach is carried out to substantiate the real and forged Information, and the analysis will determine whether the data embedded is malicious or not. To establish if an application has a secret embedded payload, you have to determine the size of the payload in terms of length and establish the real payload. The concept of steganalysis is vital, holding the fact that the application of steganalysis on these stegamalware has proven to be efficient. In our research paper, we are using the generative adversarial network to analyze stegamalware in our android applications so that we can deliver secure applications free from malware. In our paper, we will use a secure generative adversarial network for Steganalysis. The model is secure and pontifically efficient in stegamalware generation, which is easily detectable by the discriminator and can substantiate the stegamalware and something else that isn't malware. We have evaluated the effectiveness and efficiency of the generative adversarial network from applications data trained for steganalysis. The promising results demonstrate that the model is useful, and its classification is very accurate.
Keywords: PSO-particle swarm optimization, CNN-convolutional neural networks, GA-Genetic Algorithm, ACO-Ant colony optimization, GANS-generative adversarial network, SSGAN-Secure Steganography Generative Adversarial network, SGANs-secure Generative adversarial network, NN-Neural networks, API-Application programming interface, EC-Evolutionary computation, EA-Evolutionary algorithm.
Title: Steganalysis on Android Applications using Generative Adversarial Networks
Author: Coulibaly Fatoumata Siga, Zhang Jie Fu
ISSN 2350-1022
International Journal of Recent Research in Mathematics Computer Science and Information Technology
Paper Publications
