Search

Malware detection is important for computer security. However, existing signature-based malware detection systems are still not quite perfect because they are designed to recognize already established patterns of malicious codes. In this work, a malware prediction model was developed to intelligently discover future malware strains, in order to improve the capability of malware detection system. The paradigm imbibed for this method includes "malware in vision context". Particularly, the method involved the generation of new data points from a malware data distribution using generative adversarial network (GAN), parameterized with fully-connected neural network (FCN) architecture. The developed model generates malware images from a 100-dimensional Gaussian noise distribution and learns to distinguish it from real malware images. The generated malware is similar but not the same as the real malware, as it consists modified features when compared with real malware. To establish the feasibility of the proposed method for malware research, an experiment was conducted by leveraging Mallmg, an image-based malware dataset. Due to certain technical constraints as discussed in the study, 52.83% of Mallmg dataset was used to generate new malware data which yielded 224.98%, amounting to 98.66% of the original dataset. Metrics such as Mean Squared Error (MSE), Structural Similarity Index (SSIM) and a customized enhancer (ABV) were used evaluate the generated images. The best scores obtained for MSE, SSIM andABV are 0.02, 0.91 and 1.00 respectively while the worst scores are 0.07, 0.02, 0.68 respectively. Also, the uniqueness of the generated malware was established. These metrics showcase an exemplary, yet simplistic approach to malware prediction and data augmentation.

The popularity and open nature of cloud services has made them an attractive platform for cyber attacks. The increase in activities of attackers through Distributed Reflection Denial of service (DRDoS) attack is particularly alarming. There have been many reported cases of denial of service attacks on many cloud services in recent times. Most of these attacks capitalize on the open nature of cloud services and the imperfections of existing security measures. In this paper, a scheme called Common Source IP (COMSOUIP) and Incremental Replacement Strategy (IRS) are proposed to mitigate against Distributed Reflection Denial of Service (DRDoS) attack in the cloud database offered as a service. The idea presented was about reducing the surface area of attack by pushing back the attack to the attacker’s network. In COMSOUIP, attack is detected by capturing ip addresses of all machines that communicate by making service call request to the cloud database service to determine machines that have a common ip source with the command and control machine. In IRS scheme, a new set of virtual machines are used to replace the compromised machines in gradual manner. The results of the schemes evaluation show that the schemes would be effective in guarding against Distributed Reflection Denial of Service (DRDoS) attack.

Secure group communication systems typically rely on a group key, a secret shared by all members of the group. Confidentiality is provided by encrypting all messages with the group key. Meanwhile, some of the associated problems with securing group communications are rekeying when membership changes and multiple sub-group key storage by users. In this paper, an Efficient Broadcast Encryption Scheme (EBES) which uses key encapsulation mechanism was designed to eliminate these problems. Symmetric cryptosystem (AES 128), Elliptic Curve Cryptosystem in prime order Galois Field GFp simulated over the Elgammal Cryptosystem (EC-ELGAMAL) and the Chinese Remainder Theorem (CRT) are the major components of our solution. The design also introduced the concept of mutual authentication into group communications using cryptographic keys. The authentication was performed smoothly using the Elliptic Curve Digital Signature Scheme (ECDSS). The security of our solution was formally proven by reduction to the security of EC-ELGAMAL cryptosystem, which is based on Elliptic Curve Discrete Logarithm Problem (ECDLP). This paper is concluded with a set of experimental results that support our research.