Mohammed Kaosar

For large, intricate, and multi-layered networks like that of Industrial IoT, an individual instance of intrusion detection system cannot efficiently work against advanced attack strategies. The reason is that it would not be aware of the overall context, environment, and relevant incidents in other networks. This necessitates a collaborative intrusion detection system that allows multiple intrusion detection systems to communicate with each other and share information on emerging cyber-attack incidents. Thus, immunizing themselves and preventing the attack from escalating. However, the main challenge here is to manage the trust among the peers, where an insider attacker may input false attack signatures to the network, thus degrading the performance. Hence, we propose a blockchain-based trustfree collaborative intrusion detection system, in which threat alert messages will only be propagated in the network after network consensus.

Network traffic classification (NTC) has attracted considerable attention in recent years. The importance of traffic classification stems from the fact that data traffic in modern networks is extremely complex and ever-evolving in different aspects, e.g. volume, velocity and variety. The inherent security requirements of Internet-based applications also highlights further the role of traffic classification. Gaining clear insights into the network traffic for performance evaluation and network planning purposes, network behavior analysis, and network management is not a trivial task. Fortunately, NTC is a promising technique to gain valuable insights into the behavior of the network, and consequently improve the network operations. In this paper, we provide a method based on deep ensemble learning to classify the network traffic in communication systems and networks. More specifically, the proposed method combines a set of Convolutional Neural Network (CNN) models into an ensemble of classifiers. The outputs of the models are then combined to generate the final prediction. The results of performance evaluation show that the proposed method provides an average accuracy rate of 98% for the classification of traffic (e.g., FTP-DATA, MAIL, etc.) in the Cambridge Internet traffic dataset.

Test data generation is a significant part of software and/or hardware testing. It is a complex problem and researchers have proposed various solutions to generate optimum number of test data in an acceptable polynomial time. However, most of the solutions are highly complex (NP-hard), interaction limitation and takes substantial time to generate test data set. Therefore, it is an open challenge to propose the best solution. This paper proposes a novel technique called MTTG (Multi-Tuple based T-way Test Generator) which relies on Kid's Card game strategy. The proposed technique finds optimum matching value by searching through all possible combination of similarity matching, based on data generation principle. Our empirical results demonstrate that the proposed MTTG is very efficient in test data generation based on time and interaction strength/level, compared to other existing strategies.

Concealing a message and ensuring its security is inevitable in data transmission. Among various concepts, one approach is steganography that encodes secret message in indiscernible way. In this paper, we present an audio steganographic technique and propose a novel approach to hide data in the least significant bit (LSB) of the stereo-audio samples with CD-quality. Here, on the basis of stego-key and its parity, message bits are encoded into cover audio samples. In terms of security and imperceptibility, this method is a significant improvement of LSB method for hiding information in audio.

This paper introduces a novel notion in the application protocol design paradigm for wireless sensor networks (WSNs). The traditional approaches of designing application protocols tend to focus primarily on developing the protocols first, and then using them on different topologies for implementation. We, however, argue that the logical topology of WSNs should be considered before designing application protocols. The argument is made on the basis that the logical topology of WSNs dictates the communication abstraction, the structure, and the hierarchy of the network. Thus, a well-designed logical topology helps in minimising the constraints of the WSNs and provides benefits to design various application protocols. In this paper we demonstrate how a well-designed logical topology influences the performances of protocols developed in WSNs. In doing so, the logical structure and the communication abstraction of the logical topology are used to design a number of application protocols, and their performances are evaluated.