No abstract

As novel technologies continue to reshape the digital era, cyberattacks are also increasingly becoming more commonplace and sophisticated. Distributed denial of service (DDoS) attacks are, perhaps, the most prevalent and exponentially-growing attack, targeting the varied and emerging computational network infrastructures across the globe. This necessitates the design of an efficient and early detection of large-scale sophisticated DDoS attacks. Software defined networks (SDN) point to a promising solution, as a network paradigm which decouples the centralized control intelligence from the forwarding logic. In this work, a deep convolutional neural network (CNN) ensemble framework for efficient DDoS attack detection in SDNs is proposed. The proposed framework is evaluated on a current state-of-the-art Flow-based dataset under established benchmarks. Improved accuracy is demonstrated against existing related detection approaches. INDEX TERMS Software defined network (SDN), anomaly detection, distributed denial of service (DDoS), deep learning, deep convolutional neural network (CNN).

No abstract

Cloud computing has emerged as a long-dreamt vision of the utility computing paradigm that provides reliable and resilient infrastructure for users to remotely store data and use on-demand applications and services. Currently, many individuals and organizations mitigate the burden of local data storage and reduce the maintenance cost by outsourcing data to the cloud. However, the outsourced data is not always trustworthy due to the loss of physical control and possession over the data. As a result, many scholars have concentrated on relieving the security threats of the outsourced data by designing the Remote Data Auditing (RDA) technique as a new concept to enable public auditability for the stored data in the cloud. The RDA is a useful technique to check the reliability and integrity of data outsourced to a single or distributed servers. This is because all of the RDA techniques for single cloud servers are unable to support data recovery; such techniques are complemented with redundant storage mechanisms. The article also reviews techniques of remote data auditing more comprehensively in the domain of the distributed clouds in conjunction with the presentation of classifying ongoing developments within this specified area. The thematic taxonomy of the distributed storage auditing is presented based on significant parameters, such as scheme nature, security pattern, objective functions, auditing mode, update mode, cryptography model, and dynamic data structure. The more recent remote auditing approaches, which have not gained considerable attention in distributed cloud environments, are also critically analyzed and further categorized into three different classes, namely, replication based, erasure coding based, and network coding based, to present a taxonomy. This survey also aims to investigate similarities and differences of such a framework on the basis of the thematic taxonomy to diagnose significant and explore major outstanding issues.

No abstract