Search

Security and privacy in the Internet of Things (IoT) other significant challenges, primarily because of the vast scale and deployment of IoT networks. Blockchain-based solutions support decentralized protection and privacy. In this study, a private blockchain-based smart home network architecture for estimating intrusion detection empowered with a Fused Real-Time Sequential Deep Extreme Learning Machine (RTS-DELM) system model is proposed. This study investigates the methodology of RTS-DELM implemented in blockchain-based smart homes to detect any malicious activity. The approach of data fusion and the decision level fusion technique are also implemented to achieve enhanced accuracy. This study examines the numerous key components and features of the smart home network framework more extensively. The Fused RTS-DELM technique achieves a very significant level of stability with a low error rate for any intrusion activity in smart home networks. The simulation findings indicate that this suggested technique successfully optimizes smart home networks for monitoring and detecting harmful or intrusive activities. [ABSTRACT FROM AUTHOR]

Most Intrusion Detection Systems (IDS) nowadays are signature-based. They are very useful and accurate for detecting known attacks. However, they are inefficient with unknown attacks. Moreover, most of cyber attacks start with malicious URLs. Attackers try to trick users into clicking on malicious URLs. This gives attackers an easy way to launch attacks. To defend against this, companies and organizations use IDS/IPS to detect malicous URLs using blacklist of URLs. This is very efficient with known malicious URLs, but useless with unknown malicious URLs. To overcome this problem, a number of malicious Web site detection systems have been proposed. One of the most promising methods is to apply machine learning detection techniques. In this paper, we present a new lexical approach to classify URLs by using machine learning techniques which patternize the URLs. Our approach is based on natural language processing features which use word vector representation and ngram models on the blacklist word as the main features. Using this technique can help classifier distinguish benign URLs from malicious ones. Our experimentation shows that our approach can achieve a high degree of accuracy at 97.1% in the case of SVM. Moreover, it can maintain a high level of robustness with 0.97 precision and 0.93 recall scores. [ABSTRACT FROM AUTHOR]

In a Personal Area Network (PAN) populated with wearable devices, the security of stored files is an important concern. Secret sharing has received considerable attention as an effective approach for handling security as it guarantees confidentiality and integrity. A high level of data privacy is still assured when devices are compromised, lost, or stolen, and even in cases where the total count of these devices is less than the minimum threshold. However, because traditional secret sharing requires tremendous computational overhead and consumes large storage space, it is suboptimal for battery-powered wearable devices. To overcome these limitations, combinatorial-based file sharing is proposed. However, this approach is also hindered by an efficiency problem, because the preparation of file storage and retrieval involves computational costs. Additionally, this file sharing scheme was designed without taking into consideration an environment with heterogeneous wearable devices. To address the aforementioned problems, we first propose a new fog network model that delegates the calculation for the preparation of file storage and retrieval to a fog node, such as a smartphone or tablet PC. A fog node has comparatively higher computational speed, and a larger battery capacity; hence, the resources of the devices in the PAN can be managed efficiently. Additionally, we propose a new algorithm that facilitates file shares by considering the heterogeneous characteristics of wearable devices. In particular, we consider the factors of storage capacity and network speed to determine the size of the file shares. We use mathematical analysis and simulations to demonstrate that our proposed model efficiently manages stored files. [ABSTRACT FROM AUTHOR]

The adoption of network flow in the domain of Network-based Intrusion Detection System (NIDS) has steadily risen in popularity. Typically, NIDS detects network intrusions by inspecting the contents of every packet. Flow-based approach, however, uses only features derived from aggregated packet headers. In this paper, all publicly accessible and labeled NIDS data sets are explored. Following the advances in deep learning techniques, the performances of Long Short-Term Memory (LSTM) are also presented and compared with various machine learning classifiers. Amongst the reviewed data sets, the models are trained and evaluated on CIDDS-001 flow-based data set. [ABSTRACT FROM AUTHOR]

Anomaly-based intrusion detection system (IDS) is gaining wide attention from the research community, due to its robustness in detecting and profiling the newly discovered network attacks. Unlike signature-based IDS which solely relying on a set of pre-defined rules through some massive human efforts, anomaly-based IDS utilises the collected network traces in building its own classification model. The classification model can optimised when a large set of network traces is available. The ideal way of pooling the network traces is through database sharing. However, not many organisations are willing to release or share their network databases due to some privacy concerns, i.e. to avoid some kinds of internet traffic behaviour profiling. To address this issue, a number of anonymisation techniques was developed. The main usage of anonymisation techniques is to conceal the potentially sensitive information in the network traces. However, it is also important to ensure the anonymisation techniques are not over abusing the performances of IDS. To do so, the convention way is by using a Snort IDS to measure the number of alarms generated before-and-after an anonymisation solution is applied. However, this approach is infeasible for Anomaly-Based IDS. Thus, an alternative way of using machine learning approach is proposed and explored in this manuscript. Instead of manual evaluation through the usage of Snort IDS, a J48 decision tree (Weka package of C4.5 algorithm) is used. In this manuscript, two anonymisation techniques, (1) black-marker, and (2) bilateral classification are used to hide the value of port numbers; and their before-and-after performances are evaluated through a J48 decision tree. [ABSTRACT FROM AUTHOR]