Abdullah Sharaf Alghamdi

Application of artificial neural network in detection of DOS attacks

A solo attack may cause a big loss in computer and network systems, its prevention is, therefore, very inevitable. Precise detection is very important to prevent such losses. Such detection is a pivotal part of any security tools like intrusion detection system, intrusion prevention system, and firewalls etc. Therefore, an approach is provided in this paper to analyze denial of service attack by using a supervised neural network. The methodology used sampled data from Kddcup99 dataset, an attack database that is a standard for judgment of attack detection tools. The system uses multiple layered perceptron architecture and resilient backpropagation for its training and testing. The developed system is then applied to denial of service attacks. Moreover, its performance is also compared to other neural network approaches which results more accuracy and precision in detection rate.

Resource Allocation for Service Composition in Cloud-based Video Surveillance Platform

Resource allocation play an important role in service composition for cloud-based video surveillance platform. In this platform, the utilization of computational resources is managed through accessing various services from Virtual Machine (VM) resources. A single service accessed from VMs running inside such a cloud platform may not cater the application demands of all surveillance users. Services require to be modeled as a value added composite service. In order to provide such a composite service to the customer, VM resources need to be utilized optimally so that QoS requirements is fulfilled. In order to optimize the VM resource allocation, we have used linear programming approach as well as heuristics. The simulation results show that our approach outperforms the existing VM allocation schemes in a cloud-based video surveillance environment, in terms of cost and response time.

Smart Environment as a Service: Three Factor Cloud Based User Authentication for Telecare Medical Information System

The Telecare Medical Information System (TMIS) provides a set of different medical services to the patient and medical practitioner. The patients and medical practitioners can easily connect to the services remotely from their own premises. There are several studies carried out to enhance and authenticate smartcard-based remote user authentication protocols for TMIS system. In this article, we propose a set of enhanced and authentic Three Factor (3FA) remote user authentication protocols utilizing a smartphone capability over a dynamic Cloud Computing (CC) environment. A user can access the TMIS services presented in the form of CC services using his smart device e.g. smartphone. Our framework transforms a smartphone to act as a unique and only identity required to access the TMIS system remotely. Methods, Protocols and Authentication techniques are proposed followed by security analysis and a performance analysis with the two recent authentication protocols proposed for the healthcare TMIS system.

Enhancing SVM performance in intrusion detection using optimal feature subset selection based on genetic principal components

Intrusion detection is very serious issue in these days because the prevention of intrusions depends on detection. Therefore, accurate detection of intrusion is very essential to secure information in computer and network systems of any organization such as private, public, and government. Several intrusion detection approaches are available but the main problem is their performance, which can be enhanced by increasing the detection rates and reducing false positives. This issue of the existing techniques is the focus of research in this paper. The poor performance of such techniques is due to raw dataset which confuse the classifier and results inaccurate detection due to redundant features. The recent approaches used principal component analysis (PCA) for feature subset selection which is based on highest eigenvalues, but the features corresponding to the highest eigenvalues may not have the optimal sensitivity for the classifier due to ignoring many sensitive features. Instead of using traditional approach of selecting features with the highest eigenvalues such as PCA, this research applied a genetic algorithm to search the genetic principal components that offers a subset of features with optimal sensitivity and the highest discriminatory power. The support vector machine (SVM) is used for classification purpose. This research work used the knowledge discovery and data mining cup dataset for experimentation. The performance of this approach was analyzed and compared with existing approaches. The results show that proposed method enhances SVM performance in intrusion detection that outperforms the existing approaches and has the capability to minimize the number of features and maximize the detection rates.

Folksonomy-based personalized search and ranking in social media services

In recent years, social Web users have been overwhelmed by the huge numbers of social media available. Consequentially, users have trouble finding social media suited to their needs. To help such users retrieve useful social media content, we propose a new model of tag-based personalized searches to enhance not only retrieval accuracy but also retrieval coverage. By leveraging social tagging as a preference indicator, we build two models: (i) a latent tag preference model that reflects how a certain user has assigned tags similar to a given tag and (ii) a latent tag annotation model that captures how users have tagged a certain tag to resources similar to a given resource. We then seamlessly map the tags onto items, depending on an individual users query, to find the most desirable content relevant to the users needs. Experimental results demonstrate that the proposed method significantly outperforms the state-of-the art algorithms and show our methods feasibility for personalized searches in social media services.

Adaptive interaction support in ambient-aware environments based on quality of context information

Ambient-aware environments are technologically augmented with myriad sensors, devices and other emerging services in order to support users. However, users find it complex in interacting with such environments due to the presence of numerous devices and services. In this situation, providing context-aware implicit or automatic interaction support may help reducing the cognitive load of the users and facilitate easy access of the available devices and services. However, due to the imprecision in context information, implicit interactions performed by the environment often leads to mis-automation. The result of such impaired implicitness causes distrust and dissatisfaction to the user. In order to address this issue, we propose a system that considers quality of information to dynamically adjust the level of implicit interaction and allows a system to operate in different modes such as full-automation, action suggestion, simple notification, or null action. We conduct experiment in a smart home scenario in order to elicit users’ acceptance and trust regarding the proposed system. Our experiment shows that dynamic and alternative mode of interaction not only increases the satisfaction of users but also helps to avoid distrust in automated actions carried out by the environment under varying contexts.

A scalable and elastic cloud-assisted publish/subscribe model for IPTV video surveillance system

In this paper, we present a scalable and elastic content-based publish/subscribe model over cloud computing platform to support a smart, flexible and ubiquitous IPTV video surveillance system. Through this system, users of a surveillance system can subscribe to many surveillance events and receive video streams as a notification of new event occurring. This has direct impact on the way surveillance activities are carried out in different application domains including public safety and security, healthcare surveillance, etc. In the publish/subscribe model, it is challenging to match the events with the subscriptions efficiently that contains a large number of live contents. Existing algorithms on event matching are not very effective in the case of range predicates in subscriptions that are commonly used in IPTV video surveillance-based healthcare system and other areas. This paper addresses the aforementioned issue and propose an elastic and scalable algorithm for event matching in IPTV video surveillance over cloud platform. We also show the performance assessment of the proposed event matching algorithm in cloud-based IPTV video surveillance scenario and compare with various state-of-the-art approaches.

A Common Information Exchange Model for Multiple C4I Architectures

Information exchange among different architectures has been a massive challenge from the last three decades. Difficulty usually arises when an ideal model that deals with the precise exchange of data among different architectures is needed. Considering the case of military architecture frameworks on a common battle field, exchange of efficient and most precise information is the key for a perfect decision. Data exchange within different defense nodes is still a challenge. In this paper, we propose a general reference model to integrate different Command Control Communication Computer & Intelligent (C4I) systems operating under different defense forces of different coalition groups involved in a common-battle-field. The purpose of this study is to introduce a general reference model utilizing W3C Semantic Web Specification technique for meta-data models based on a software engineering method to develop and maintain usable and precise domain ontology in order to overcome interoperability issues of different C4I architectures exchanging information on a common battle field.

Bio-chaotic Stream Cipher-Based Iris Image Encryption

Conventional cryptography uses encryption key, which are long bit strings and are very hard to memorize such a long random numbers. Also it can be easily attacked by using the brute force search or technique. Instead of traditional cryptography, biometric e.g. fingerprint, iris, face, voice etc uniquely identifies a person and a secure method for stream cipher, because Biometric characteristics are ever living and unstable in nature (with respect to recognition). In this paper we used the idea of bio-chaotic stream cipher which encrypts the images over the electronic media by using a biometric key and a bio-chaotic function. It enhances the security of the images and it should not be compromised. The idea also gives birth to a new kind of stream cipher named bio-chaotic stream cipher. The paper also describes how to generate a key from a biometric string and how to encrypt and decrypt the desired data by using the bio-chaotic function.

Comparative Analysis of Intrusion Detection Approaches

Information security is a serious issue especially in present age because a solo attack may cause a big harm in computer and network systems. Several intrusion detection approaches exist to tackle this critical issue but the problem is which one is more suitable in the field of intrusion. Further, these approaches are used in intrusion detection systems. Therefore, in this paper, we evaluated them so that a suitable approach may be advised to intrusion detection systems. This work describes the concepts, tool and methodology being used for evaluation analysis of different intrusion detection approaches using multi-criteria decision making technique. Moreover, conclusion on results is made and direction for future works is presented.