Muhammad Fermi Pasha

EFIS: evolvable-neural-based fuzzy inference system and its application for adaptive network anomaly detection

This paper presents an application of a new type of fuzzy inference system, denoted as evolvable-neural-based fuzzy inference system (EFIS), for adaptive network anomaly detection in the presence of a concept drift problem. This problem cannot be avoided to happen in every network. It is a problem of modeling the behavior of normal traffic while it keeps changing over time in continuous manner. EFIS can solve the concept drift problem by having dynamic network traffic profile creation and adaptation. The profile is then being further used to detect anomaly. An enhanced evolving clustering method (ECMm), which is employed by EFIS for online network traffic clustering, is also presented. It is demonstrated, through experiments, that EFIS can evolve in a growing network and also successfully detect network traffic anomalies.

Vertical Handover Scheme for Car-to-Car Communication Based on IEEE 802.21 Standard

Car-to-car communication starting to became important in the attempt to improve road safety. Not only it is able to reduce car accident rate, but also it is useful to minimize traffic jam. The communication media between the cars varied but the low cost solution is to use existing wireless networks such as 3G, Wi-Fi, WiMAX, etc. Car-to-car communication implementation over these existing heterogeneous wireless networks requires a fast and reliable handovers when one network becomes unavailable. This paper presents a study on applicability of using the IEEE 802.21 standard as the handover protocol to expedite vertical handover in a car-to-car communication. A fast and simple car-to-car communication handover scheme is proposed and analyzed to show the effectiveness and how it can be implemented further.

A Feasibility Study Scheme of an Android-based Integrated Wearable ECG Monitoring System☆

Abstract Ubiquitous healthcare system is becoming essential in the attempt to improve healthcare delivery. Not only it is able to facilitate two-way communication between physician and patient, but it is also useful for the caregivers to maximize efficiency in running daily activities. Managing ischemia heart disease, one of the deadly diseases, would definitely benefit from such ubiquitous healthcare system. The current treatment method for ischemia requires patients to be physically present at their respective treatment centres for regular monitoring. This increases the need for remote monitoring method, due to the urgent nature of treating ischemia heart attack. Therefore, this paper presents preliminary system architecture and feasibility study of HeartPals , a low-cost Android-based integrated wearable electrocardiography (ECG) monitoring system that enables ischemia patients to be remotely monitored by their physicians. We also propose TOFUSER, a T echnology- O riented F easibility st U dy S cheme for E nd-use R s as an extended feasibility study scheme to allow users to verify the preliminary requirements of the technology and users’ requirements of HeartPals .

Liver tumor segmentation using kernel-based FGCM and PGCM

Low contrast between tumor and healthy liver tissue is one of the significant and challenging features among others in the automated tumor delineation process. In this paper we propose kernel based clustering algorithms that incorporate Tsallis entropy to resolve long range interactions between tumor and healthy tissue intensities. This paper reports the algorithm and its encouraging results of evaluation with MICCAI liver Tumor Segmentation Challenge 08 (LTS08) dataset. Work in progress involves incorporating additional features and expert knowledge into clustering algorithm to improve the accuracy.

iNet-Grid: A real-time Grid monitoring and troubleshooting system

The purpose of Grid monitoring and management is to monitor services in Grid environment for fault detection, performance analysis, performance tuning, load balancing and scheduling. This paper emphasis on presenting a new framework namely iNet-Grid deployed for Grid monitoring and troubleshooting purposes. The iNet-Grid is integrated on top of Ganglia. iNet-Grid has been tested and successfully accomplished on USM network with the preliminary results have shown the positive outcomes.

Adaptive Real-Time Network Monitoring System: Detecting Anomalous Activity with Evolving Connectionist System

When diagnosing network problems, it is desirable to have a view of the traffic inside the network. This can be achieved by profiling the traffic. A fully profiled traffic can contain significant information of the network’s current state, and can be further used to detect anomalous traffic and manage the network better. Many has addressed problems of profiling network traffic, but unfortunately there are no specific profiles could lasts forever for one particular network, since network traffic characteristic always changes over and over based on the sum of nodes, software that being used, type of access, etc. This paper introduces an online adaptive system using Evolving Connectionist Systems to profile network traffic in continuous manner while at the same time try to detect anomalous activity inside the network in real-time and adapt with changes if necessary. Different from an offline approach, which usually profile network traffic using previously captured data for a certain period of time, an online and adaptive approach can use a shorter period of data capturing and evolve its profile if the characteristic of the network traffic has changed.

Enhanced local binary pattern for chest X-ray classification

The Local Binary Pattern (LBP) descriptor encodes the complementary information of the spatial patterns and intensity variations in a local image neighborhood. The richness of this multidimensional information offers many possible variations to the encoding process. Taking advantage of this, several variants of the LBP have been proposed. This work attempts to further optimize the discriminative power of the LBP specifically for the medical image classification task. It proposes an LBP variant that takes into account the salient edge features that are found particularly in chest X-ray images. In addition, it introduces a semi-global histogram to replace the commonly used global histogram which normally represents the spatial distribution of the generated codes in the LBP encoding process. The proposed LBP variant has been applied to the task of classifying X-rays of the ImageCLEFmed 2009 dataset into different chest categories. Experimental results show that while reducing the computational load the proposed LBP variant achieved an accuracy of 99.19% as compared to the best reported LBP based results of 98.73%, in classifying chest x-ray images of the ImageCLEFmed 2009 dataset.

Profiling the features of pre-segmented healthy liver CT scans: Towards fast detection of liver lesions in emergency scenario

Automating the detection of lesions in liver CT scans requires a high performance and robust solution. With CT-scan start to become the norm in emergency department, the need for a fast and efficient liver lesions detection method is arising. In this paper, we propose a fast and evolvable method to profile the features of pre-segmented healthy liver and use it to detect the presence of liver lesions in emergency scenario. Our preliminary experiment with the MICCAI 2007 grand challenge datasets shows promising results of a fast training time, ability to evolve the produced healthy liver profiles, and accurate detection of the liver lesions. Lastly, the future work directions are also presented.

A Distributed Approach of Intelligent Network Traffic Monitoring and Anomaly Detection Application

Monitoring a large corporate network connecting thousands of computers which generate billions of packets everyday is a challenge and difficult task! This paper proposes a distributed approach of intelligent network traffic monitoring and anomaly detection system. By utilizing a distributed client-server scheme, our proposed system can monitor multiple network segments and distribute the workload among the intelligent clients to monitor and detect anomaly. In this way, the complexity of having analyzing enormous traffic at once can be reduced. The servers primary task is only to manage all different profiles from different network segment used by the intelligent clients in their respective segments