@article { author = {Elkilani, W}, title = {A SECURE ARCHITECTURE FOR DEFENDING ARP SPOOFING ATTACKS USING A SWITCH BASED METHODOLOGY}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {1-19}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15759}, abstract = {In local networks, security weaknesses in the data link layer enable internal attacks. Althoughswitches and routers have some built in security features, they are not enough to fully ensure thesecurity of local networks. Moreover, these features require network administrators' involvement andare prone to miss-configuration. In addition, data link layer protocols used in local area networks(LANs) are not designed with built-in security features. The most dangerous attacks on layer 2 are ARPspoofing and MAC flooding attacks. Several schemes to mitigate, detect and prevent these attacks havebeen proposed, but each has its limitations. This paper proposes a detection and prevention system forARP spoofing attacks. The system consists of two back to back servers. An application on the serversallows authentication of users to a centralized server. The server, in turn, retrieves logged users to theswitch. Hence filtering of untrusted users is performed by telneting the main switch. The Performancestudy has shown the efficiency and superiority of the proposed system, as compared to the previouswork. Several performance metrics have been measured to show its fast response to detection andprevention of the ARP spoofing attacks. The system has been compared to one of the famouscommercial tools. The comparison has shown the superiority of our system, since the system detectiontime is 20 time faster than that of the commercial tool.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15759.html}, eprint = {} } @article { author = {Girgis, M and Mahmoud, T and Eliwa, E}, title = {MEDICAL DECISION SUPPORT SYSTEM FOR HEPATITIS C VIRUS PREDICTION USING DATA MINING TECHNIQUES}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {21-35}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15760}, abstract = {The healthcare environment is generally perceived as being ‘information rich’ yet ‘knowledgepoor’. Which, unfortunately, are not “mined” to discover hidden information for effective decisionmaking by healthcare practitioners. The health-care knowledge management can be improved throughthe integration of data mining and decision support. In this paper, we present a prototype Hepatitis CVirus Decision Support System (HCVDSS) that uses three data mining classification techniques,namely, Decision Trees, Naïve Bayes and Neural Network. Results show that each technique has itsown strength in realizing the objectives of the defined mining goals. HCVDSS can answer complex“what if” queries. Using medical profiles such as gender, residence, Alt and Ast the proposed HCVDSScan predict the likelihood of patients getting HCV disease. It enables significant knowledge, e.g.,patterns, relationships between medical factors related to HCV disease, to be established. The proposedHCVDSS, which is implemented on the .Net platform, is windows application, user-friendly, scalable,reliable and expandable.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15760.html}, eprint = {} } @article { author = {Tarrad, I}, title = {PERFORMANCE EVALUATION OF HYBRID PTS-SLM TECHNIQUES VERSUS CONVENTIONAL PTS}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {37-50}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15763}, abstract = {Orthogonal Frequency Division Multiplexing (OFDM) is an attractive technique for wirelesscommunication over frequency-selective fading channels. Suffer from high Peak-to-Average PowerRatio (PAPR) limiting OFDM usage and reduces the efficiency of High Power Amplifier (HPA) orworse degrade BER. Many PAPR reduction techniques have been proposed in the literature, amongthem Partial Transmit Sequence (PTS) and Selected Mapping (SLM) are highly successful PAPRreduction techniques. However, the highly computational complexities of these techniques limit theirPAPR reduction capability. Recently different hybrid schemes combine PTS and SLM aim to reduce thecomputational complexity or obtain a better PAPR reduction performance compared with conventionalPTS. In this paper, a fairly comparison between PTS and the other hybrid techniques is done,considering technique that can reduce number of required additions and multiplications or at least oneof them less than PTS, as the best technique among others. Results show that only SH which canachieve this condition while the other hybrid techniques can't achieve that.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15763.html}, eprint = {} } @article { author = {Selem, E and Khedr, A and Nassar, H}, title = {HM-LEACH: ENERGY EFFICIENT HETEROGENEOUS MOBILE LEACH PROTOCOL FOR HETEROGENEOUS WIRELESS SENSOR NETWORKS}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {51-61}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15764}, abstract = {in wireless sensor networks (wsns) sensor nodes with sensing, computing and communicationinfrastructure are randomly deployed and organized as clusters. most of the existing sensor networksfocus on homogeneous in which the cluster heads are changed periodically. to improve the lifetime ofenergy constraint battery powered wsn; clustered heterogeneous sensor networks are analyzed. in thispaper, we study the impact of heterogeneity of nodes in terms of their energy in mobile wsns that arehierarchically clustered. we assume that a percentage of the population of sensor nodes is equippedwith the additional energy resources. we also assume that the sensor nodes are randomly distributedand are mobile, the coordinates of the sink and the dimensions of the sensor field are known.homogeneous clustering protocols assume that all the sensor nodes are equipped with the same amountof energy and as a result, they cannot take the advantage of the presence of node heterogeneity.adapting this approach, we introduce an energy efficient heterogeneous mobile leach protocol forheterogeneous wsns based on weighted election probabilities of each node to become a cluster headaccording to the residual energy in each node. Finally, the simulation results demonstrate that ourproposed heterogeneous clustering approach is more effective in prolonging the network lifetime andreduce energy consumption compared with m-leach protocol.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15764.html}, eprint = {} } @article { author = {Amin, Ahmed}, title = {AUTOMATIC MACHINE FAULT DIAGNOSIS BASED ON WAVELET TRANSFORM AND PROBABILISTIC NEURAL NETWORKS}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {63-79}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15766}, abstract = {Machinery parts always put imprint on the product during the production processing.Industry in the developing country the discovery of defects depends on the human experience andspectrum analysis (SA). Fast Fourier Transform (FFT) is basis for SA and uses to extract the frequencyfeatures which help the experts to identifying the causes of defected parts in machine. In this paperpresents a new technique to automatic fault diagnosis. The proposed technique is constituted of twostages architecture: the first stage is analysis the product signals to extract the features by usingwavelet transform (WT). The second stage is devoted to the classification of defect from the features byusing probabilistic neural network (PNN). Naïve Bayesian algorithm and Bayesian net algorithm istaken for classification and compared. The novelty of the proposed method resides in the ability notonly with higher precision, but also with dimensionality reduction and higher speed than method ofFourier transform and mathematical statistics.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15766.html}, eprint = {} } @article { author = {Koriem, A and Tarrad, I and Farahat, H}, title = {PETRI NET PERFORMANCE AND RELIABILITY MODELS FOR WIRELESS NETWORKS}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {81-107}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15768}, abstract = {Since the dynamic behavior of wireless networks inherits different communication patternsfor the execution processes of the desired parallel tasks and the occurrence processes of handoffoperations, the available reliability metrics are not applicable to study and analyze such networks.Therefore, in this paper, we propose a novel Petri net modeling methodology for accurately calculatingthe reliability of wireless networks that incorporate theses aspects, such as wireless CORBA network.The proposed reliability methodology is developed based on using the End-to-End ExpectedInstantaneous Reliability (EIR) measure. The EIR has the capability to measure the reliability ofwireless networks that incorporate the handoff processes as well as the service processes. In theproposed technique, we firstly build Generalized Stochastic Petri Net (GSPN) performance model fordescribing the different communication pattern that can occur among the various network resourcesunder the execution of parallel tasks. Secondly, we build GSPN performance model for describing thehandoff communication processes among the various network resources. Finally, we build genericreliability model consists of the following parts: the first part describes the execution of the desiredparallel tasks, the second part illustrates the mobility characteristic introduced by handoff processes,and the third part describes the dynamic behavior of failure processes of network resources. We studyand analyze the various reliability results as well as the performance results obtained from thedeveloped GSPN models under the effect of communication distance among the CORBA networkresources.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15768.html}, eprint = {} } @article { author = {El-Dahshan, K and Masameer, E and Elhabshy, Ab}, title = {A NEW ENHANCED AUTHENTICATED KEY AGREEMENT PROTOCOL}, journal = {International Journal of Intelligent Computing and Information Sciences}, volume = {14}, number = {1}, pages = {109-115}, year = {2014}, publisher = {Ain Shams University, Faculty of Computer and Information Science}, issn = {1687-109X}, eissn = {2535-1710}, doi = {10.21608/ijicis.2014.15769}, abstract = {In 2010, Deng and Deng investigated several Seo-Sweeny related protocols and presented anew improvement to Tseng’s protocol to avoid discovered attacks such as modification and off-linepassword guessing attacks. In this paper, we show that Deng-Deng protocol is still vulnerable to offlinepassword guessing attack, and then we present an enhancement to prevent such attacks.}, keywords = {}, url = {https://ijicis.journals.ekb.eg/article_15769.html}, eprint = {} }