Vol. 33 · No. 1

In this paper, we present an effective tree based subspace clustering technique (TreeCLUSS) for finding clusters in network intrusion data and for detecting known as well as unknown attacks without using any labelled traffic or signatures or training. To establish its effectiveness in finding the appropriate number of clusters, we perform a cluster stability analysis. We also introduce an effective cluster labelling technique (CLUSSLab) to label each cluster based on the stable cluster set obtained from TreeCLUSS. CLUSSLab is a multi-objective technique that employs an ensemble approach for labelling each stable cluster generated by TreeCLUSS to achieve high detection rate. We also introduce an effective unsupervised feature clustering technique to identify the dominating feature set from each cluster. We evaluate the performance of both TreeCLUSS and CLUSSLab using several real world intrusion datasets to identify known as well as unknown attacks and find that results are excellent.

Reusing the best practices and know-how capitalized from existing software process models is a promising solution to model high quality software processes. This paper presents a part of AoSP (Architecture oriented Software Process) for software processes reuse based on software architectures. The solution is proposed after the study of existing works on software process reusing. AoSP approach deals with the engineering ``for'' and ``by'' reusing software processes, it exploits the progress of two research fields that promote reusing in order to improve the software process reusing: domain ontologies and software architectures. AoSP exploits a domain ontology to reuse software process know-how, it allows retrieving, describing and deploring software process architectures. This article details the engineering ``for'' reusing SPs step of AoSP, it explains how the software process architectures are described and discusses the software process ontology conceptualization and software process knowledge acquisition.

In the last decades linear logic became a useful logical system for various usage in computer science. Its ability to handle resources and its competence to describe dynamics of processes predetermine it for describing behavior of programs and program systems. Linear logic can be apprehended as a multiplicative and additive extension of usual logic. We show the possibilities how these fragments can be enriched to describe behavior and to achieve knowledge on an example of simplified Intrusion Detection System (IDS). We construct Kripke model over a coalgebra of modal linear logic for pursuing observable behavior of IDS. Using the same Kripke frame we show how knowledge and belief in the terms of epistemic linear logic can be achieved.

Can specific behaviour strategies be induced from low-level observations of two adversary groups of agents with limited domain knowledge? This paper presents a domain-independent Multi-Agent Strategy Discovering Algorithm (MASDA), which discovers strategic behaviour patterns of a group of agents under the described conditions. The algorithm represents the observed multi-agent activity as a graph, where graph connections correspond to performed actions and graph nodes correspond to environment states at action starts. Based on such data representation, the algorithm applies hierarchical clustering and rule induction to extract and describe strategic behaviour. The discovered strategic behaviour is represented visually as graph paths and symbolically as rules. MASDA was evaluated on RoboCup. Both soccer experts and quantitative evaluation confirmed the relevance of the discovered behaviour patterns.

In this study we discuss how to enable grid sites for the support of data-intensive workflows. Usually, within grid sites, tasks and resources are administrated by local resource managers (LRMs). Many of LRMs have been designed for managing compute-intensive applications. Therefore, data-intensive workflow applications might not perform well on such environments due to the number and size of data transfers between tasks. To improve the performance of such kind of applications it is necessary to redefine the scheduling policies integrated on LRMs. This paper proposes a novel scheme for efficiently supporting data-intensive workflows in LRMs within grid sites. Such scheme is partially implemented in our grid middleware LOGOS and used to improve the performance of a well known LRM: HTCondor. The core of LOGOS is a novel communication-aware scheduling algorithm (PPSA) capable of finding near-optimal solutions. Experiments conducted in this study showed that our approach leads to performance improvements up to 52

The paper presents rough fuzzy subspace clustering algorithm and experimental results of clustering. In this algorithm three approaches for handling missing values are used: marginalisation, imputation and rough sets. The algorithm also assigns weights to attributes in each cluster; this leads to subspace clustering. The parameters of clusters are elaborated in the iterative procedure based on minimising of criterion function. The crucial parameter of the proposed algorithm is the parameter having the influence on the sharpness of elaborated subspace cluster. The lower values of the parameter lead to selection of the most important attribute. The higher values create clusters in the global space, not in subspaces. The paper is accompanied by results of clustering of synthetic and real life data sets.

A variety of important engineering and scientific tasks may be formulated as non-linear, constrained optimization problems. Their solution often demands high computational power. It may be reached by means of appropriate hardware, software or algorithm improvements. The Evolutionary Algorithms (EA) approach to solution of such problems is considered here. The EA are rather slow methods; however, the main advantage of their application is observed in the case of non-convex problems. Particularly high efficiency is demanded in the case of solving large optimization problems. Examples of such problems in engineering include analysis of residual stresses in railroad rails and vehicle wheels, as well as the Physically Based Approximation (PBA) approach to smoothing experimental and/or numerical data. Having in mind such analysis in the future, we focus our current research on the significant EA efficiency increase. Acceleration of the EA is understood here, first of all, as decreasing the total computational time required to solve an optimization problem. Such acceleration may be obtained in various ways. There are at least two gains from the EA acceleration, namely i) saving computational time, and ii) opening a possibility of solving larger optimization problems, than it would be possible with the standard EA. In our recent research we have preliminarily proposed several new speed-up techniques based on simple concepts. In this paper we mainly develop acceleration techniques based on simultaneous solutions averaging well supported by a non-standard application of parallel calculations, and a posteriori solution error analysis. The knowledge about the solution error is used to EA acceleration by means of appropriately modified standard evolutionary operators like selection, crossover, and mutation. Efficiency of the proposed techniques is evaluated using several benchmark tests. These tests indicate significant speed-up of the involved optimization process. Further concepts and improvements are also currently being developed and tested.

RF-Inhomogeneity Correction (aka bias) artifact is an important research field in Magnetic Resonance Imaging (MRI). Bias corrupts MR images altering their illumination even though they are acquired with the most recent scanners. Homomorphic Unsharp Masking (HUM) is a filtering technique aimed at correcting illumination inhomogeneity, but it produces a halo around the edges as a side effect. In this paper a novel correction scheme based on HUM is proposed to correct the artifact mentioned above without introducing the halo. A wide experimentation has been performed on MR images. The method has been tuned and evaluated using the simulated Brainweb image database. In this framework, the approach has been compared successfully against the Guillemaud filter and the SPM2 method. Moreover, the method has been successfully applied on several real MR images of the brain (0.18 T, 1.5 T and 7 T). The description of the overall technique is reported along with the experimental results that show its effectiveness in different anatomical regions and its ability to compensate both underexposed and overexposed areas. Our approach is also effective on non-radiological images, like retinal ones.

Profiting from the development of space remote sensing technology, the amount of remote sensing image data obtained by satellite is increasing dramatically; however, how to deal with these data quickly and efficiently has turned out to be a great computational challenge. With the rapid development of general-purpose GPU computing technology, researchers improved remote sensing applications based on GPU, and obtained good speedup. However, the current GPU parallel processes are not well adapted to the remote sensing image processing; furthermore, they have data loading, storage, and I/O problems. To solve these bottlenecks, this paper proposes three corresponding optimization strategies, and their effectiveness is confirmed by further experiments.

This paper contains research and development results concerning application of hysteretic control principles to solve SIP servers overload problem, which is known from a number of IETF standards and scientific papers published over the past few years. The problem is that SIP protocol, being the application layer protocol, by default has no build-in means of overload control, as, for example, SS7, MTP2 and MTP3 protocols. It was the SS7 network, where a threshold mechanism of hysteretic signaling load control was first implemented. In this paper we describe the main up-to-date solutions of an overload control problem in a signaling network, and develop analytical models of hysteretic control, which are useful in the development of load management functions of SIP servers. We also propose the design of Open SIP signaling Node (OSN) software architecture which is intended to be used for simulations and comparison of various overload control mechanisms.