International Journal of Applied Mathematics and Informatics

ISSN: 2074-1278
Volume 11, 2017


Notice: As of 2014 and for the forthcoming years, the publication frequency/periodicity of NAUN Journals is adapted to the 'continuously updated' model. What this means is that instead of being separated into issues, new papers will be added on a continuous basis, allowing a more regular flow and shorter publication times. The papers will appear in reverse order, therefore the most recent one will be on top.

Main Page

Submit a paper | Submission terms | Paper format

 


Volume 11, 2017


Title of the Paper: Exact Computation of Residual Noise and Collateral Distortion in Nonlinear Image Filtering: A Case Study

 

Authors: Fabrizio Russo

Pages: 126-134

Abstract: Residual noise and collateral distortion are two key features for any image denoising filter. The former is the amount of noise still affecting the data after filtering, the latter represents the price to be paid in terms of detail blur. Measuring these effect is of paramount importance for the validation of a denoising algorithm. This work focuses of Non-Local Means (NLM) filtering that represents one of the most effective approaches to grayscale image denoising. The exact values of residual noise and collateral distortion are derived from NLM theory and an in-depth analysis of these features is provided for different input data and different parameter settings.


Title of the Paper: Integrated LTspice and NS-3 Power Management Simulation for Energy Harvesting

 

Authors: Moner Alsader, Al Savvaris

Pages: 120-125

Abstract: LTspice is a popular and free circuit simulating package while NS-3 is an open source network simulator. Integrating these two popular simulation packages is an ideal scenario where an output from LTspice such as power can be an input to network modules in NS-3. The aim of this paper is, therefore, to integrate LTspice and NS-3 simulators in order to forge a strong research collaboration between two domains (electronics and networking). The integration of these simulators have be evaluated and validated by using power management use case in energy harvesting. Preliminary results based on Fuzzy control logics for the power management system have shown that the integration to be effective. The downtime ratio and energy efficiency utilization were preferable compared to the scenario where fuzzy control system was not implemented.


Title of the Paper: Exact Computations as Background for Numerical Methods

 

Authors: Valentin Golodov

Pages: 116-119

Abstract: This paper continues a couple of author’s papers devoted to development of the exact rational computing and its applications in numerical methods of analysis and solving illconditioned tasks. Examples of application of the exact rational computing in numerical methods of solving ill-conditioned tasks are presented. Ill-conditioned linear equation systems, linear equation systems with interval uncertainty of the coefficients are examples of such tasks. Application of the interval regularization procedure to the Firordt method of the spectrophotometric analysis of the non-separated mixtures is one examples where exact computing allows improving of the result robustness.


Title of the Paper: Interval Entropy Profile Method Based on Maximum Entropy Principle for Estimating Evolutionary Information Content in Genetic Sequences for their Comparison

 

Authors: Uddalak Mitra

Pages: 108-115

Abstract: Two classes of computational algorithms have been applied to solve the most fundamental problem in Bioinformaticssequence comparison. Due to computational limitation in alignment methods, alignment-free methods become a predominant approach for devising sequence similarity. Of the alignment-free methods, return time distribution(RTD) based approaches are widely applied to answer various biological queries. However, existence of sequence noise becomes a principle bottle-neck to achieve precision in results for RTD approaches. Additionally, use of single length for k-mer strings often neglects evolutionary information that can be extracted using k-mers of other lengths. I inspect these flaws and propose a new method that considers multiple lengths of k-mer strings for generating features. Markovian dependency is assumed to estimate expected distribution for k-mer strings and to reduce sequence noise. Additionally, a criteria based on uniform independent model of sequence composition for selecting maximum k value is used. Experiments on benchmark datasets of 18s-rRNA sequence and whole genome mitochondrial(mtDNA) sequence of eutherians orders show superiority of the proposed method over all existing RTD and two start-of-the-art non-RTD methods.


Title of the Paper: Banking Information System: Residual Risks Modeling

 

Authors: Marie Ndaw, Gervais Mendy, Samuel Ouya

Pages: 100-107

Abstract: Banking Products and services are more and more innovative using Internet and mobile technologies. For this reason, bank information system must be protected against several risks considering physical or logical attacks which can cause important financial losses. In this paper, we propose a quantification model of information system residual risks which are the risks remaining after the response of management. Our model provides an automatic calculation of security measures impact on information system risks and it is based on FMECA (Failure Modes and Effect Criticality Analysis Method) which is an inductive reasoning method studying causes, effects of failures and criticality. For testing, the values obtained with the model were compared to reference values given by assessors during different working sessions. The result is satisfactory, facilitates risk analysis, helps to maintain information system security, to increase banks profit and customers confidence.


Title of the Paper: Attraction in Differential Systems Arising in Network Regulatory Theory

 

Authors: Eduard Brokan, Felix Sadyrbaev

Pages: 96-99

Abstract: We consider systems of ordinary differential equations that appear in gene regulation networks. Interrelation between nodes of a network is described by the regulatory matrix W: We provide description of attracting sets for various choices of regulatory matrix W or, equivalently, for various types of interrelation in a network. The related examples are considered.


Title of the Paper: Optimizing a Single Vendor-Multi Buyer Bi-Objective Supply Chain Problem with Stochastic Demand and Routing Under Storage Capacity and Transportation Constraints Using NSGA-II Algorithm

 

Authors: Mohammadreza Shahriari

Pages: 87-95

Abstract: One of the most important factors for succession in a supply chain is decreasing their costs. Combining the decision making in inventory control and routing areas in distribution network may decrease the cost of supply chain and increase level of service. In this paper we work on a bi-objective supply chain consisting single vendor-multi buyer in infinite horizon. The vendor produces all products and presents them to the buyers through a heterogeneous fleet of transportation. The first objective of this problem deals with decreasing the cost of inventory and transportation and the second one with increasing customer satisfaction level. To draw the problem close to the real situations some practical constraints like storage capacity and transportation equipment are added to the problem. This model is a non-linear integer problem, so Non-dominated sorting genetic algorithm (NSGA-II) is used for solving the presented model.


Title of the Paper: Implementation of an Automatic Appliance Control and Energy Monitoring System

 

Authors: Rula Sharqi, Muhammad Hassaan Ayyub

Pages: 82-86

Abstract: The concept of home automation has recently sparked consumer interest due to rapid innovation, development of new technologies, growing trend towards sustainability and efficiency, to serve this issue, a ‘stepping-stone’ needs to be developed, not necessarily providing a comprehensive automation solution, but instead aiming to ease the transition to such a solution by minimizing the investment required by the consumer, along with installation and maintenance overheads. In this paper, such a product is developed by improving a low-cost device that already exists in most homes and has the potential to be useful for automation purposes – the power strip. The developed product is a power strip that allows users to remotely control and automate appliances, and track the power consumption of each appliance using a web interface connected to the Internet. The developed system serves as a simple and low-cost entry into the world of automation and will aid in presenting this concept to consumers who are not ready to invest the substantial amount required by existing solutions.


Title of the Paper: Stabilized Column Generation for the Construction of Rotations with Resource Constraints

 

Authors: Andelakder Lamamri, Imene Mehamdia

Pages: 77-81

Abstract: Column generation has proven to be efficient in solving the linear programming relaxation of large scale instances of the construction of Rotations with resource constraints. However difficulties arise when the instances are highly degenerate. Recent research has been devoted to accelerate column generation while remaining within the linear programming framework. This paper presents an efficient approach to solve the linear relaxation of the construction of Rotations with resource constraints. It combines column generation, preprocessing variable fixing, and stabilization. The outcome shows the great potential of such an approach for degenerate instances.


Title of the Paper: Existence and Uniqueness of the Weak Solution of a (Time-Dependent) Advection-Diffusion PDE

 

Authors: Atefeh Hasan-Zadeh, Neda Asasian, Mohammad Ali Aroon

Pages: 71-76

Abstract: This paper deals with time-dependent advection-diffusion equation which covers gas absorption, solid dissolution, heat and mass transfer in falling film or pipe and other similar equations of transport phenomena. Among various works about the solution of these PDEs by numerically and somewhat analytically methods, a general analytic framework for the equations is proposed. In fact, drawing upon advanced ingredients of Sobolev spaces, weak solutions and some important integral inequalities, an analytic methodology is presented for the existence and uniqueness of the weak solution of these PDEs which is the best solution in the proposed structure. Then, the weak solution of the general parabolic boundary value problem, covering transport phenomena PDEs, can be obtained by a reduced system of ODE. Furthermore, the new approach supports infinite propagation speed of disturbances of (time–dependent) advection–diffusion equations in semi–infinite media.


Title of the Paper: Presenting a Mathematical Model for Joint Production and Purchasing in a Multi-Product Problem and Determining the Optimal Order Value under Practical Constraints

 

Authors: Mani Sharifi, Ali Haji Kazem Lavasani, Mohammadreza Shahriari

Pages: 65-70

Abstract: Production and economic order is one of the most important topics in problems of production and inventory control, and it has long been of interest to researchers. Economic and profitable production requires that a comprehensive and detailed plan is taken and implemented for all stages of production. In the manufacturing companies that are producing a group of products, it is possible that in some periods, customers’ demands exceed the production rate. In such conditions, the companies can response to such demand by more production, purchase, or accepting shortage of backordering. In case of purchase, the products stock in the warehouse at the beginning of programing period. In this study, a model is presented to determine the optimal amount of production and purchase in a multi-product system and with the objective of optimizing the costs of inventory system under the limitations of warehouse space and budget. Also, the start production time and the optimal amount of the shortage are determined. The presented mathematical model is of nonlinear integer type. The metaheuristic algorithm was used to solve the models, and the method of design of the experiments has been used to configure the parameters of the proposed method. At the end, the proper performance of the proposed methods will be proved by some numerical examples and comparing with random search method.


Title of the Paper: A Formal Framework for Data Fusion

 

Authors: Ildar Baymuratov, Nataly Zhukova

Pages: 56-64

Abstract: In order to develop a formal framework for data fusion field main data fusion models are observed and a logical model of data fusion is suggested. It is shown that considered data fusion models are representable in the logical model. The logical model being insufficiently general to compare different methods of data fusion is reformulated on category theory language. After that a number of information theoretic measures defined on morphisms are suggested as universal criteria for evaluation of data fusion methods.


Title of the Paper: Two Term Control Strategy for Position Control of Twin Rotor System

 

Authors: Sonal Sing, Shubhi Purwar

Pages: 50-55

Abstract: In this paper, a two term control law is developed for position control of Twin rotor multiple-input-multiple-output (MIMO) system (TRMS). The proposed controller is an added delay term in conventional Composite Nonlinear Feedback (CNF) control which improves robustness of the controller with fast transient response and better damping characteristics. Proposed control law is compared with conventional CNF to prove its superiority which is validated via computer simulation in MATLAB environment. Lyapunov-Krasovskii functional is proposed for the study of robust stability conditions which create boundaries of the closed-loop system.


Title of the Paper: Worst-Case Test Network Optimization for Community Detection Methods

 

Authors: Iveta Dirgová Luptáková, Marek Šimon, Jiří Pospíchal

Pages: 44-49

Abstract: In control and automation, applications of network community detection range from black-start recovery of power sys-tems, truss structure manufacturing, neural motor control, water dis-tribution, up to image segmentation for finding cracks in bridges. For such important technical applications, it is necessary to know the limits of various methods, how much they can deviate from optimum. Network community detection finds communities/clusters of densely connected nodes with few edges outside the cluster. The robustness of the community detection methods were already compared on a number of test networks, both real world and artificial. However, these test networks were implicitly average cases. Here we evolve networks, which produce a maximum error of modularity measure for selected methods of community detection. This worst-case test networks were evolved for Edge betweenness, Fast greedy, Infomap, Louvain, and Walktrap modularity detection. Such a comparison provides a tougher test of robustness than previous approaches. A bit surprisingly, after the Blondel’s Louvain method, the next best result was provided by fast greedy, while otherwise favored Infomap fared rather poorly.


Title of the Paper: Extending Population Oriented Extremal Optimisation to Permutation Problems

 

Authors: Marcus Randall

Pages: 38-43

Abstract: Extremal optimisation in its canonical form is based on the manipulation of a single solution. This solution is changed iteratively by gradually replacing poor components of it so that over time it improves. Many successful evolutionary optimisers are population based, so it appears a reasonable exercise to extend extremal optimisation in this way. Scaling it up to an entire population presents many challenges, and only a few works have examined possible models. In this paper, a recent approach is expanded upon which extends the approach from assignment type problems (such as the generalised assignment problem) to permutation oriented ones. Using the asymmetric travelling salesman problem as a test case, it is found that improvements over a canonical extremal optimisation algorithm were realised.


Title of the Paper: An Empirical Study on the Current Practice in Software Project Management – the Requirement Specifications

 

Authors: Yusmadi Yah Jusoh, Nor Hidayah Zainal Abidin, Che Suhana Che Wil, Fatimah Abdullah Moalim, Nurul Huda Wahab, Noraini Che Pa

Pages: 27-37

Abstract: Requirements specification is one of the most crucial processes in software development projects. Without well-specified requirements, the project manager could not planning and design a good project and team members could not understand what and how to do and things get worse when the user may not know what can be expected from the project. The objective of the study is to identify factors that influence the specification requirements associated with the successful of project development from the literature review and to verify the best factors identified through the real-world practice. Identifying such factors can help companies around the world in an effort to improve the quality of software development, especially in the specification requirements that lead to the success of project management. To deal with research question, we use the literature review and empirical survey approach. The result shows the factors that most significantly affect the requirements specifications is effective requirements communication chain and the lowest factor is the defined project/application domain. This study expected to address some of the factors that affect the specification requirements that is identified can help practitioners to develop a good strategy in the planning for the project success.


Title of the Paper: Computation of Stabilizing PI and PID Parameters for Multivariable System with Time Delays

 

Authors: Nour El Houda Mansour, Sami Hafsi, Kaouther Laabidi

Pages: 20-26

Abstract: In this paper, a new approach to multivariable Smith control design is proposed for the computation of all stabilizing by Proportional-Integra (PI) and Proportional-Integral-derivative PID controllers for Multi-Inputs Multi-Outputs (MIMO) plants with multiple time delays. First, the Smith decoupler is used to eliminate interactions. Second the decoupled models are approximated by second-order plus time delay (SOPDT) models using the standard recursive least square approach. We seek finally, by using a generalization of Hermite-Biehler theorem, the set of complete stabilizing PI and PID parameters.


Title of the Paper: Controlling the Angle of Attack of an Aircraft Using Genetic Algorithm Based Flight Controller

 

Authors: S. Swain, P. S Khuntia

Pages: 11-19

Abstract: In this paper, the unstable angle of attack of a FOXTROT aircraft is controlled by using Genetic Algorithm based flight controller and the result is compared with the conventional techniques like Tyreus-Luyben (TL), Ziegler-Nichols (ZN) and Interpolation Rule (IR) for tuning the PID controller. Also the performance indices like Mean Square Error (MSE), Integral Square Error (ISE), and Integral Absolute Time Error (IATE) etc are improved by using Genetic Algorithm. It was established that the error by using GA is very less as compared to the conventional techniques thereby improving the performance indices of the dynamic system.


Title of the Paper: Comparative Study of Krill Herd Algorithm and Flower Pollination Algorithm

 

Authors: Viktor Tuba, Romana Capor-Hrosik, Milan Tuba

Pages: 6-10

Abstract: Numerous applications deal with the hard optimization problems. In recent years various solutions for this kind of problems were proposed. Swarm intelligence algorithms represent efficient metaheuristics for finding the optimal solution for hard optimization algorithms. In this paper two recent swarm intelligence algorithms, krill herd algorithm and flower pollination algorithm will be tested and compared. Both algorithms were tested on CEC 2013 benchmark functions. Flower pollination algorithm obtained better and more stable results for some test functions while krill herd algorithm made some larger mistakes for some test functions.


Title of the Paper: Assessing the Results of Exposure to Computers on Problem Solving Skills

 

Authors: Michael Gr. Voskoglou

Pages: 1-5

Abstract: Problem solving affects our daily lives in direct or indirect ways for ages. Nowadays, in our knowledge society the attitude to think critically has become a necessary condition for solving non-routine problems. However, Critical Thinking is not always a sufficient condition too for problem solving, especially when tackling complicated technological problems, where computers are frequently used as a supporting tool. In this case the need for Computational Thinking is another prerequisite for problem solving. The present paper utilizes the traditional calculation of the mean values and of the Grade Point Average (GPA) index for assessing the effect of the student exposure to computers when taught mathematics on their mathematical problem solving skills. A classroom experiment is performed in which the outcomes of the two assessment methods are compared to each other and interesting conclusions are drawn.