Journal of Modeling and Simulation
Volume:51 Issue: 1, Winter and Spring 2019

The Wireless Sensor Networks (WSNs) include a number of fixed sensor nodes so that each sink moves to collect data between nodes. It is necessary to determine the optimum route and residence location of mobile sinks to reduce energy consumption and increase the value of collected data, which causes increasing the lifetime of WSNs. Using Network Coding (NC), this paper presents a Mixed Integer Linear Programming Model to determine the multi cast Sink Optimal Route (SOR) of Source Sensor Nodes (SSNs) to mobile sinks in WSNs which determines the time and location of sinks to collect maximum coded data and reduce the delay in sinks movement and energy consumption. Since solving this problem is not possible in polynomial time due to the multiple parameters and the limited resources of WSNs, therefore, several heuristic, greedy and fully distributed algorithms are proposed to determine the movement of sinks and their residence location based on maximizing the Value of Collected Coded Data (VCCD) and the type of data deadline. It is demonstrated, by simulation, that the optimal model and the use of NC and proposed algorithms, causes reducing the energy consumption and increasing the Value of Collected Coded Data and network lifetime than non-NC methods.

In this paper, the multi-project scheduling problem is studied. The duration of the activities is subjected to the considerable uncertainty and the robust optimization approach is considered to deal with the uncertainty. The maximum total tardiness of the projects is defined as the objective function which should be minimized. In order to allocate the constrained resources to the multi-projects, two models are proposed. In the first model, the projects are scheduled separately while in the second model, the multi-project approach is applied and the resource sharing policy is used. It is demonstrated that how the tardiness of the projects will be decreased when the multi-project approach is applied. Also, the Adaptive Bee Genetic Algorithm (ABGA) is designed as a hybrid metaheuristic algorithm and proposed in this paper to solve the first stage model of the Robust Resource Constrained Multi-Project Scheduling Problem (RRCMPSP). The results of ABGA is compared with the results of scenario-relaxation algorithm as an exact algorithm for the small size problems. Also, the performance of ABGA is studied compared to the Genetic Algorithm (GA) and Artificial Bee Colony (ABC) as two basic algorithms for the large size problems. The results show the effectiveness of the proposed algorithm in solving the RRCMPSP.

With the development of network-based technologies, intrusion detection plays an important role in modern computer systems. Intrusion Detection System (IDS) is used to achieve higher security, and detect abnormal activities in computers or networks. The efficiency of intrusion detection systems mainly depends on the dimensions of data features. So, in the implementation of the IDS, by applying the feature selection phase irrelevant and redundant features are eliminated, and as a result, the speed and accuracy of the intrusion detection system increases. Applying appropriate search strategy and evaluation measure are significantly effective to feature selection. In this paper, we propose a feature selection method which uses a combination of filter and wrapper feature selection method. This method applies a modified ant colony algorithm as a search strategy on filter phase and fuzzy rough sets to calculate the information gain ratio and acquire the evaluation measure in the ant colony algorithm. Then, on the wrapper phase the minimal subsets of features with first order and second order accuracies are selected. To confirm the efficiency of our proposed method, we compared this method with three other methods and with a method which is based on artificial neural networks. Finally, we compared the proposed method with an ant colony optimization based method. Considering the results, the proposed method, on average, has a higher accuracy than the other methods and also selects a subset of features which have a minimum length.

Emotion recognition has many applications in relation between human and machine. A facial emotion recognition framework for 6 basic emotions of happiness, sadness, disgust, surprise, anger and fear is proposed in this paper. The proposed framework utilizes the histogram estimate of shape and textural characteristics of face image. Instead of direct processing on the original gray levels of face image which may have not significant information about facial expression, the processing is done on transformed images containing informative features. The shape features are extracted by morphological operators by reconstruction and the texture ones are acquired by computing the gray-level co-occurrence matrix (GLCM), and applying Gabor filters. The use of whole face image may provide non-informative and redundant information. So, the proposed emotion recognition method just uses the most important components of face such as eyes, nose and mouth. After textural and shape feature extraction, the histogram function is applied to the shape and texture features containing emotional states of face. The simple and powerful nearest neighbor classifier is used for classification of fused histogram features. The experiments show the good performance of the proposed framework compared to some state-of-the-art facial expression methods such as local linear embedding (LLE), Isomap, Morphmap and local directional pattern (LDP).

For the past decades, we have witnessed an extraordinary advancement in the field of Wireless Sensor Networks (WSNs) in both academic and industrial settings. Moreover, the network lifetime is regarded as one of the most critical issues in this field, and quite a large number of researchers are increasingly exploring this topic of interest. In this study, the linear-scaling method is initially implemented onto the Gravitational Search Algorithm (GSA) for the mass calculation. This is due to the fact that the exploitation and exploration abilities of the algorithm can be fully controlled using this approach. The results obtained from the simulation revealed that this novel GSA achieves the same level of performance as that of conventional GSA and significantly outperforms the state-of-the-art meta-heuristic search algorithms. Additionally, this improved GSA can be readily utilized to solve the K-of-N lifetime problem in two-tiered WSN architecture. In our proposed method, the novel GSA was employed in order to find the optimum location of the base station to enhance network lifetime. Furthermore, the simulation results indicated that despite the simplicity in implementation, our proposed method has a higher level of performance compared to other approaches used to address K-of-N lifetime problem in two-tiered WSN architecture.

Traditional methods for granting credit to loan applicants are based on personal judgment. Nevertheless, the current financial crisis alongside the efforts of banks and financial institutes for decreasing the percentage of overdue loans emphasis the importance of Credit Scoring (CS) models. This paper provides a credit scoring model by means of Modular Neural Network (MNN) established upon combined hybrid-ensemble learning. The proposed model is composed of four powerful neural networks that construct collectively the Adaptive Hierarchical Mixture of Experts (AHME). Training process is a hybrid way for learning the modular model and adaption to the CS model based on the modulation of learning rules specific to each module and particular HME online learning algorithm. Binary Particle Swarm Optimization (BPSO), using Taguchi reasoning scheme for tuning the governing parameters, is also applied for reducing dimensionality and decomposing the problem among the various modules. The proposed model’s performance is compared with that of Multi-Layer Perceptron (MLP) and Laterally Connected Neural Network (LCNN) models. The aforementioned models are evaluated using the data obtained from one of the Iranian banks. Results demonstrate that the AHME outperforms other methods in terms of prediction accuracy as well as the Area Under the ROC Curve (AUC) and the Mean Squared Error (MSE) rate.