No abstract

One of the known dangerous attacks against wireless sensor networks (WSNs) is node replica. In this attack, adversary captures one or more normal nodes of the network, generates copies of them (replicas) and deploy them in the network. These copied nodes are controlled by the adversary which can establish a shared key with other nodes of the network easily and exchange information. In this paper, a novel algorithm is proposed to defend against this attack in static sensor networks. The proposed algorithm employs a multi-tree architecture to assign ID to the nodes dynamically and prevent attachment of the injected replica nodes to the network by the adversary. The efficiency of the proposed algorithm is evaluated in terms of memory, communication, and computation overheads and the results are compared with other existing algorithms. Comparison results indicate the superiority of the proposed algorithm in terms of mentioned measures. In addition, the proposed algorithm is simulated and its efficiency is evaluated in terms of probability of detecting replica nodes. Experiment results show that the proposed algorithm has favorable performance in detection of replica nodes.

Today, with the emergence of data mining technology and access to useful data, valuable information in different areas can be explored. Data mining uses machine learning algorithms to extract useful relationships and knowledge from a large amount of data and offers an automatic tool for various predictions and classifications. One of the most common applications of data mining in medicine and health-care is to predict different types of breast cancer which has attracted the attention of many scientists. In this paper, a hybrid model employing three algorithms of Naive Bayes Network, RBF Network, and K-means clustering is presented to predict breast cancer type. In the proposed model, the voting approach is used to combine the results obtained from the above three algorithms. Dataset used in this study is called Breast Cancer Wisconsin taken from data sources of UCI. The proposed model is implemented in MATLAB and its efficiency in predicting breast cancer type is evaluated on Breast Cancer Wisconsin dataset. Results show that the proposed hybrid model achieves an accuracy of 99% and mean absolute error of 0.019 which is superior over other models.

Wireless Sensor Network (WSN) is a type of ad hoc networks which consist of hundreds to thousands of sensor nodes. These sensor nodes collaborate to surveillance environment. WSNs have a variety of applications in military, industrial and other fields and they are fit to study environments that presence of human being is costly or dangerous. Sensor nodes have memory, energy and processing limitations. According to sensors' limitations and also increasing use of these networks in military fields, establishing a secure WSN is very important and challenging. Applying Key Predistribution Schemes (KPSs) is one of the effective and useful mechanisms to provide security in WSN. In this paper, a hybrid KPS is proposed that support three various keys, primary pairwise, polynomial, and ordinary. The proposed scheme has been implemented using J-SIM simulator and its performance has been evaluated in terms of maximum supportable network sizes and resiliency against node capture attack, by performing some experiments. Simulation results have been compared with Basic, q-Composite, RS, Cluster-Based, QS, and Double-Key Hash schemes. The compared results showed that the proposed scheme has a better resiliency against links disclosing via enemies.

Aerodynamic is a branch of fluid dynamics that evaluates the behavior of airflow and its interaction with moving objects. The most important application of aerodynamic is in aerospace engineering, designing and construction of flying objects. Reduction of noise emitted by aerodynamic objects is one of the most important challenges in this area and many efforts have been to reduce its negative effects. The prediction of noise emitted from these aerodynamic objects is a low-cost and fast approach that can partially replace the "fabrication and testing" phase. One of the most common and successful tools in prediction procedures is data mining technology. In this paper, the performance of different data mining algorithms such as Random Forest, J48, RBF Network, SVM, MLP, Logistic, and Bagging is evaluated in predicting the amount of noise emitted from aerodynamic objects. The experiments are conducted on a dataset collected by NASA, which is called "Airfoil Self-Noise". The obtained results illustrate that the proposed hybrid model derived from the combination of Random Forest and Bagging algorithms has better performance compared to other methods with an accuracy of 77.6% and mean absolute error of 0.2279.