Abstract Connectivity and coverage are two crucial problems for wireless sensor networks. Several studies have focused on proposing solutions for improving and adjusting the initial deployment of a wireless sensor network to meet these two criteria. In our work, we propose a new hierarchical architecture for sensor networks that facilitates the gathering of redundancy information of the topology. Several mobile robots... Connectivity and coverage are two crucial problems for wireless sensor networks. Several studies have focused on proposing solutions for improving and adjusting the initial deployment of a wireless sensor network to meet these two criteria. In our work, we propose a new hierarchical architecture for sensor networks that facilitates the gathering of redundancy information of the topology. Several mobile robots must then relocate, in an optimized way, redundant sensors to achieve optimal connectivity and coverage of the network. Mobile robots have to cooperate and coordinate their movement. A performance evaluation is conducted to study the trade-off between the number of required robots and its impact on the rate of network connectivity and coverage. Read More Read Less

Abstract Clustering peers based on the semantics of their content is one of the most difficult and important taskin P2P data integration systems because it enhances data search and integration significantly. Currently super-peer networks, such as the Edutella network, do not provide sophisticated means for such a “semantic clustering” of peers.In fact, most solutions try only to combine the advantages of... Clustering peers based on the semantics of their content is one of the most difficult and important taskin P2P data integration systems because it enhances data search and integration significantly. Currently super-peer networks, such as the Edutella network, do not provide sophisticated means for such a “semantic clustering” of peers.In fact, most solutions try only to combine the advantages of data integration and P2P technologies to overcome centralized solutions shortcomings without taking into account content based semantic and efficiency to deal with network dynamicity. In this paper, we investigate P2P computing and data integration fundamentals and detail the challenges that face the P2P data integration process. In addition, we presentour approach for semantic clustering in a super peer architecture based on a vector space model forour P2P semantic data integration framework. In a first stage, we detail the various modules of our framework and specify the functions of each one. Then, we detailed the different levels of knowledge we take into consideration for the semantic clustering that we adopted using ontology alignment techniques. Finally, we explain how we manage network dynamicity and how semantic should be adjusted accordingly. Read More Read Less

Abstract In recent years, Wireless Ad-hoc networks have been considered as one of the most important technologies. The application domains of Wireless Ad-hoc Networks gain more and more importance in many areas. One of them is the control and management of the traffic of packets. In this paper our goal is to control the performance of different sections of the pipeline... In recent years, Wireless Ad-hoc networks have been considered as one of the most important technologies. The application domains of Wireless Ad-hoc Networks gain more and more importance in many areas. One of them is the control and management of the traffic of packets. In this paper our goal is to control the performance of different sections of the pipeline of a factory by checking the network, periodically. Along the factory the traffic is modeled with a Poisson process. With this model we can produce packets. The numbers of packets which are sending from a typical node to the other nodes are controlled by a fuzzy system. We show that having the traffic packets at time (t) for each node in Wireless Ad-hoc Network, we can completely train a Neural Network and successfully predict the traffic at time (t+1) for each node. By this means we can recognize the inefficient sections of the factory and try to fix them. The results of different experiments have shown that the proposed model has acceptable performance. Read More Read Less

Abstract Multi-Layered Perceptron (MLP) is a useful supervised neural network for data classification. Error Back Propagation (EBP) algorithm is the common technique for training MLP. Standard EBP algorithm has challenges for largescale and heterogeneous data such as lack of memory and low–speed convergence, besides, computational load is high. In this paper, to overcome these drawbacks, a modified version of EBP has... Multi-Layered Perceptron (MLP) is a useful supervised neural network for data classification. Error Back Propagation (EBP) algorithm is the common technique for training MLP. Standard EBP algorithm has challenges for largescale and heterogeneous data such as lack of memory and low–speed convergence, besides, computational load is high. In this paper, to overcome these drawbacks, a modified version of EBP has been proposed. It decreases the time and space complexity, and somewhat increases convergence speed of the standard EPB by partial and random updating of some of weights instead of all of them. Result of experiments on two standard dataset, confirms the effectiveness of the proposed algorithm. Read More Read Less

Abstract In this paper, a new classification method that uses a clustering method to reduce the train set of K-Nearest Neighbor (KNN) classifier and also in order to enhance its performance is proposed. The proposed method is called Nearest Cluster Classifier (NCC). Inspiring the traditional K-NN algorithm, the main idea is to classify a test sample according to the tag of... In this paper, a new classification method that uses a clustering method to reduce the train set of K-Nearest Neighbor (KNN) classifier and also in order to enhance its performance is proposed. The proposed method is called Nearest Cluster Classifier (NCC). Inspiring the traditional K-NN algorithm, the main idea is to classify a test sample according to the tag of its nearest neighbor. First, the train set is clustered into a number of partitions. By obtaining a number of partitions employing several runnings of a simple clustering algorithm, NCC algorithm extracts a large number of clusters out of the partitions. Then, the label of each cluster center produced in the previous step is determined employing the majority vote mechanism between the class labels of the patterns in the cluster. The NCC algorithm iteratively adds a cluster to a pool of the selected clusters that are considered as the train set of the final 1-NN classifier as long as the 1-NN classifier performance over a set of patterns included the train set and the validation set improves. The selected set of the most accurate clusters are considered as the train set of final 1-NN classifier. After that, the class label of a new test sample is determined according to the class label of the nearest cluster center. Computationally, the NCC is about K times faster than KNN. The proposed method is evaluated on some real datasets from UCI repository. Empirical studies show an excellent improvement in terms of both accuracy and time complexity in comparison with KNN classifier. Read More Read Less