Abstract. Minimization of the number of cluster heads in a wireless sensor network is a very important problem to reduce channel contention and to improve the efficiency of the algorithm when executed at the level of cluster-heads. This paper proposes a Self Organizing Sensor (SOS) network based on an intelligent clustering algorithm which does not require many user defined parameters and random selection to form clusters like in Algorithm for Cluster Establishment (ACE) [2]. The proposed SOS algorithm is compared with ACE and the empirical results clearly illustrate that the SOS algorithm can reduce the number of cluster heads.
Abstract:Wireless sensor networks are composed of a huge number of sensor nodes, which have limited resources -energy, memory and computation power. Energies are directly related to the lifetime of sensor network. If sensor nodes can be grouped to clusters, cluster member sensor nodes only need to communicate with cluster center (head) and this leads to energy conservation of the member sensors. So, how to compose clusters with minimal number of cluster heads, while including each node in a cluster is an important research issue. We propose a new advanced optimization algorithm for sensor network clustering. Using the proposed optimization algorithm, redundant cluster heads are eliminated, and unnecessarily overlapped clusters are merged. Optimization algorithm can be used as a clustering algorithm by itself and also manage the dynamic changes like node addition or die-out, while the network is even on the working state. We tested the proposed method as a clustering algorithm and compared it with two other recent sensor network clustering algorithms, Algorithm for Cluster Establishment (ACE) and Self Organizing Sensor network algorithm (SOS). The experiments results not only illustrate that the proposed algorithm could result in clusters with smaller number of cluster heads than others with any density of sensor networks, but also that the performance is more stable, which is also verified through repeated experiments.
Abstract. We show that excluding outliers from the training data significantly improves kNN classifier, which in this case performs about 10% better than the best know method-Centroid-based classifier. Outliers are the elements whose similarity to the centroid of the corresponding category is below a threshold.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.