The paper presents the criteria that affect the performance of network lifetime in Wireless Sensor Networks. In sensor networks the nodes can be organized into clusters for one hop energy-efficient data routing. This ...
详细信息
The paper presents the criteria that affect the performance of network lifetime in Wireless Sensor Networks. In sensor networks the nodes can be organized into clusters for one hop energy-efficient data routing. This work focuses on adaptive clustering of sensor nodes where the role of cluster head changes among nodes. The nodes take turn to become cluster head based on one or several parameters such as random selection, nodes residual energy, nodes ID, node proximity to other nodes or nodes with different energy levels (advanced nodes). Ideally, the number of elected cluster heads must match the optimal number of cluster heads which is determined a priori or via simulation for energy-efficient routing. Achieving the optimal number of cluster heads is crucial as this will ensure the minimum energy consumption for data transmission to base station. This paper presents simulation results of existing adaptive clustering algorithms. The simulation results show how the election criteria for cluster heads election such as node residual energy, node proximity to other nodes, random election and nodes with different energy level affect the number of cluster heads elected, their distribution in the network and the network lifetime. Simulation results are provided to show the comparative effectiveness of different clustering algorithm on network lifetime, cluster head selection and their distribution in the network. Modifications on the cluster head election criteria are done that combined the node residual energy and nodes proximity to other nodes and simulation shows that the network lifetime is improved.
We propose a novel method for the analysis of the magnetoencephalogram (MEG) of epileptic patients. The proposed method was based on the reconstruction of the phase space from the one-dimension signals to higher-dimen...
详细信息
We propose a novel method for the analysis of the magnetoencephalogram (MEG) of epileptic patients. The proposed method was based on the reconstruction of the phase space from the one-dimension signals to higher-dimension phase space. An especially developed clustering algorithm was applied on the reconstructed data in order to investigate the distribution of epileptic MEG dynamics in higher order reconstructed phase spaces.
In this paper we present CALSAGOS: clustering algorithms Applied to Galaxies in Overdense Systems which is a PYTHON package developed to select cluster members and to search, find, and identify substructures. CALSAGOS...
详细信息
In current scenario wireless sensor networks can be used in several areas like disaster management, intelligence surveillance and border security and so on. Wireless sensor networks (WSNs) are recognized as crucial an...
详细信息
ISBN:
(纸本)9781467394178
In current scenario wireless sensor networks can be used in several areas like disaster management, intelligence surveillance and border security and so on. Wireless sensor networks (WSNs) are recognized as crucial and critical way to facilitate computing environment in effective and efficient manner for numerous applications. The most considerable constraint in wireless sensor network is limited energy. Wireless sensor networks (WSNs) play vital role in computing environment due to its energy saving attributes. In such applications, a collection of sensor nodes are required which are unattended and work independently. clustering is a significant technique which is applied in sensor networks to achieve better utilization of nodes. It is also helpful to get better load distribution in network. In this paper, we survey and discuss various dimensions and approaches of some broadly discovered algorithms for clustering. This paper also presents a comparative study of various clustering algorithms and discussion about the potential research areas and the challenges of clustering in WSNs.
Hierarchical Agglomerative clustering (HAC) algorithms are used in many applications where clusters have a hierarchical relationship between them. Their parallelization is challenging due to the dependence of every ag...
详细信息
ISBN:
(数字)9781728108582
ISBN:
(纸本)9781728108599
Hierarchical Agglomerative clustering (HAC) algorithms are used in many applications where clusters have a hierarchical relationship between them. Their parallelization is challenging due to the dependence of every agglomeration step on all previous agglomerations. Although a few parallel algorithms have been proposed for SLINK HAC algorithm, only limited work has been done to parallelize other HAC algorithms. In this paper, we present a high-level abstraction, which provides a uniform way to specify any HAC algorithm, and a framework for automatic parallelization of the same for distributed memory systems. The abstraction is supported by constructs in a high level, domain specific language, and a compiler translates algorithms expressed in this language to efficient parallel code targeting distributed systems. Our experiments on multiple HAC algorithms proves that the runtime performance achieved is comparable with state-of-the-art manual parallel implementations on Spark and MPI while requiring only a fraction of the programming effort. At runtime, master-slave execution is used, and load is balanced among the slaves in an algorithm-agnostic way, which is a significant contrast to custom load-balancing techniques seen in the literature on parallel HAC algorithms.
clustering algorithms are some of the most important algorithms used to describe data attributes and very effective ways to mine and analyze big data set. In this paper, the whole description of vector space of Intern...
详细信息
ISBN:
(纸本)9781479998937
clustering algorithms are some of the most important algorithms used to describe data attributes and very effective ways to mine and analyze big data set. In this paper, the whole description of vector space of Internet users is acquired by clustering and analyzing user behavior data set. Moreover, all users are divided into different clusters according to KPI which is to correlate different users in terms of their degree of video completion. Our research shows that KPI of total length correlates degree of completion better than other KPIs. This KPI is drastically negatively correlated with user degree of video completion. Finally, we compare the accuracy and efficiency of three different algorithms which we used to cluster our research data in this paper.
Several fast algorithms for clustering very large data sets have been proposed in the literature. CLARA is a combination of a sampling procedure and the classical PAM algorithm, while CLARANS adopts a serial randomize...
详细信息
ISBN:
(纸本)0769504930
Several fast algorithms for clustering very large data sets have been proposed in the literature. CLARA is a combination of a sampling procedure and the classical PAM algorithm, while CLARANS adopts a serial randomized search strategy to find the optimal set of medoids. GAC-R/sup 3/ and GAC-RAR/sub w/ exploit genetic search heuristics for solving clustering problems. In this research, we conducted an empirical comparison of these four clustering algorithms over a wide range of data characteristics. According to the experimental results, CLARANS outperforms its counterparts both in clustering quality and execution time when the number of clusters increases, clusters are more closely related, more asymmetric clusters are present, or more random objects exist in the data set. With a specific number of clusters, CLARA can efficiently achieve satisfactory clustering quality when the data size is larger, whereas GAC-R/sup 3/ and GAC-RAR/sub w/ can achieve satisfactory clustering quality and efficiency when the data size is small, the number of clusters is small, and clusters are more distinct or symmetric.
This paper presents a segment detection and grouping scheme that allows incremental and online learning of indoor environment maps by mobile robots. In this study, the modeling is refined by first dividing the world i...
详细信息
This paper presents a segment detection and grouping scheme that allows incremental and online learning of indoor environment maps by mobile robots. In this study, the modeling is refined by first dividing the world into discrete regions as local models. The line segments in local models are extracted by clustering algorithm. The local models are grouped together by a hierarchical fuzzy system. Adjusting the membership functions that establish the grouping criteria controls the degree of approximation in such combination. The performance of the algorithm is validated in indoor office environments using a Pioneer II mobile robot.
The objective of this paper is the establishment of the relationships among the main variables of a wastewater biological treatment process using, as a first approach, data from a continuous aerobic reactor that was s...
详细信息
The objective of this paper is the establishment of the relationships among the main variables of a wastewater biological treatment process using, as a first approach, data from a continuous aerobic reactor that was simulated by means of the activated sludge model No 1 (ASM no 1). Self-organizing map (SOM) and clustering techniques were used to carry out this task.
Through Wireless Sensor Networks (WSN) formation, industrial and academic communities have seen remarkable development in recent decades. One of the most common techniques to derive the best out of wireless sensor net...
详细信息
暂无评论