Cluster Head Sensor Nodes Research Articles

Similarity‐based clustering and data aggregation with independent component analysis in wireless sensor networks

AbstractWireless sensor networks are consisted of group of sensor nodes, which are scattered in inaccessible areas with limited resources. Data aggregation and clustering methods minimize sensor network's energy usage by reducing the quantity of data transmission. Machine learning techniques like reinforcement learning, neural networks, and swarm intelligence significantly cut down the transmission of data and make use of network's distributive features. Quantity of aggregated data can be reduced if the measured data are similar. Based on this concept, a similarity based clustering method is proposed in which sensor node with similar data are grouped as a cluster for data aggregation. Then an algorithm is proposed that aggregates the data using independent component analysis which is performed on cluster head sensor nodes. Data aggregation process is executed on clusters which are having higher data similarity. Implementation shows that proposed method is performing better in terms of computational overhead, aggregation ratio, and energy consumption.

Energy Efficient Clustering Algorithm Based on Particle Swarm Optimization Technique for Wireless Sensor Networks

Maximizing network lifetime in wireless sensor networks is one of the critical issues, particularly for transmitting multimedia data. The wireless sensor network's lifetime is directly linked to energy conservation at each sensor node in the network. Clustering is the most energy-efficient technique for saving energy in sensor networks. The appropriate method for selecting the cluster head is still lagging. The sink node divides the deployment region into the optimal number of sub-regions depending upon its placement in the sensing region. The initial cluster heads are chosen randomly in each region, and this is not an energy-efficient method. The sink node adopts particle swarm optimization technique to select the cluster head in each region efficiently. The chosen cluster head in each region advertises its role to member nodes. Then, the cluster head node is chosen forms the new cluster. PSO optimization technique with the optimization parameters of clustering coefficient, the sensor node's remaining energy, and the distance from the sink and the head of the cluster to the members is adopted to select the cluster head sensor node. The cluster head spends most of its energy aggregating and transferring the data to the sink node. For unloading the cluster head responsibilities, the assistant cluster head and super cluster head are selected for the aggregation and transfer of data, respectively. The proposed energy efficient cluster head selection algorithm has improved the network lifetime by an average of 65 percent better than the existing clustering algorithms.

Clustering Algorithms for Wireless Sensor Networks Survey

This paper aims to discuss a comprehensive survey on clustering algorithms for wireless sensor networks (WSN). The several real-time applications adopted the WSN with the advance features. But the capacity and size of the battery used in the sensor nodes are limited. Battery replacement or recharging is very difficult in most outdoor applications. Hence handling this kind of network is one of the issues. One of the best solutions to the energy issue is Clustering. Clustering is to balance the energy consumption of the whole network by cluster-based architecture to prolong the network lifetime. Sensor nodes grouped into clusters; one sensor node selects as the cluster head for each cluster. The cluster head sensor node collects the data from their sensor member nodes and forwards them to the sink node. In cluster-based architecture, cluster formation and the selection of the cluster head node decides the network lifetime. The paper discusses the for and against various clustering algorithms. It suggests the vital parameters for developing energy-efficient clustering algorithms and steps to overcome the limitations.

An Efficient Proposed Approach of Dynamic Clustering for Target Tracing in Wireless Sensor Networks

This paper proposes and evaluates decentralized dynamic clustering algorithm for tracing a movable target. Here firstly we proposed dynamic K mean clustering algorithm. In this algorithm a fixed number of sensor nodes is choose and then cluster is created. When the cluster is created then a cluster head (CH) is active. This active CH sensor nodes will create new cluster and that new cluster is also formed a new mean value of cluster head. But, the newly created cluster is only active when a moving objected is trace. According from the position of cluster head, few sensor nodes is active, where as few sensor nodes are inactive. According from the CH nodes newly cluster is created. So, creation of dynamic cluster is less energy efficient and stability of cluster will more than static cluster with sensor nodes. On the other hand, movable object tracing sensor nodes are familiar with energy utilization of sensor nodes. Here we proposed an energy efficient target tracing approach which follow network stability as well as energy saving. As we use dynamic clustering technique, so optimization of energy each sensor nodes with cluster head is maximum. So all the sensors with cluster head sensor nodes will continue more time for object tracing. In simulation result we show that our proposed dynamic K mean clustering algorithm is more accurate and more stable.

Implicit authentication protocol and self-healing key management for WBANs

In recent years, the rapid development of computer science and technology brings health care into the information era. As a typical representation, wireless body area networks (WBANs) has also promoted the development of medical technology into a higher field. However, collecting data from users seriously affects personal privacy and security, that WBANs needs to overcome. In order to resolve the flaws of the existing secure authentication protocols in WBANs, we propose a lightweight implicit authentication protocol based on the Elliptic Curve Qu-Vantone (ECQV) algorithm. Moreover, we take advantage of the bidirectional key chain to design a group key management protocol between the personal digital assistance (PDA) and each of cluster head sensor nodes. The security and performance analysis show that our protocol can be performed with perfect forward security and backward security in data communication. In addition, the experimental simulation and theoretic analysis show that our protocol is more suitable for WBANs.

An Efficient Data Compression Model Based on Spatial Clustering and Principal Component Analysis in Wireless Sensor Networks.

Wireless sensor networks (WSNs) have been widely used to monitor the environment, and sensors in WSNs are usually power constrained. Because inner-node communication consumes most of the power, efficient data compression schemes are needed to reduce the data transmission to prolong the lifetime of WSNs. In this paper, we propose an efficient data compression model to aggregate data, which is based on spatial clustering and principal component analysis (PCA). First, sensors with a strong temporal-spatial correlation are grouped into one cluster for further processing with a novel similarity measure metric. Next, sensor data in one cluster are aggregated in the cluster head sensor node, and an efficient adaptive strategy is proposed for the selection of the cluster head to conserve energy. Finally, the proposed model applies principal component analysis with an error bound guarantee to compress the data and retain the definite variance at the same time. Computer simulations show that the proposed model can greatly reduce communication and obtain a lower mean square error than other PCA-based algorithms.

Data Fusion Based on Node Trust Evaluation in Wireless Sensor Networks

Abnormal behavior detection and trust evaluation mode of traditional sensor node have a single function without considering all the factors, and the trust value algorithm is relatively complicated. To avoid these above disadvantages, a trust evaluation model based on the autonomous behavior of sensor node is proposed in this paper. Each sensor node has the monitoring privilege and obligation. Neighboring sensor nodes can monitor each other. Their direct and indirect trust values can be achieved by using a relatively simple calculation method, the synthesis trust value of which could be got according to the composition rule of D-S evidence theory. Firstly, the cluster head assigns different weighted value for the data from each sensor node, then the weight vector is set according to the synthesis trust value, the data fusion processing is executed, and finally the cluster head sensor node transmits the fused result to the base station. Simulation experiment results demonstrate that the trust evaluation model can rapidly, exactly, and effectively recognize malicious sensor node and avoid malicious sensor node becoming cluster head sensor node. The proposed algorithm can greatly increase the safety and accuracy of data fusion, improve communication efficiency, save energy of sensor node, suit different application fields, and deploy environments.

무선센서노드의 센싱 데이터 필터링을 사용한 메모리 관리 기법에 대한 연구

Recently Wireless sensor networks have been used for many purposes and is active for this study. The various methods to reduce energy consumption, which are actively being studied Wireless sensor network to reduce energy consumption, leading to improve transport efficiency, Cluster can be viewed using the research methods. Cluster method researches consists of a sensor node to the cluster and in among those they take out the Cluster head node and Cluster head node is having collects sensing information of circumferential nodes sensing to sink node transmits. Selected as cluster head sensor nodes so a lot of the energy consumption is used as a cluster head sensor nodes is lose a shorter life span have to be replaced by another sensor node. In this paper, to complement the disadvantages of a cluster-mesh method, proposes to manage memory efficiently about filtering method for sensing data. Filtering method to store the same data sensing unlike traditional methods of data filtering system sensing first sent directly by the hashing algorithm to calculate the hash table to store addresses and Sensing to store data on the calculated address in a manner to avoid duplication occurred later, and sensing data is not duplicated by filtering data to be stored in the hash table is a way.