Current Issue : October - December Volume : 2017 Issue Number : 4 Articles : 6 Articles
Outlier detection is an important data mining task, whose target is to find the abnormal or atypical objects from a given dataset.\nThe techniques for detecting outliers have a lot of applications, such as credit card fraud detection and environment monitoring.\nOur previous work proposed the Cluster-Based (CB) outlier and gave a centralized method using unsupervised extreme learning\nmachines to compute CB outliers. In this paper, we propose a new distributed algorithm for the CB outlier detection (DACB). On\nthe master node, we collect a small number of points from the slave nodes to obtain a threshold. On each slave node, we design a\nnew filtering method that can use the threshold to efficiently speed up the computation. Furthermore, we also propose a ranking\nmethod to optimize the order of cluster scanning. At last, the effectiveness and efficiency of the proposed approaches are verified\nthrough a plenty of simulation experiments....
In wireless sensor networks (WSNs), each sensor node can estimate the global parameter\nfrom the local data in a distributed manner. This paper proposed a robust diffusion estimation\nalgorithm based on a minimum error entropy criterion with a self-adjusting step-size, which\nare referred to as the diffusion MEE-SAS (DMEE-SAS) algorithm. The DMEE-SAS algorithm\nhas a fast speed of convergence and is robust against non-Gaussian noise in the measurements.\nThe detailed performance analysis of the DMEE-SAS algorithm is performed. By combining the\nDMEE-SAS algorithm with the diffusion minimum error entropy (DMEE) algorithm, an Improving\nDMEE-SAS algorithm is proposed for a non-stationary environment where tracking is very important.\nThe Improving DMEE-SAS algorithm can avoid insensitivity of the DMEE-SAS algorithm due to the\nsmall effective step-size near the optimal estimator and obtain a fast convergence speed. Numerical\nsimulations are given to verify the effectiveness and advantages of these proposed algorithms....
The lifetime of wireless sensor networks can be improved by imposing low\nduty cycle, but doing so could not solve unbalanced energy consumption and\nwill increase transmission latency. To avoid this, this paper gives a new method\nto collect data by mobile sink. The proper data collection route is selected\naccording to the sink speed and buffer size of the sensors. The sensors\nonly wake up when the sink approaches them. When certain sensors detect an\nemergency, the sink catches the message quickly and moves to the hotspot to\ndecrease message relay in the network. The result of simulation by OPNET\nshows that this protocol can reduce transmission data in the network and\nprolong the network lifetime....
In this paper, we analyse the deployment of middlebox. For a given network\ninformation and policy requirements, an attempt is made to determine the\noptimal location of middlebox to achieve the best performance. In terms of\nthe end-to-end delay as a performance optimization index, a distributed middlebox\nplacement algorithm based on potential game is proposed. Through\nextensive simulations, it demonstrates that the proposed algorithm achieves\nthe near-optimal solution, and the end-to-end delay decreases significantly....
This paper presents a monitor and a remote control of water level system with automatic operation, this system is considered an application of wireless sensor network. The network of the presented system is consist of two main parts: the main station part which is considered as the gateway node and Raspberry Pi used for this purpose and the second part is the control node which is used Arduino Uno board to control the level of water according to the value of the connected sensor. The ultrasonic sensor is used to detect the water level and a relay board is used to change the state of the water pump. ZigBee technology is used as the wireless communication between the main station and the control node which provided low data rate with low power and low cost. The overall system provided accessing from a remote location by using any web browser and it used a contactless method to detect the water level with increasing the safety and reliability....
This paper focuses on the convergence rate and numerical characteristics of the nonlinear\ninformation consensus filter for object tracking using a distributed sensor network. To avoid the\nJacobian calculation, improve the numerical characteristic and achieve more accurate estimation\nresults for nonlinear distributed estimation, we introduce square-root extensions of derivative-free\ninformation weighted consensus filters (IWCFs), which employ square-root versions of unscented\ntransform, Stirling�s interpolation and cubature rules to linearize nonlinear models, respectively.\nIn addition, to improve the convergence rate, we introduce the square-root dynamic hybrid consensus\nfilters (DHCFs), which use an estimated factor to weight the information contributions and shows a\nfaster convergence rate when the number of consensus iterations is limited. Finally, compared to the\nstate of the art, the simulation shows that the proposed methods can improve the estimation results\nin the scenario of distributed camera networks....
Loading....