An Efficient K-Means Clustering Approach in Wireless Network for Data Sharing
Authors: Priyanka Tavse, Anil Khandelwal
Abstract
In a wireless network the data is to be transmitted through network to the base station. The data clustering can be applied for the data transmission for achieving better efficiency, network lifetime and stability. We can improve the data selection mechanism by efficient use of clustering techniques. So in this paper we have applied K-means Clustering approach in Wireless Network for data sharing. In this approach we have used distributed as well as centralized scheme. In one nature the data is in the control of one station which is used to control the data, nodes and key. The distributed environment is used for cluster data sharing which help is improving the performance in terms of time and efficiency.
Speech Parameterization Based on AM-FM Model and its Application in Speaker Identification Using AANN
Authors: D. Giften Francis Samuel, D. Synthiya Vinothini
Abstract
This paper presents the parameterization of speech based on amplitude and frequency modulation (AM-FM) model and its application to speaker identification. Speech parameterization is based on three different bandwidths viz 400Hz, 266mel, 106mel. The feature obtained by this parameterization is termed as PYKFEC which is not directly used as a feature instead its average of each filter is used as the feature and termed as FAP. The speaker identification is done using auto associative neural network and Gaussian mixture model. The AANN/GMM is trained using the SOLO speaking style from CHAINS CORPUS database and a network/model is created for each speaker. The created model is tested using different speaking style like FAST and WHSP of the speaker. The identification rate of FAP is better than PYKFEC, and AANN performs well with these features.
Effect of Nano-clay on Mechanical Properties and Microstructure of Ordinary Portland Cement Mortar
Authors: M. S. Morsy, S. H. Alsayed, M. Aqel
Abstract
The effect of nano-clay on the mechanical properties and microstructure of Portland cement mortar was investigated. The main objective of this research is to constitute a blended cement mortar with high mechanical properties. The nano-clay used in this investigation was nano-kaolin. The nano-metakaolin (NMK) was prepared by thermal activation of kaolin clay for 2 hours at 750 oC. The blended cement used in this investigation consists of ordinary Portland cement (OPC) and nano- metakaolin. The OPC was partially substituted by NMK of 0, 2, 4, 6 and 8% by weight of cement. The blended cement mortar was prepared using cement-sand ratio of 1:2 by weight with water-binder ratio (w/b ratio) as 0.5. The fresh mortar pastes were first cured at 100% relative humidity for 24 hours and then cured in water for 28 days. The compressive strength, tensile strength, phase composition and microstructure of mortar were investigated. The results showed that the compressive strength and the tensile strength of the cement mortars with NMK were higher than plain cement mortar with the same w/b ratio. The enhancement in tensile strength was 49%, whereas the enhancement in compressive was 7% at 8% NMK
Depth Recovery and Motion Estimation Algorithms
Authors: Shivani kumra, Niraj Patil
Abstract
The proposed work is the literature survey done on Depth extraction from a video using many approaches. A detail study is made on the various algorithms used and develop the techniques that automatically generates plausible depth maps from the videos. The techniques which is applicable to both single images as well as videos. The survey also deals with the techniques that can be used to automatically convert the monoscopic video into the stereo for 3D visualization. An algorithm for depth estimation from a monocular video sequence containing moving and deformable objects is also presented
An Efficient Data Mining and Ant Colony Optimization technique (DMACO) for Heart Disease Prediction
Authors: Animesh Dubey, Rajendra Patel, Khyati Choure
Abstract
India is set to witness a spike in deaths due to heart diseases. Early stage detection may prevent the death due to the heart diseases. In this paper we provide an efficient approach which is based on Data Mining and Ant Colony Optimization technique (DMACO) for Heart Disease Prediction so that we can prevent it in the earlier stages. For this we first taken the concept of data mining to finding the support, generated support is used as a weight of the symptom which will be the initial pheromone value of the ant. Then we consider Pain in the chest, Discomfort radiating to the back, choking feeling (heartburn), Nausea, Extreme weakness and Irregular heartbeats as the factor of heart attack. According to risk level identified we find the max pheromone value), max pheromone value is the addition of weight and the risk level. After applying the DMACO algorithm we can observe the increasing detection rate. So by this approach we can increase the detection probability in the early stage which is not generally detected in the earlier stage.