Title: Transaction Based Clock Synchronization Algorithm for Cloud Based Dynamic Distributed Systems [TCSDDS]
Authors: Suthan Amaresan and D.Kesavaraja
Abstract: Clock Synchronization is a very important part in the functioning of a distributed system. The clock synchronization takes a important role since it can affect the reliability of a application that works on distributed concepts. In this paper we propose a scheme for clock synchronization, in a cloud based environment using clustering of nodes. All the nodes in a distributed application will not be communicating with every other node in the group. So an algorithm is formed to find out the relative nodes which communicate with each other and a local cluster of nodes is formed based upon the relativity in communication and irrespective of geographical locations. Timing signals is exchanged within this cluster of nodes and a cluster moderator synchronizes the timing signals which are again passed to the individual nodes. The local nodes adjust their time based upon the new synchronized time.
Title: Efficient Design of New Novel Full Adder in Gate Diffusion Input Techniques
Authors: M. Rajarajan, Dr. A. Rajaram, A.Saravanakumar, C. Sathiyam and C. Elavarasu
Abstract: We describe the design and performance of the new adder based on the Gate Diffusion Input (GDI) technique is efficient technique for digital design circuits it consumes less power compare with CMOS technique. GDI has an advantage of minimum area, less complexity and propagation delay for any digital circuit design. We propose a new adder using GDI technique its performance is compared with the ripple carry adder and propagate adder it achieves the best power-delay. The new novel adder is simulated using gate diffusion input technique and their performance is higher than the existing GDI adders, with an area requirement comparable with ripple carry adder and carry look ahead adder. Circuit is designed and simulated using CADENCE tool at 180nm technology.
Title: An Efficient Color Based Image Segmentation with Intensity Values Using Conformal Transformations and Otsu Thresholding
Authors: D. Napoleon, V.Mageshwari and U.Lakshmi Priya
Abstract: Images are replications of real world objects. Image can be represented in the area of digital processing, which can be told as another reproduction of an object. Image samples have been considered and it is transformed using conformal transformation. Traditional salt and pepper noise has been added. It has been represented by bright pixels in dark regions and dark pixels in bright regions. Implementation of median filter filters the unwanted disturbances in the Image. In this work color image segmentation has been extensively applied using K-Means Clustering. Color segmentation is done by color features in the image in order to classify different colors in an Image. Thresholding is the basic approach in segmentation of an image. In this work Otsu Thresholding has been used to segment the Image. The images are being segmented according to the intensity value of the pixels.