Dependency Mapper based Efficient Job Scheduling and Load Balancing in Green Clouds

Cloud computing is usually recognized as a technology which has significant impact on IT. However, cloud computing still has many crucial problems. In a cloud computing system, Load balancing is the most central issue in the system i.e. to distribute the load in an efficient manner. It plays a very important role in the realization of efficient and robust cloud computing platform. In this paper, new load balancing mechanisms have proposed based on character/ nature of jobs along with priority consideration. Furthermore Virtualization is considered in more practical way to avoid the wastage of resources over the network. Finally the performance of the proposed algorithm is analyzed and compared with existing Load balancing and Scheduling policies.

An Analytical Model for IEEE 802. 15. 4/ ZigBee Wireless Sensor Networks with Duty Cycle Mechanism for Performance Prediction and Configuration of MAC Parameters to Achieve QoS and Energy Efficiency

The IEEE 802.15.4 standard defines a Physical (PHY) and Medium Access Control (MAC) layer protocol for low data rate wireless network with low power requirement which makes it ideal PHY/MAC for WSNs. Currently WSNs with combination of IEEE 802.15.4 and ZigBee are being used extensively in industrial applications such as factory automation and control, environmental monitoring etc. Such applications need Quality of Service (QoS) in terms of reliability and latency along with energy efficiency for longevity of the network. The slotted IEEE 802.15.4 MAC protocol uses the duty cycle mechanism to save energy. However the duty cycle mechanism reduces the effective bandwidth leading to possibility of higher packet collision in the active periods and an improper choice of duty cycle may result in low packet delivery ratio and higher packet latency. Similarly there other MAC parameters such as the beacon interval which have bearing on performance of the WSN. It is therefore necessary to make accurate analysis of IEEE 802.15.4 MAC protocol, with duty cycle mechanism enabled, so that the MAC parameters such as Duty Cycle, beacon interval etc. can be set so as to ensure satisfaction of QoS requirements of applications while achieving maximum energy efficiency. We present here an analytical model developed as a Markov Chain Process for slotted IEEE 802.15.4 MAC protocol assuming duty cycle mechanism is enabled in the protocol. Expressions are derived based on the model to compute percentage of packet successfully delivered within a given latency and the energy consumed by the sensor nodes in the process. The model is expected to be useful in deciding on the key MAC parameter values for applications of known traffic load and QoS requirements such as Packet Delivery Ratio(PDR) within given latency while maximizing energy efficiency. The analytical model presented is validated through simulation study in NS2.

Jordanian Cloud-Government between Implementation and Challenges

computing is a modern style of innovation in network environment which takes the advantage of multi-technologies as distributed systems, wireless communication, parallel processing in term of providing service architecture. The main services are concerned about sharing communication and resources. E-governments are available online and they have to adapt new technology that will enhance the service provided to their citizens. This paper focuses on cloud computing and e-government concepts and identifies the ways for implementing cloud computing in Jordanian e-government through a set of stages regarding the main challenges for this adaptation Keywords-government, Cloud environment.

A Fuzzy C-Means based GMM for Classifying Speech and Music Signals

Handoff Study of Ring Shaped Cellular Configuration Designed for High Altitude Platform Communications

Clustering Multi-Attribute Uncertain Data using Probability Distribution

Hybrid Algorithm for Edge Detection using Fuzzy Inference System

Prediction and Filtering of Delay Error on a Corporate Network by using Simulation Model

In this work, a model of a corporate network has been developed, simulated and implemented using optimized network engineering tool (OPNET) technology, in a simulation environment of 100m x 100m office network topology. Delay signal was monitored, neural network (NN) was used to predict the error in the delay signal, one- dimensional (1D) multilevel wavelet de-noising technique to filter the error, autocorrelation function (ACF) and fast Fourier transform (FFT) energy spectrum to validate the result of the filtering after the stages of the decomposition and the reconstruction of the delay signal. The result of the filtering revealed that the error in the data delay is de-noised successfully, since the coefficient of the ACF grows above zero and energy rate in the FFT- spectrum increased.

Rapid 2D-3D Conversion for Low-Cost 3D Television

This work develops a real-time 2D-to-3D converter that exploits motion parallax naturally available in a normal 2D moving image sequence to produce a 3D side-by-side motion picture suitable for viewing on low-cost 3D Television displays at conversion processing rates that can reach high speeds of up to 100 frames per second. The novel paradigm presented in this paper enhances 3D perceptibility by ensuring continuous synchronization between the left and the right image views even if motion in the 2D video abruptly freezes or transitions rapidly, with a minor single frame initial 2D broadcast, equivalent to an initial 2D delay of 1/30 of a second for a 30 frames per second video stream, before full 3D takes effect, and neither depends on computationally expensive depth map extraction nor does it require any special hardware setup such as multicore processors or special purpose graphics processing units.

FPGA Implementation of OFDM Transceiver using Verilog - Hardware Description Language