Chandigarh University

About: Chandigarh University is a education organization based out in Mohali, India. It is known for research contribution in the topics: Computer science & Chemistry. The organization has 1358 authors who have published 2104 publications receiving 10050 citations.

An efficient approach for green cloud computing using genetic algorithm

Abstract: Cloud computing delivers consumers a proficient way to efficiently complete their service demands. The use of energy aware virtual machine allocation algorithms for cloud computing framework provides great elasticity with the capability to migrate virtual machines across physical machines. However, data center hosting services consume huge amount of electricity. The ongoing growth of energy consumption by Information and Communication Technology (ICT) at data center side forces researchers to propose innovative ways to reduce the power consumption. A major challenge for cloud providers is proper VM allocations while taking into account of less SLA violations within better response time. In this paper a energy aware hybrid algorithm for VM allocation is proposed with considering SLA parameters; response time, throughput. As number of algorithm proposed for energy aware VM allocations like Minimization of Migration (MM). In this paper, a hybrid energy aware algorithm using genetic algorithm has been proposed. We evaluate the proposed algorithm with three performance metrics; energy consumption, SLA violations, throughput. The results demonstrate the proposed algorithm obtains less energy consumption and SLA violations with high throughput than the MM algorithm.

A high performance multiband BAW filter

Abstract: In this paper, a multi band filter using TFBAR is proposed for 4G/LTE IEEE standard at mobile front end terminal. Initial layout of FBAR is designed and analyzed in Coventorware software. The electromagnetic properties and equivalent circuit characteristics of proposed multi-band filter are studied using Advanced Design System, Agilent software. The filter works well for frequency range 2.3–2.4 and 1.97–2.24 GHz. In this research work, the effect of inserting passive elements i.e. inductor (L) and capacitor (C) either alone or both in series and shunt arms of equivalent circuit of filter are also studied. As a result of introducing LC components, the filter responses at different frequency range and converges the frequency response either narrow or wideband of the proposed filter. The measured insertion loss is better than − 2 dB and return loss is better than − 20 dB. The novelty of this work is designing of miniaturized multiband filter which can be deployed in future mobile RF front end terminal that not only cover upcoming frequency band but also cover present standard IEEE communication band.

A Novel Cost-Effective Magnetic Characterization Tool for Soft Magnetic Materials Used in Electrical Machines

Abstract: Energy losses during the magnetization process of soft magnetic laminations depend on the excitation waveform. Manufacturers of magnetic materials may not provide losses for different excitation waveforms. This may affect the accuracy of the predicted losses which are used to estimate the performance of electrical equipment during the predesign stage. Thus, a characterization tool which is capable of measuring hysteresis loops under controlled magnetization is required. Digital characterization systems available in literature use expensive commercial power amplifiers. In this work, a cost-effective power amplifier circuit is developed and interfaced with a LabVIEW program to characterize soft magnetic materials. The measurements for a sinusoidal excitation from the proposed setup are validated with those from a standard commercial measuring setup. The system is also used to measure hysteresis loops for arbitrary magnetization conditions (magnetic flux density ( $ {B}$ ) with harmonics). The proposed setup can operate over a frequency range 35–500 Hz. A small Epstein frame is also designed for measurements at higher frequencies (>200 and <500 Hz). A provision is also made to predict three components of dynamic hysteresis losses at any frequency in the above range using the loss separation approach. This capability of the setup enables the user to predict the losses at frequencies beyond the capability of the tool. The predicted losses using this algorithm are validated with measured losses. The proposed setup is cost-effective and can be developed easily using basic electronic components.

Lossless compression of DICOM images using genetic algorithm

Abstract: The numerous medical images generated by various imaging devices require a huge amount of storage space and minimum transmission bandwidth. So, it is necessary to use an appropriate technique for compression of medical images in order to meet the above concerns. This paper proposes a technique for lossless compression of significant parts by extracting the region of interest in DICOM images. The extracted region is compressed using Huffman encoding and genetic algorithm for the enhancement of compression ratio. Additionally, the value of PSNR has also been calculated. When the results of the proposed technique are compared with the previous method, it is found that there is an increase of more than 400% in the compression ratio. It also provides an average increase of 20% in the compression ratio by using Huffman encoding.