A. Elchouemi

Enhancing static noise margin while reducing power consumption

The unrelenting scaling of CMOS devices has brought their dimensions down to few tens of nanometers. In such sizes, the reliability margins drop ominously and the leakage power dissipation increases significantly. This paper presents a non-conventional transistor-sizing method for improving reliability by increasing the static noise margin, while simultaneously reducing the power consumption. Simulations results have been used to compare the static noise margin, the power consumption, and the performance of classical CMOS gates with the proposed scheme in the 22 nm technology. The results show that modifying the channel lengths of transistors in inverters and other gates can improve the noise margin by nearly 40% over the conventional one, while reducing the power consumption by 47%. The robustness (measured here in terms of noise margin) of the classical and the new gates are also compared when their transistors are subject to threshold voltage variations.

SIFT based approach on Bangla sign language recognition

This paper presents a SIFT-based geometrically computational approach to vigorously recognize Bangla sign language (BdSL). Gaussian distribution and grayscaling techniques are applied for image processing and normalizing the sign image. After this pre-processing, features are extracted from the sign image by implementing scale invariant feature transform. Acquiring all descriptors from the sign image, k-means clustering is executed on all the descriptors which are previously computed by SIFT. Based on the sample training set, each of the cluster denotes as a visual word. Considering the histograms of the clustering descriptors, Bag of words model is introduced on this hybrid approach which develops a set of visual vocabulary. Finally for each of sign word, a binary linear support vector machine (SVM) classifier is trained with a respective training data set. Considering these binary classifiers, we obtained a respective recognition rate on both Bangla signs of expressions and alphabets.

Early diagnosis of breast cancer using contrast limited adaptive histogram equalization (CLAHE) and Morphology methods

The main aim of this paper is to enhance using hybrid solution for early diagnosis of breast cancer using mammogram images. For the CAD system for any image processing system follows mainly four steps that is image pre processing, segmentation, features extraction and classification and evaluation. For this research purpose follows the same structure using best possible methods in each stage found in literature review and proposed solution. From the experiment hybrid image enhancement method using CLAHE and Morphology method helps to enhance image for further computation in CAD system to early diagnosis of breast cancer using mammogram images while maintaining processing time as previous best solution. From the study depict that the purposed solution can help to enhance mammogram images for the further processing for segmentation, feature extraction and classification. However, further testing and implementation of proper classification method needed for before clinical use.

A collaborative platform for facilitating standard cell characterization

Circuit designers need to characterize their designs based on different requirements of area, power, delay, and transition times. To do so, they may need to create a large number of circuits (defined as Spice netlists) in different configurations. The netlists not only contain the unit-under-test but also the mechanisms for providing the input stimuli and measuring variables (such as voltage, current, and time). The characterization is usually carried out by a team of engineers who may be located on the same site or in different geographical locations. A web-based collaborative platform proposed in this paper facilitates and expedites the characterization process by producing the required netlists in a consistent and error-free fashion.

Experiments on the MFCC application in speaker recognition using Matlab

Speaker recognition is a very important research area where speech synthesis, and speech noise reduction are some of the major research areas. Speaker recognition is a new challenge for technologies. Many algorithms have been suggested and developed for feature extraction. This paper presents a feature extraction technique for speaker recognition using Mel Frequency Cepstral Coefficients (MFCC). Further, this paper evaluates experiments conducted along each step of the MFCC process. Finally, the paper compares hamming window and rectangular window technique based on the number of filters for accurate and efficient result in a Matlab environment. The result indicates that using a 32 filter with hamming window has more accuracy and efficiency compared to other windowing techniques and number of filters.

Early diagnosis of Alzheimer using mini mental state examination method: MMSE

This study aims to analyse the current method in diagnosing early Alzheimer disease and offer a new method to improve the performance of bioinformatics techniques. It proposes a hybrid MRI image processing method to improve the image quality for Alzheimer disease classification. This hybrid method has four stages consisting of image pre-processing, segmentation, feature extraction, and classification. Experimental results using a hybrid image pre-processing method apply on original MRI image data. The study shows that the Mini Mental State Examination (MMSE) method together with the residual intensity non-uniformity method in the SPM machine could make MRI image more effective with high accuracy and sensitive rates. The study shows that a well-designed pre-processing stage could improve the image quality of the subject for final classification. It can help researchers to gain high value of diagnosing early Alzheimer disease by filtering most of the image noise.

Impact of using mobile devices in earthquake

The natural disaster is an unstoppable reality that people are facing. Nowadays, mobile devices (MDs) such as smartphone, tablets, laptops are common to people, and they are using it in day-to-day life. This research work explores the use of MDs at the situation of the catastrophic earthquake in developing countries for rescue. The main concern of this research is to limited knowledge, infrastructure and access to MDs and the internet in developing countries of both people and rescue team. The universal proposed framework is introduced with low-cost balloon network, Finder Radar and Drone. The main contribution of this framework is the rescue operation after the catastrophic earthquake when all another internet connection will be interrupted. To establish communication between earthquake victims and rescue teams, balloon Wi-Fi network provides free Wi-Fi signals. Furthermore, Drone has considered as an earthquake area monitoring MDs and Finder Radar to search trapped earthquake survivals.

Impact of wireless communications technologies on elder people healthcare: Smart home in Australia

Over the last three decades, there has been a dramatic rise in ageing populations in most countries. Older people are remaining in nursing home care due to the fact that general services and medical support are provided. However, these costly environments often negatively affect older people due to high cost, limited staff and the social impacts they have. A way to overcome these challenges is to place the elderly instead in a smart home environment. The aim of this study is to describe the impact of wireless communications technologies on elderly people in a smart home. This has been compared with conditions in nursing homes. Using smart wireless sensors, wireless communications and ambient intelligent system, it is possible to create systems capable of measuring vital signs of the person in their own home. The findings indicate that wireless technology is the most suitable basis for a communications framework in a smart home to assist older people.

Upgrading Internet service provider (ISP) network in multiprotocol label switching (MPLS) and border gateway protocol (BGP) environment

This report describes a project which creates an upgraded topology based on the Internet service provider (ISP) environment. In the current service provider network, reliability, availability and management have become issues which need to be considered. Many technologies have been presented to solve these issues. This project is based on the virtual environment finished by simulators which are tools necessary to accomplish the project. In this project, two simulators are applied. One simulator creates a network topology which can be used to simulate real architecture, check all the protocols and process packets exchange. The second simulator deals with traffic. In the virtual environment, the simulated traffic will be created, collected and analysed. From this operation, a report about the advantages and disadvantages in both the current and the upgraded network environments will be generated. Detailed results are then compared, explained and clearly presented using data format.

Development of the Wireless Cisco Networking Laboratory pod

This paper aims to examine the issues related to the Traditional Cisco Laboratory system in particular and evaluate the issues regarding current solutions. This paper reports on an efficient and economical Laboratory system which can be accessed wirelessly by students eliminating the cabling issues and time wastage. The newly proposed Cisco Networking Laboratory is characterized by incorporating Virtual Network Computing (VNC) and Remote Desktop Protocol (RDP) as access protocol and Raspberry Pis as access nodes which are connected to the Cisco Devices. These Raspberry Pi computers act as the interface between the students and Cisco Devices and as well as they serve as a terminal or end computer in the Lab work. The implementation also demonstrates that the newly designed Wireless Cisco Networking Laboratory is perceived positively by IT students of the Study Group, Sydney. The evaluation of the proposed model will be further discussed for further use.