Mohamed N. Moustafa

Energy Efficient Routing protocol For Wireless Sensor Networks

In the last few years, wireless sensor networks have gained a lot of interest in the research field. WSNs consist of small nodes with sensing, computation, and wireless communications capabilities. Many routing protocols have been proposed for WSNs. Most of the hierarchical algorithms proposed for WSNs concentrate mainly on maximizing the lifetime of the network by trying to minimize the energy consumption, but delay is also an important metric that should be considered. In this paper, we propose pairs energy efficient routing protocol (PEER), a new routing protocol for WSNs that uses dual power management and focuses on minimizing both the energy dissipated and the delay cost. We have evaluated the performance of our protocol for two different cases: for all nodes having the same energy level at startup and for nodes starting with different energy levels. In terms of energy consumption, network lifetime, and average delay, our protocol has performed better than LEACH (a well-known hierarchical sensor network protocol) by more than 200%, 200%, and 400% respectively.

CIFAR-10: KNN-Based Ensemble of Classifiers

In this paper, we study the performance of different classifiers on the CIFAR-10 dataset, and build an ensemble of classifiers to reach a better performance. We show that, on CIFAR-10, K-Nearest Neighbors (KNN) and Convolutional Neural Network (CNN), on some classes, are mutually exclusive, thus yield in higher accuracy when combined. We reduce KNN overfitting using Principal Component Analysis (PCA), and ensemble it with a CNN to increase its accuracy. Our approach improves our best CNN model from 93.33% to 94.03%.

Gender Classification Using Mixture of Experts from Low Resolution Facial Images

In this study, we propose a novel two-stages mixture of experts scheme estimating gender from facial images. The first stage combines a couple of complementary gender classifiers with a third arbiter in case of decision discrepancy. Experimentally, we have verified the common thinking that one appearance-based (Haar-features cascade) classifier with another shape-based (landmarks positions metrology with SVM) classifier form a complementary couple. Subsequently, the second stage in our scheme is a Bayesian framework that is activated only when the arbiter cannot take a confident decision. We demonstrate that the proposed scheme is capable of classifying gender reliably from faces as small as 16x16 thumbnails on benchmark databases, achieving 95% gender recognition on FERET database, and 91.5% on the Labeled Faces in the Wild dataset.

GAME based dynamic resource scheduling in QoS aware radio access networks

Radio resource management and QoS are finally inseparable in wideband CDMA networks. In this paper, we propose a novel wireless resource scheduler, called GAME-C, that integrates our genetic algorithm for mobiles equilibrium (GAME) with the standard CDMA transmitter closed loop power control (CLPC). GAME assigns optimally both transmitting power and bit rate to every mobile station. Optimal allocation is in the sense that every user gets only enough resources necessary for meeting or exceeding its QoS requirements while minimizing interference to other users. Having done that, we gain further benefits as well. In addition to QoS provisioning, lower transmitting power extends a mobile station battery life. Moreover, the base station coverage efficiency is improved by decreasing the probability of blocking new connections or dropping current ones. In short, GAME-C expands the number of QoS-satisfied mobile stations in a cell. Various simulations show improvements achieved over the established (CLPC) basic scheme.

Landmarks-SIFT Face Representation for Gender Classification

Existing methods for gender classification from facial images mostly rely on either shape or texture cues. This paper presents a novel face representation that combines both shape and texture information for gender classification. We propose extracting the Scale Invariant Feature Transform (SIFT) descriptors at specific facial landmarks positions, hence encoding both the face shape and local-texture information. Moreover, we propose a decision-level fusion framework combining this Landmarks-SIFT with Local Binary Patterns (LBP) descriptor extracted for the whole face image. LBP is known of being tolerant against uncontrolled image capturing conditions. Competitive correct classification rates for both controlled (97% for FERET) and uncontrolled (95% and 94% for LFW and KinFace) benchmark datasets were achieved using our proposed decision-level fusion.

Outcome of venom bradykinin potentiating factor on rennin-angiotensin system in irradiated rats

Abstract Purpose: The objective of this work was to compare the effect of a bradykinin potentiating (BPF) isolated from venom of Egyptian scorpion Androctonus amoreuxi as a natural angiotensin converting enzyme inhibitor (ACEI) with losartan (LOS), a chemical angiotensin receptor blocker (ARB), in the modulation of radiation-induced damage. Material and methods: Rats were grouped into: (i) Control: untreated; (ii) + CBPF: Received intraperitoneally (i.p.) BPF 1 μg/g body weight (b.w.) (twice/week) during 3 weeks; (iii) + CLOS: Received i.p. LOS 5 μg/g b.w. (twice/week) during 3 weeks; (iv) R: Irradiated at 4 Gy; (v) R + BPF and (vi) R + LOS: Received BPF or LOS post-irradiation for 3 weeks. Results: BPF or LOS treatment induced a significant drop of sodium and uric acid. Irradiation induced a significant elevation of malondialdehyde (MDA) and advanced oxidation protein product (AOPP) associated with a significant decrease of glutathione (GSH) content in the kidney. Serum aldosterone, sodium, urea and creatinine levels showed a significant increase while a significant drop was recorded for haematological values, calcium and uric acid levels. Treatment of irradiated animals with BPF or LOS significantly improved radiation-induced changes. Conclusion: It could be concluded that the use of BPF as a natural product is comparable to the chemical compound LOS.

Software Quality model Analysis Program

It is vital that data is obtained so that actions can be taken to improve the performance. Such improvement can be measured in terms of improved quality, increased customer satisfaction and decreased cost of quality. Different researchers have proposed software quality models to help measure the quality of software products. These models often include metrics for this purpose. Some of the classical and recent models are discussed and analyzed in this paper showing the points of strength and weakness of each model type. A new comprehensive model is proposed and analyzed. A complete solution is discussed through the paper to enable an effective and efficient use of the proposed model to help the development team in prioritizing the important metrics while developing the software products according to some inputs from the user and the objectives of the software being developed. The solution developed is called the quality model analysis program (QAP) and is a fuzzy system that weights the proposed model attributes according to certain rules. The solution enables software project managers to better utilize their resources and take specific actions to better improve the quality of the software produced.

House Price Estimation from Visual and Textual Features.

Most existing automatic house price estimation systems rely only on some textual data like its neighborhood area and the number of rooms. The final price is estimated by a human agent who visits the house and assesses it visually. In this paper, we propose extracting visual features from house photographs and combining them with the houses textual information. The combined features are fed to a fully connected multilayer Neural Network (NN) that estimates the house price as its single output. To train and evaluate our network, we have collected the first houses dataset (to our knowledge) that combines both images and textual attributes. The dataset is composed of 535 sample houses from the state of California, USA. Our experiments showed that adding the visual features increased the R-value by a factor of 3 and decreased the Mean Square Error (MSE) by one order of magnitude compared with textual-only features. Additionally, when trained on the benchmark textual-only features housing dataset, our proposed NN still outperformed the existing model published results.

Reactive Collision Avoidance using Evolutionary Neural Networks.

Collision avoidance systems can play a vital role in reducing the number of accidents and saving human lives. In this paper, we introduce and validate a novel method for vehicles reactive collision avoidance using evolutionary neural networks (ENN). A single front-facing rangefinder sensor is the only input required by our method. The training process and the proposed method analysis and validation are carried out using simulation. Extensive experiments are conducted to analyse the proposed method and evaluate its performance. Firstly, we experiment the ability to learn collision avoidance in a static free track. Secondly, we analyse the effect of the rangefinder sensor resolution on the learning process. Thirdly, we experiment the ability of a vehicle to individually and simultaneously learn collision avoidance. Finally, we test the generality of the proposed method. We used a more realistic and powerful simulation environment (CarMaker), a camera as an alternative input sensor, and lane keeping as an extra feature to learn. The results are encouraging; the proposed method successfully allows vehicles to learn collision avoidance in different scenarios that are unseen during training. It also generalizes well if any of the input sensor, the simulator, or the task to be learned is changed.

A Mixture of Two Gender Classification Experts

This paper presents a novel method for combining the outputs of different gender classification techniques based on facial images. Merging the methods is performed by a committee machine using the Bayesian theorem. We implement and compare several well-known individual classifiers on four different datasets, then we experiment the proposed machine, and show that it significantly improves the accuracy of classification compared to individual classifiers. We also include results that address the effect of scale on the performance of classifiers.