45 research outputs found
Advanced Space-Time Coding: Near-Capacity Sphere-Packing, Multi-Functional MIMOs and Cooperative Space-Time Processing
No full textNear-capacity wireless transceivers and cooperative communications in the MIMO era: evolution of standards, waveform design, and future perspectives
Get PDFClassic Shannon theory suggests that the achievable channel capacity increases logarithmically with the transmit power. By contrast, the MIMO capacity increases linearly with the number of transmit antennas, provided that the number of receive antennas is equal to the number of transmit antennas. With the further proviso that the total transmit power is increased proportionately to the number of transmit antennas, a linear capacity increase is achieved upon increasing the transmit power, which justifies the spectacular success of MIMOs. Hence we may argue that MIMO-aided transceivers and their cooperation-assisted distributed or virtual MIMO counterparts constitute power-efficient solutions. In a nutshell, since the conception of GSM in excess of three orders of magnitude bit-rate improvements were achieved in three decades, which corresponds to about a factor ten for each decade, because GSM had a data rate of 9.6 Kb/s, while HSDPA is capable of communicating at 13.7 Mb/s. However, the possible transmit power reductions remained more limited, even when using the most advanced multistage iterative detectors, since the required received signal power has not been reduced by as much as 30 dB. This plausible observation motivates the further research of advanced cooperation-aided wireless MIMO transceivers, as detailed in this treatis
Near-Capacity Three-Stage Turbo Detection of Irregular Convolutional Coded Joint Sphere-Packing Modulation and Space-Time Coding
No full textThe value of journaling on multimodal materials: a literacy narrative and autoethnography of an experienced Saudi high school English teacher
No full textThe author reflects on his literacy in both Arabic and English while tracing his path to becoming an English major at King Khalid University and an English teacher in boys’ high schools—first in the village of Hajur and then in Jeddah. He describes his first two years teaching in this remote village made up of two main tribes in western Saudi Arabia. In fourth grade in a class of six boys in Alkanahbilah—also a small village—the author discovered his love for the Arabic language as Mohammed Aziz, his beloved teacher, read Arabic poems aloud to the boys every day, and they memorized the daily poem each evening. Providing his own written artifacts—primarily journals from graduate seminars in the United States—he explains the benefits of daily writing in English and daily viewing of multimodal materials while offering five original lessons he intends to teach alongside the national core curriculum upon returning to his homeland
Contextual relabelling of detected objects
Get PDFThis is the author accepted manuscript. The final version is available from IEEE via the DOI in this recordContextual information, such as the co-occurrence of objects and the spatial and relative size among objects provides deep and complex information about scenes. It also can play an important role in improving object detection. In this work, we present two contextual models (rescoring and re-labeling models) that leverage contextual information (16 contextual relationships are applied in this paper) to enhance the state-of-the-art RCNN-based object detection (Faster RCNN). We experimentally demonstrate that our models lead to enhancement in detection performance using the most common dataset used in this field (MSCOCO)
Educating Saudi Women through Communicative Language Teaching: A Bi-literacy Narrative and An Autoethnography of a Saudi English Teacher
No full textThis project is the bi-literacy narrative and autoethnography of a novice English teacher who earned the undergraduate degree in English Language at King Abdulazziz University in Jeddah, Saudi Arabia. The author offers a brief literacy narrative of her experience learning Arabic in girls’ schools on a military base in Tabuk in northern Saudi Arabia and at the university. Through strategic contemplation, she reflects on her mother’s illiteracy, which was common for women of her mother’s generation, while focusing on her mother’s strengths, especially her commitment to educating her six children—all of whom have earned college degrees. In the autoethnography, she traces her growth as a teacher and her interest in applying Communicative Language Teaching (CLT) to the education of women in Saudi Arabia, focusing on creating materials that reinforce Islamic values. She offers 11 written artifacts created during her graduate studies as well as three sample lessons she designed in her curriculum seminar and on lesson co-created by her and her thesis chair and internship supervisor. Rather than including just the four traditional macro-skills—reading, writing, speaking, and listening—she adds viewing and visually representing (Tompkins, 2016). In so doing, she refined and expanded basic CLT—sometimes described as a “fuzzy” method by applying the NCTE whole language educational philosophy to her original curriculum based on biographies of women
Kinetics analysis of degradation of polynorbornene containing aryl or hetaryl azo dye
No full textThe Vyazovkin model-free method were applied to study non-isothermal degrdn. of an aryl and hetaryl azo dye-contg. polynorbornene using a differential scanning calorimetric (DSC) technique. The samples were heated in nitrogen, with heating rates in the range 5-50 K min-1. The isoconversional method of Vyazovkin was used to det. the variation of the activation energy for degrdn. with temp., E α(T). The values of E α(T) were found to be almost independent of temp. The results of isoconversional methods used suggest that the transition follows a single-step kinetics reaction mechanism.Source type: Electronic(1
Implicit and Explicit Attention for Zero-Shot Learning
Get PDFThis is the author accepted manuscript. The final version is available from Springer via the DOI in this recordPattern recognition: 43rd DAGM German Conference, DAGM GCPR 2021, 28 September - 1 October, Bonn, Germany, edited by Christian Bauckhage, Juergen Gall, and Alexander SchwingMost of the existing Zero-Shot Learning (ZSL) methods focus on learning a compatibility function between the image representation and class attributes. Few others concentrate on learning image representation combining local and global features. However, the existing approaches still fail to address the bias issue towards the seen classes. In this paper, we propose implicit and explicit attention mechanisms to address the existing bias problem in ZSL models. We formulate the implicit attention mechanism with a self-supervised image angle rotation task, which focuses on specific image features aiding to solve the task. The explicit attention mechanism is composed with the consideration of a multi-headed self-attention mechanism via Vision Transformer model, which learns to map image features to semantic space during the training stage. We conduct comprehensive experiments on three popular benchmarks: AWA2, CUB and SUN. The performance of our proposed attention mechanisms has proved its effectiveness, and has achieved the state-of-the-art harmonic mean on all the three datasets.Defence Science and Technology LaboratoryAlan Turing Institut
Preparation and characterisation of thin film CdS/CdTe solar cells produced by close space sublimation
No full textThis work concerns the preparation and characterisation of n-CdS/p-CdTe thin film solar cells using a close space sublimation system (CSS) to deposit the p-CdTe layer. The effects of preparation conditions of CdTe on growth rate and the texture of the film were investigated. The cells were characterised by measurements of I-V, C-V, spectral response, photocapacitance and electroluminescence. The growth rate of CdTe increased as the source temperature increased with an activation energy #approx# 1.9 eV, whereas it was constant and independent of the substrate temperature, up to some breakpoint temperature, above which the rate decreased rapidly to zero. Increasing the pressure during growth allows growth to take place at higher temperatures but at the expense of growth rate. The growth rate increased as the separation between the source and the substrate decreased. The CdTe films grown at 335 deg. C substrate temperature showed a highly preferred (111) orientation and small grain size, whereas the films deposited at higher temperature showed a progressive loss of (111) orientation and an increase in grain size. The CdCl_2 treatment for the CdTe layer increased the efficiency of the solar cells from about 3% to around 10%. The I-V dark characteristics suggest that the CdCl_2 treatment changed the current transport mechanism across the junction from emission/recombination with activation energy around 0.07eV to a tunnelling mechanism. Solar cells fabricated with SnO_2 give higher efficiencies than cells fabricated with ITO due to the interface between the ITO and CdS layer, which becomes rectifying after heating at 400 deg. C in air. There are many preparation parameters which affect the interdiffusion between the CdTe and CdS layer, and so modify the spectral response. In the present study, the effects of the CdCl_2 treatment of CdTe and CdS layers, the thickness of CdCl_2 and the time of annealing, the thickness of the CdS layer and the substrate temperature were investigated in terms of the spectral response. These studies tended to confirm that interdiffusion had taken place, with significant effects on the spectral response. The photocapacitance measurements indicated the existence of some levels below the conduction band at 0.4, 0.2 eV in p-CdTe and at 0.55eV in n-CdS, also some deep levels above the valance band at 0.24, 0.1 eV in p-CdTe and 0.71 eV in n-CdS. (author)SIGLEAvailable from British Library Document Supply Centre-DSC:DXN029011 / BLDSC - British Library Document Supply CentreGBUnited Kingdo
A Large-Scale Category-Based Evaluation of A Visual Language for Adaptive Hypermedia
Get PDFAdaptive Hypermedia (AH) provides a personalised and customised approach, enhancing the usability of hypermedia, by building a model of various qualities of a user and applying this information to adapt the content and the navigation to their requirements. However, authoring adaptive materials is not a simple task, as an author may be pressed for time, or simply lack the skills needed to create new adaptive materials from scratch. The most challenging part is the authoring of the adaptation specification (adaptive behaviour rules). This paper tackles this challenge by proposing and evaluating (on a large scale) a visual language for authoring of adaptive hypermedia
