The most important fact in the mobile industry at the moment is that demand for wireless services will continue to expand in the coming years. Therefore, it is vital to find more spectrums through cognitive radios for the growing numbers of services and users. However, the spectrum reallocations, enhanced receivers, shared use, or secondary markets-will not likely, by themselves or in combination, meet the real exponential increases in demand for wireless resources. Network operators will also need to re-examine network architecture, and consider integrating the fibre and wireless networks to address this issue. This thesis involves driving fibre deeper into cognitive networks, deploying microcells connected through fibre infrastructure to the backbone LTE networks, and developing the algorithms for diverting calls between the wireless and fibre systems, introducing new coexistence models, and mobility management. This research addresses the network deployment scenarios to a microcell-aided cognitive network, specifically slicing the spectrum spatially and providing reliable coverage at either tier. The goal of this research is to propose new method of decentralized-to-distributed management techniques that overcomes the spectrum unavailability barrier overhead in ongoing and future deployments of multi-tiered cognitive network architectures. Such adjustments will propose new opportunities in cognitive radio-to-fibre systematic investment strategies. Specific contributions include: 1) Identifying the radio access technologies and radio over fibre solution for cognitive network infrastructure to increase the uplink capacity analysis in two-tier networks. 2) Coexistence of macro and microcells are studied to propose a roadmap for optimising the deployment of cognitive microcells inside LTE macrocells in the case of considering radio over fibre access systems. 3) New method for roaming mobiles moving between microcells and macrocell coverage areas is proposed for managing spectrum handover, operator database, authentication and accounting by introducing the channel assigning agent entity. The ultimate goal is to reduce unnecessary channel adaptations.

Master of Science / Department of Statistics / Paul I. Nelson / Capture-Recaputre (CR) experiments stemmed from the study of wildlife and are widely used in areas such as ecology, epidemiology, evaluation of census undercounts, and software testing, to estimate population size, survival rate, and other population parameters. The basic idea of the design is to use “overlapping” information contained in multiple samples from the population. In this report, we focus on the simplest form of Capture-Recapture experiments, namely, a two-sample Capture-Recapture design, which is conventionally called the “Petersen Method.” We study and compare the performance of three methods of constructing confidence intervals for the population size based on a Capture-Recapture design, asymptotic normality estimation, Chapman estimation, and “inverting a chi-square test” estimation, in terms of coverage rate and mean interval width. Simulation studies are carried out and analyzed using R and SAS. It turns out that the “inverting a chi-square test” estimation is better than the other two methods. A possible solution to the “zero recapture” problem is put forward. We find that if population size is at least a few thousand, two-sample CR estimation provides reasonable estimates of the population size.

As Web services grow in maturity and use, so do the methods which are being used to test and maintain them. Regression Testing is a major component of most major testing systems but has only begun to be applied to Web services. The majority of the tools and techniques applying regression test to Web services are focused on test-case generation, thus ignoring the potential savings of regression test selection. Regression test selection optimizes the regression testing process by selecting a subset of all tests, while still maintaining some level of confidence about the system performing no worse than the unmodified system. A safe regression test selection technique implies that after selection, the level of confidence is as high as it would be if no tests were removed. Since safe regression test selection techniques generally involve code-based (white-box) testing, they cannot be directly applied to Web services due to their loosely-coupled, standards-based, and distributed nature. A framework which automates both the regression test selection and regression testing processes for Web services in a decentralized, end-to-end manner is proposed. As part of this approach, special consideration is given to the concurrency issues which may occur in an autonomous and decentralized system. The resulting synchronization method will be presented along with a set of algorithms which manage the regression testing and regression test selection processes throughout the system. A set of empirical results demonstrate the feasibility and benefit of the approach.

The aim of this work is to examine the application Topocoding API and Google Maps API. Analyze system requirements to define the coverage of wireless connectivity to allow the ISP so as to allow the ISP to file AP and customers, as well as exploring the possibilities of joining the AP ISP and view coverage maps. Further, this application design, implement and test.

With advances in technology, modern integrated circuits have higher complexities and reduced transistor sizing. In deep sub-micron, the parameter variation-control is difficult and component delays vary from one manufactured chip to another. Therefore, the delays are not discrete values but are a statistical quantity, and statistical evaluation methods have gained traction. Furthermore, fault injection based gate-level fault coverage is non-scalable and statistical estimation methods are preferred. This dissertation focuses on scalable statistical methods to select critical paths in the presence of process variations, and to improve the defect coverage for complex integrated circuits. In particular, we investigate the sensitization probability of a path by a test pattern under statistical delays. Next, we investigate test pattern generation for improving the sensitization probability of a path, selecting critical paths that yield high defect coverage, and scalable method to estimate fault coverage of complex designs using machine learning techniques.

Submitted in fulfilment of the requirements for the degree of Master of Art in Journalism and Media Studies, University of the Witwatersrand, Johannesburg, 2016 / This is a story of undocumented migrants’ journeys between Malawi and Johannesburg, exposing the challenges they face and the corruption that takes place along the borders. Modern migration between the two countries has taken place for close to two hundred years. In the past, migrants, most of whom worked in the mines, were protected by law and that eased their movements. These days, low skilled migrants do not qualify for work permits so they stay in the country illegally. The South African law qualifies such migrants as undesirable visitors and bans them from returning to the country for some time. This research project documents how such migrants return home and come back to South Africa even before their bans expire. The research exposes how the migrants are aided by corrupt officers at the borders. The story shows how some of the migrants utilised a weakness in the old Malawi identification and passport system to obtain new travel documents under false names and return to South Africa undetected. This research project adopted an ethnographic approach. The findings are presented in a longform narrative story, which forms the first part of this document. The story is accompanied by a method document, which provides the theoretical framework and explains the methodology. / GR2017

The research was premised on the notion that gender is a social construct that is inextricably linked with masculinity and violence. This view of gender was used in an attempt to understand how masculinity and violence are constructed in the tabloid newspaper, the Daily Sun. The research used a discourse analysis, specifically looking at the various subject positions that are afforded to both men and women in this publication and how such positions foreclose other ways of thinking about and understanding gendered violence. The data set consisted of 23 information rich articles that were selected through the use of a purposeful sampling strategy. Results suggest while the Daily Sun positions itself as a medium that aims to ‘educate’ men (particularly working class men) against committing gendered violence, the Daily Sun may serve to reinforce the system of patriarchy that it claims to want to dismantle. In particular, the report highlights the problematic nature and implications of providing individual psychological explanations for gendered violence without acknowledging the contextual issues that are at play.

A research report submitted to the Faculty of Humanities, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Masters of Journalism and Media Studies, 2017 / This study investigates China’s media image in Zimbabwe. It does so through the lens of two leading newspapers: The Herald and NewsDay during the intense election campaign period of 2013. It deploys both quantitative and qualitative content analysis of newspaper articles drawn from The Herald and NewsDay in order to investigate the image of China in the Zimbabwean media. In terms of theoretical underpinnings, it draws from various media and communication theories. These include the broad area of image studies and framing theory to understand the media-constructed image of China before, during and after the 2013 electioneering period. These communication theories are tested and applied as a means of gaining insights into how the media shape images of China in Zimbabwe and in Africa generally. The findings of this study suggests that during this period the media image of China was projected in a more positive light in The Herald when compared to the negative images and depictions in the NewsDay. It shows that the economic consequence frame dominated the coverage of China thereby portraying the East Asian country as an economic partner, donor, economic saviour, development source and investor in Zimbabwe. The study also reveals that the construction of the media image of China is influenced by interplay of several factors such as the editorial slant of news producers, economic and political pressures influencing the polarised media environment in Zimbabwe. / XL2018

Research report submitted to the Faculty of Humanities, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Masters of Art (Journalism and Media Studies), Johannesburg, 2016 / Rape is a predominant crime and a social issue in South Africa today. South Africa’s incidence of rape is among the highest in the world. Identifying and understanding the dominant rape narratives in news media is useful in pinpointing how the media represents the crime of rape. It is understood through agenda-setting theory that news media plays an important role in how topics come onto the national agenda, giving news media a particular influence in society. Further, through discourse analysis and narrative theories, research has shown how what people read and hear can influence their understanding of those matters, and can drive social change or maintain the stability of social structures. Some theorists take this further, arguing that narrative fundamentally informs how humans make sense of the world, that reality is discursively constructed. The research below attempts to access, reveal and unpack these dominant narratives as they pertain to rape, using a combination of corpusbased analysis and critical discourse analysis techniques on two corpora of South African newspaper text from the first quarter of 2013, and tied to a specific case study, the rape and murder of Anene Booysen. The resultant findings also provide a snapshot of the dominant ideology and social practices in South Africa over the time period studied, as discourse and narrative are implicitly tied to power in society / GR2017

A brucelose bovina, causada principalmente pela Brucella abortus (B. abortus), é uma enfermidade infectocontagiosa responsável por perdas reprodutivas e que possui implicações com a saúde pública. O estado de Mato Grosso é um dos estados brasileiros mais importantes da pecuária nacional, sendo também o estado que apresenta a mais alta prevalência de brucelose bovina em nível nacional. Este estudo teve por objetivo propor um método para cálculo de cobertura vacinal contra brucelose bovina, de modo que fosse possível avaliar o impacto da vacinação sobre a dinâmica da enfermidade na população bovina existente no estado de Mato Grosso a partir da análise de informações registradas junto ao Instituto de Defesa Agropecuária do Mato Grosso INDEA/MT referentes aos exercícios de 2006-2010. Observou-se que em torno de 30% do total de fêmeas pertencem aos grupos etários em idade vacinal e que 47,18% do rebanho estadual foi vacinado durante os meses em que ocorreu a vacinação contra febre aftosa, demonstrando uma possível associação entre os procedimentos, que pode ser justificada pela economia e a maior praticidade de se manejar o rebanho uma única vez para se realizar as duas imunizações. O cálculo das vacináveis foi realizado com base na aproximação proporcional do efetivo de fêmeas considerando para cada grupo etário o intervalo da janela vacinal de seis meses. A taxa de cobertura vacinal foi estimada com base em método que considera a evolução de coortes de fêmeas nascidas ao longo de um ano. A obtenção das taxas reais de mortalidade e natalidade possibilitou a realização de novas simulações de modo a avaliar o real impacto dessa estratégia na redução da prevalência de brucelose bovina. Os resultados obtidos com base neste cálculo fornecem a melhor estimativa possível com base nos dados atualmente disponíveis na maioria dos estados. Conclui-se que é necessária uma padronização nacional na forma de cálculo de cobertura vacinal, que o estado do Mato Grosso manteve-se com níveis acima de 80% de cobertura vacinal entre os anos de 2006 a 2010 e considerando-se as taxas de natalidade e mortalidade reais e a perspectiva de manutenção deste nível de cobertura vacinal, estima-se demoraria um pouco mais de dez anos para reduzir em 50% a proporção de fêmeas positivas da população inicial e em torno de duas décadas para reduzir a prevalência ao nível de 2%. Vale ressaltar que este modelo considerou apenas o efeito da estratégia de vacinação, desconsiderando outros pontos de ação do programa como, por exemplo, a certificação de propriedades livres e o controle da movimentação animal no Estado, fatores estes que poderiam auxiliar na aceleração do processo de redução da prevalência no estado. / Bovine brucellosis, mainly caused by Brucella abortus (b. abortus), is an infectious disease responsible for reproductive losses and with public health implications. The State of Mato Grosso is one of the most important Brazilian states on livestock farming, as it is also the state that has the highest prevalence of bovine brucellosis in the nation. This study aimed to propose a vaccination coverage calculation method against bovine brucellosis, so that it is possible to evaluate the impact of vaccination on the disease dynamics within the bovine population existing in the State of Mato Grosso from the analysis of information recorded by the Institute of Agricultural Defense of Mato Grosso-INDEA/MT for the years 2006-2010. It was observed that around 30% of the total females belong to vaccination age groups and 47.18% of the State herd was vaccinated during the months in which foot-and-mouth disease vaccination occurred, demonstrating a possible association between the procedures, which can be justified economically and by the convenience of managing the herd only once to perform the two immunizations. The number of animals belonging to each vaccinated age group was calculated proportionally, considering a vaccination window of six months for each age group. The vaccination coverage rate was estimated using a method that considers the evolution of female cohorts born over the course of a year. Obtaining the actual birth and death rates led to the creation of new simulations in order to assess the actual impact of this strategy in the reduction of the prevalence of bovine brucellosis. The results obtained using this calculation provide the best possible estimate based on the data currently available in most states. It is concluded that national standardization is required in the form of vaccination coverage calculation and that the State of Mato Grosso remained with levels above 80% vaccination coverage from 2006 to 2010. Additionally, considering the actual birth and death rates and the prospect of maintaining this level of vaccination coverage, it is estimated that it would take slightly longer than ten years to reduce the proportion of the population initially positive for brucellosis by 50% and approximately twenty years to reduce the prevalence of the disease to 2%. It is worth mentioning that this model considers only the vaccination strategy, disregarding other programs such as the certification of brucellosis free herds and the control of animal movement in the state. These factors could also assist in the acceleration of the reduction of brucellosis prevalence in the state.