[pt] Atualmente, há uma demanda crescente por serviços que exploram a integração entre as tradicionais redes de pacotes (como a Internet) e as redes celulares de última geração. Além das aplicações que visam ao transporte de dados em si, há linhas de pesquisa no sentido de oferecer serviços de voz sobre IP em redes celulares, ou seja, explorando o paradigma da transmissão de voz em pacotes. Uma destas aplicações é o Push to Talk over Cellular, ou PoC, serviço no qual, durante uma sessão, apenas um usuário por vez pode falar em modo half duplex. No PoC, assim como em toda aplicação baseada em voz sobre IP, o atraso fim-a-fim, a variação do atraso (jitter) e as taxas de perdas (além de outros fatores) influenciam a qualidade da recepção da voz nos extremos finais da comunicação. Os resultados aqui apresentados estão focados no requisito de atraso fim-a-fim dos pacotes de voz, cujos valores variam dependendo do tamanho destes pacotes, do número de usuários por canal e, principalmente, das condições do meio (razão C/I). Também são apresentados resultados referentes às simulações que validaram o modelo da rede EGPRS, mais especificamente, referentes às características da interface aérea baseada no EDGE. Durante as simulações, considerou-se que os canais eram dedicados às sessões PoC, isto é, sem nenhuma espécie de tráfego concorrente. / [en] Nowadays, there is a increasing demand for services that join traditional packet data networks (like Internet) and last generation cellular systems. Besides applications based on data only, there are researches about offer services of voice calls in cellular networks based on packet switched transmission mode. One of that is Push to Talk over Cellular, or PoC, that is a real-time service based on delivery of voice traffic in packets between two or more parties in a half duplex way. Like traditional VoIP services, the end-to-end delay, the jitter and the lost rates can affect the transmission and reception of the voice between the extremes of the communication. The results are focused on end-to-end delay of the voice packets in a PoC session, whose values can change drasticly depending on the size of the packets, the number of users sharing a channel and, mainly, the conditions of the medium (C/I rates). The model used for the EGPRS network, in special, the EDGE air interface, is studied by simulations and the results are presented and analysed too. This work presents a mean of estimating the capacity of EGPRS/EDGE channels of supporting PoC sessions without considering any other type of traffic.

This thesis investigates embedded databases and graphical interfaces for the MicroBaseJ project. The project aim is the development of an integrated database and GUI user interface for a typical 3G, or 2.5G, mobile phone with Java MIDP2 capability. This includes methods for data acquisition, mobile data and information communication, data management, and remote user interface. Support of phone delivered informatics will require integrated server and networking infrastructure research and development to support effective and timely delivery of data for incorporation in mobile device-based informatics applications. A key research and development (R&D) challenge is to support effective and timely delivery of data for incorporation in mobile device-based informatics applications. Another important aspect of the project is determining how to develop efficient graphics for the small mobile screen. The research investigates and analyses the architecture of a mobile monitoring system. The project developed a generic solution that can be implemented in a number of commercial sectors, such as horticulture, building management and pollution/water management. The developed concept is tested using data relevant to the horticultural area of application. The system also addresses the main issues related to mobile monitoring, including real-time response, data integrity, solution cost, graphical presentation, and persistent storage capabilities of modern mobile devices. Four embedded databases based on J2ME have been investigated. Two of the four have been evaluated and analysed. The Insert function, Sequence Search, and Random Search of Perst List and RMS (Record Management System) databases have been tested. The size of the processed data was limited to 20,000 records when using the wireless toolkit simulator, and 11,000 records when using a mobile phone. Perst Lite reflects good performance and has out-performed RMS in all tests. User interface software such as J2ME Polish for mobile phones has been investigated. Custom J2ME class for graphical interface is developed. This provides the graphical presentation of the data collected from the sensors; including temperature, wind speed, wind direction, moisture, and leaf wetness. The graphical interface, bar charts, and line charts with trace ball for collected data have been designed and implemented. The embedded database performance and project performance have been investigated and analysed. The performances of Perst Lite and RMS are evaluated in terms of the insert, sequence search, and random search functions based on simulation and real devices. The record numbers vary from 1,000 to 20,000. The project performance contains data receiving and storage, and data presentation and configuration. The performance of data storage and configuration can be negated due to the running mode and the response time. Thus, data presenting performance is the key focus in this project. This performance was divided into the categories of initial, data search, data selection, and charting. The initial performance includes the initialisation of the project parameters, and the reaching of the welcome interface. Data search performance refers to the retrieval of the specified data from the embedded database, measured on 48 data points, which only can be presented on the mobile screen from the retrieved data. These four performance types are measured in thousands of record numbers, varying from 1,000 to 18,000 record numbers, with the retrieved data range varying from 1 day to 30 days.

M. Tech. Business Administration / The objective of this study was to identify factors that are responsible for employee absenteeism in Vodacom SA, which is the largest mobile cell phone company in South Africa. In the literature, it has been extensively reported that absenteeism is a major cause of loss of income and revenues among mobile cell phone operators in South Africa. The purpose of this study was to identify the root causes of absenteeism in the cellular phone industry in South Africa. Data was collected from a random sample of 120 employees of Vodacom SA working in Midrand.

M. Tech. Electrical Engineering. / Discusses Call Admission Control (CAC) is a standard operating procedure responsible for accepting or rejecting calls based on the availability of network resources. It is also used to guarantee good Quality of Service (QoS) to ongoing users. However, there are a number of imprecisions to consider during the admission and handoff processes. These uncertainties arise from the mobility of subscribers and the time-varying nature of key admission factors such as latency and packet loss.These parameters are often imprecisely measured, which has a negative impact on the estimation of a channel spectral efficiency. In mobile networking, greater emphasis is towards delivering good QoS to real-time (RT) applications. It has become increasingly necessary to develop a model capable of handling uncertainties associated with the network in order to improve the quality of decisions relating to CAC. Type-1 and Type-2 Fuzzy Logic Controllers (FLCs) were deployed to allow the CAC to make better decisions in the presence of numerous uncertainties. The model successfully proposed associated meanings and degrees of certainty to the measured values of loss and latency by means of fuzzy sets and Membership Functions (MFs). The results obtained show that the fuzzy-based CAC performs better by reducing the call blocking and call dropping probabilities which are some of the key measurement parameters of QoS in wireless networking.

D. Tech. Electrical Engineering. / Objectives of the study is to propose a suitable feature extraction and classication approach that is capable of adapting to the non-linear nature and the noise contained in the time series data. The end goal of subscriber classication in this study is to utilise the subscriber information extracted for a new radio resource optimisation model that focuses on the Channel Allocation Problem CAP. Although there they have been various models proposed in literature for solving of the CAP problem, the utilisation of subscriber related information in the CAP has not been directly considered.

M. Tech. Information Technology / Voting is a widely spread and democratic way of making decisions. For centuries, South Africa has been using the popular paper-based voting system though it does not provide the desirable blend of accessibility and efficiency. Missing ballot papers, invalid votes and miscount are some of the challenges that come with the paper-based voting system. Time is precious to everyone and correct information is even more precious in the fast-paced information technology era. Merging all these together we look in for easier, effective, efficient and time managing way of casting a vote. Mobile phones facilitate communication and rapid access to information and their diffusion has reached a larger proportion of the population in a short period of time. In this thesis, we propose designing and developing a mobile phone voting system, which allows users to vote spontaneous and timeous, using existing mobile phone networks and technologies via their mobile phones.

WCDMA promises to achieve high-speed internet, high quality image transmission and high-speed data services with larger system capacity. However, Multiple Access Interference is one of the major causes of transmission impairment, which reduces the link capacity in WCDMA systems. The Adaptive Antenna Array (AAA) technique reduces multiple access interference by directing antenna beam nulls towards the interfering signals by weighting the received signals from all antennas before combining the signals. With the very rapid advancement of wireless personal communications services, a new challenge to the cellular industry is the integration of multiple systems and applications on a single device. A software radio technique offers a possible solution to achieve this goal including international roaming and multiple standard operations within the same geographical area. The main attraction of a software radio is it's flexibility, in that it can be programmed for emerging cellular standards allowing it to be updated with new software without any changes in the hardware infrastructure. A software radio incorporating adaptive array beamforming at the receiver can increase the total carried traffic in a system and transmit power while the probability of call blocking and forced termination can also be decreased. This dissertation examines WCDMA, AAA, power control and software radio techniques in the world of wireless communication systems. Once the theoretical background of CDMA and AAA has been substantiated, the thesis establishes the need for power control in mobile systems by examining simulation results. An AAA receiver with six antenna elements is proposed and evaluated in different environments as a precursor to implementation. It can be inferred that when the link is interference limited, the link capacity can be increased and it has been shown that the AAA receiver with six antenna elements increases the link capacity to about 2.9 times that of the single antenna RAKE receiver. This thesis also examines the basic concepts of VHDL and considers this as the principle means to program reconfigurable core FPGA's in the software radio. A three-layered (PC/DSP/FPGA) software radio test bed is used to implement an AAA receiver. The architecture of the test bed is designed in such a way that it can be used to evaluate the performance of various FPGA based transceivers and coding schemes etc. Many of the desirable features and flexibilities inherent in the software radio concept are available on this test bed and the system has proved to be capable of high speed digital processing and is ideally suited to the development of time critical system components. The bit error rate achieved using the implemented receiver is assessed and compared to simulation results in an environment incorporating Rayleigh fading and AWGN. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2001.

I den föreliggande studien undersöktes mobilljudens effekt på minnesbehållning i klassrumsmiljö. I Studien undersöktes även vilken teori – unitary teorin eller duplex-mekanism teorin – som lämpligast förklarade mobilljudens eventuella effekter på kognitiva mekanismer. Metoden bestod i att deltagarna tittade på en kort dokumentär och därefter utförde ett minnestest, som berörde dokumentärens innehåll. Experimentgruppen fick titta på dokumentären, samtidigt som ringsignaler och aviseringsljud spelades upp under fyra perioder i dokumentären, medan kontrollgruppen inte utsattes FÖR buller. Deltagarnas hågkomst av innehållet av dessa fyra perioder prövades genom testfrågor. 27 försökspersoner deltog i experimentet, varav 11 personer var i kontrollgruppen och resten i experimentgruppen. Resultatet visade inte någon effekt av mobilljud på deltagarnas minnesbehållning. Resultatet stödjer dock duplex-mekanism teorin, vilket betyder att unitary teorin inte stöds i den föreliggande studien. Däremot, försvårar metodologiska brister och lågt antal deltagare tolkningen av resultatet. Förslag för framtida forskning inbegriper metodologiska förbättringar och undersökningsområden. / The aim of the present study was to investigate the effect of cell phone noise on memory retention in classroom environment. A second purpose of the study was to determine which theory – unitary account or duplex-mechanism account – would best explain the effect of cell phone noise on cognitive mechanisms. The method required the participants to watch a short documentary and afterwards complete a memory test, consisting of gist questions. The experiment group was simultaneously being exposed to ringtones and text message signals of a cell phone during four episodes of the documentary, whereas the control group was not exposed to noise. The participants’ memory retention of the content of those four episodes was tested via so called test items. 27 participants participated in the experiment, where 11 of them were in the control group and the rest in the experiment group. The results failed to find an effect of cell phone noise on participants’ memory retention. The results support the duplex-mechanism account, which means that the unitary account is not supported in the present study. However, methodological flaws and the small sample complicate the interpretation of the results. Recommendations for future research involve methodological improvements and research areas.

This thesis investigates embedded databases and graphical interfaces for the MicroBaseJ project. The project aim is the development of an integrated database and GUI user interface for a typical 3G, or 2.5G, mobile phone with Java MIDP2 capability. This includes methods for data acquisition, mobile data and information communication, data management, and remote user interface. Support of phone delivered informatics will require integrated server and networking infrastructure research and development to support effective and timely delivery of data for incorporation in mobile device-based informatics applications. A key research and development (R&D) challenge is to support effective and timely delivery of data for incorporation in mobile device-based informatics applications. Another important aspect of the project is determining how to develop efficient graphics for the small mobile screen. The research investigates and analyses the architecture of a mobile monitoring system. The project developed a generic solution that can be implemented in a number of commercial sectors, such as horticulture, building management and pollution/water management. The developed concept is tested using data relevant to the horticultural area of application. The system also addresses the main issues related to mobile monitoring, including real-time response, data integrity, solution cost, graphical presentation, and persistent storage capabilities of modern mobile devices. Four embedded databases based on J2ME have been investigated. Two of the four have been evaluated and analysed. The Insert function, Sequence Search, and Random Search of Perst List and RMS (Record Management System) databases have been tested. The size of the processed data was limited to 20,000 records when using the wireless toolkit simulator, and 11,000 records when using a mobile phone. Perst Lite reflects good performance and has out-performed RMS in all tests. User interface software such as J2ME Polish for mobile phones has been investigated. Custom J2ME class for graphical interface is developed. This provides the graphical presentation of the data collected from the sensors; including temperature, wind speed, wind direction, moisture, and leaf wetness. The graphical interface, bar charts, and line charts with trace ball for collected data have been designed and implemented. The embedded database performance and project performance have been investigated and analysed. The performances of Perst Lite and RMS are evaluated in terms of the insert, sequence search, and random search functions based on simulation and real devices. The record numbers vary from 1,000 to 20,000. The project performance contains data receiving and storage, and data presentation and configuration. The performance of data storage and configuration can be negated due to the running mode and the response time. Thus, data presenting performance is the key focus in this project. This performance was divided into the categories of initial, data search, data selection, and charting. The initial performance includes the initialisation of the project parameters, and the reaching of the welcome interface. Data search performance refers to the retrieval of the specified data from the embedded database, measured on 48 data points, which only can be presented on the mobile screen from the retrieved data. These four performance types are measured in thousands of record numbers, varying from 1,000 to 18,000 record numbers, with the retrieved data range varying from 1 day to 30 days.

Thesis (Ph. D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2008. / Laskar, Joy, Committee Chair ; Cressler, John, Committee Member ; Kornegay, Kevin, Committee Member ; Tentzeris, Emmanouil, Committee Member ; Lee, Chang-Ho, Committee Member.