With open source router firmware being used for various tasks that would be hard to achieve for the standard closed source router firmware, it is important to compare the two in terms of performance. This study aimed to study the differences in performance between open source router firmware and that of closed source router firmware. In addition to have measured bandwidth, packet loss and response time has also been measured in order to make it easier for companies/people to make informed decisions regarding whether to use open source router firmware or not. To further help with decision making, a qualitative study was made to gather data regarding how easy each firmware is to configure and how secure they are. There were some larger differences between the open source router firmware and that of the closed source router firmware. The closed source router firmware performed better when it came to bandwidth, whereas the open source firmware got better response time and overall better stability.

The growth of the Internet has led to research and development of several new and useful applications including video streaming. Commercial experiments are underway to determine the feasibility of multimedia broadcasting using packet based data networks alongside traditional over-the-air broadcasting. Broadcasting companies are offering low cost or free versions of video content online to both guage and at the same time generate popularity. In addition to television broadcasting, video streaming is used in a number of application areas including video conferencing, telecommuting and long distance education. Large scale video streaming has not become as widespread or widely deployed as could be expected. The reason for this is the high bandwidth requirement (and thus high cost) associated with video data. Provision of a constant stream of video data on a medium to large scale typically consumes a significant amount of bandwidth. An effect of this is that encoding bit rates are lowered and consequently video quality is degraded resulting in even slower uptake rates for video streaming services. The aim of this dissertation is to investigate peer-to-peer streaming as a potential solution to this bandwidth problem. The proposed peer-to-peer based solution relies on end user co-operation for video data distribution. This approach is highly effective in reducing the outgoing bandwidth requirement for the video streaming server. End users redistribute received video chunks amongst their respective peers and in so doing increase the potential capacity of the entire network for supporting more clients. A secondary effect of such a system is that encoding capabilities (including higher encoding bit rates or encoding of additional sub-channels) can be enhanced. Peer-to-peer distribution enables any regular user to stream video to large streaming networks with many viewers. This research includes a detailed overview of the fields of video streaming and peer-to-peer networking. Techniques for optimal video preparation and data distribution were investigated. A variety of academic and commercial peer-to-peer based multimedia broadcasting systems were analysed as a means to further define and place the proposed implementation in context with respect to other peercasting implementations. A proof-of-concept of the proposed implementation was developed, mathematically analyzed and simulated in a typical deployment scenario. Analysis was carried out to predict simulation performance and as a form of design evaluation and verification. The analysis highlighted some critical areas which resulted in adaptations to the initial design as well as conditions under which performance can be guaranteed. A simulation of the proof-of-concept system was used to determine the extent of bandwidth savings for the video server. The aim of the simulations was to show that it is possible to encode and deliver video data in real time over a peer-to-peer network. The proposed system achieved expectations and showed significant bandwidth savings for a sustantially large video streaming audience. The implementation was able to encode video in real time and continually stream video packets on time to connected peers while continually supporting network growth by connecting additional peers (or stream viewers). The system performed well and showed good performance under typical real world restrictions on available bandwith capacity. / Dissertation (MEng)--University of Pretoria, 2009. / Electrical, Electronic and Computer Engineering / unrestricted

In this thesis a technique that is being used in another area of technology to optimize light reception in a photographic camera was also applied to the dielectric resonator antenna. The technique consisting of the use of thin film to couple the media and camera impedances resulted in a dielectric resonator antenna bandwidth enhancement technique. The bandwidth enhancement technique was found when thin film dielectric layer structure was used to couple the dielectric resonator and its feed mechanism. Remarkable good performance was detected with a coplanar waveguide fed cylindrical dielectric resonator antenna which resulted in an improvement to its fractional bandwidth from 7.41% to 50.85%. Extensive experimental work was undertaken in order to explore the extent offered in bandwidth performance by using thin film dielectric layer structure in the dielectric resonator antenna performance. The experimental tasks were designed in order to investigate the influence of the thin film dielectric layer structure in relation to its size, shape, thickness, position and direction. Experimental results were supported with simulation work with the computer simulation technology microwave studio. The pieces of the material used for undertaking this experimental work were manually handcrafted. Four different dielectric resonator antenna designs were used in order to carry out the experimental work including the coplanar waveguide fed cylindrical dielectric resonator antenna. The other three dielectric resonator antennas were implemented using the same microstrip feed mechanism. Improved performance in bandwidth was achieved for all the designs. Optimization of the incoming signal was observed when a piece of thin film dielectric layer structure was placed in position between the feed mechanism and the dielectric resonator antenna. The optimization was observed as an enhancement in both the return loss level and the bandwidth of work. Different unexpected operational modes from were activated, such modes being called perturbed modes. Two different shapes were used in this project. Cylindrical dielectric resonator antenna (ɛr = 37) from a commercial provider and two novel rectangular dielectric resonator antennas. The novel rectangular dielectric resonator antennas were created with the methodology presented in this thesis. The rectangular dielectric resonator antennas were elaborated with transparent ceramic material (ɛr = 7) and TMM10i (ɛr = 9.8) from the Rogers Corporation company. The bandwidth enhancement technique was tested in novel embedded dielectric resonator antennas. A coplanar waveguide fed embedded cylindrical dielectric resonator antenna achieved a maximum bandwidth enhancement of 156.77% around f = 3.79 GHz with a thin film dielectric layer structure modified rectangular piece on one edge. Escalation to dielectric resonator antenna design at millimeter wave frequencies was achieved by using thin film dielectric layer structure bandwidth enhancement technique and a handcrafted printed circuit board millimeter wave feed mechanism. The millimeter wave feed mechanisms were achieved using a low cost alternative technique conceived as part of this project. Millimeter wave dielectric resonator antennas were implemented using thin film dielectric layers structure. The antennas deliver an adequate performance in bandwidth. The work presented in this thesis demonstrates dielectric resonator antenna simpler geometry, simple couple schemes, small size, low profile, light weight, and ease of excitation and orientation. Other parameters have also been investigated covering reduced complexity, high degree of flexibility, ease of fabrication and the use of low cost technology to escalate to millimeter wave frequencies.

abstract: With the push for integration, a slew of modern switching power management circuits are operating at higher switching frequencies in order to reduce passive filter sizes. But while these switching regulators provide power conversion at high efficiencies, their output is prone to ripples due to the inherent switching behavior. These switching regulators use linear-low dropout regulators (LDOs) downstream to provide clean supplies. Typically, these LDOs have good power supply rejection (PSR) at lower frequencies but this degrades at higher frequencies. Therefore, some residual ripple is still manifested on the output. Because of this, high power supply rejection (PSR) with a wide rejection frequency band is becoming a critical requirement in linear low-dropout regulators (LDOs) used in complex systems- on-chip (SOCs). Typical LDOs achieve higher PSR within their loop-bandwidth; however, their supply rejection performance degrades with reduced loop-gain outside their loop- bandwidth. The LDOs with external filtering capacitors may also have spectral peaking in their PSR response, causing excess system- level supply noise. This work presents an LDO design approach, which achieves a PSR of higher than 68 dB up to 2 MHz frequency and over a wide range of loads up to 250 mA. The wide PSR bandwidth is achieved using a current-mode feedforward ripple canceller (CFFRC) amplifier which provides up to 25 dB of PSR improvement. The feedforward path gain is inherently matched to the forward gain of the LDO, not requiring calibration. The LDO has a fast load transient response with a recovery time of 6.1μs and has a quiescent current of 5.6μA. For a full load transition, the LDO achieves settling with overshoot and undershoot voltages below 27.6 mV and 36.36 mV, respectively. The LDO is designed and fabricated in a 180 nm bipolar/CMOS/DMOS (BCD) technology. The CFFRC amplifier helps to achieve low quiescent power due to its inherent current mode nature, eliminating the need for supply ripple summing amplifiers and adaptive biasing. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2020

<p>A large volume of data is continuously being generated by data centers, humans, and the internet of things (IoT). In order to get useful insights, such enormous data must be processed in time with high throughput, low latency, and high accuracy. To meet such performance demands, a large body of new hardware is being shipped by vendors, such as multi-core CPUs, 3D-stacked memory, embedded microcontrollers, and other accelerators.</p><br><p>However, traditional operating systems (OSes) and data analytics frameworks, the key layer that bridges high-level data processing applications and low-level hardware, fails to deliver these requirements due to quickly evolving new hardware and increases in explosion of data. For instance, general OSes are not aware of the unique characters and demands of data processing applications. Data analytics engines for stream processing, e.g., Apache Spark and Beam, always add more machines to deal with more data but leave every single machine underutilized without fully exploiting underlying hardware features, which leads to poor efficiency. Data analytics frameworks for machine learning inference on IoT devices cannot run neural networks that exceed SRAM size, which disqualifies many important use cases.</p><br><p>In order to bridge the gap between the performance demands of data analytics and the new features of emerging hardware, in this thesis we exploit runtime system designs for high-level data processing applications by exploiting low-level modern hardware features. We study two important data analytics applications, including real-time stream processing and on-device machine learning inference, on three important hardware platforms across the Cloud and the Edge, including multicore CPUs, hybrid memory system combining 3D-stacked memory and general DRAM, and embedded microcontrollers with limited resources. </p><br><p>In order to speed up and enable the two data analytics applications on the three hardware platforms, this thesis contributes three related research projects. In project StreamBox, we exploit the parallelism and memory hierarchy of modern multicore hardware on single machines for stream processing, achieving scalable and highly efficient performance. In project StreamBox-HBM, we exploit hybrid memories to balance bandwidth and latency, achieving memory scalability and highly efficient performance. StreamBox and StreamBox-HBM both offer orders of magnitude performance improvements over the prior state of the art, opening up new applications with higher data processing needs. In project SwapNN, we investigate a system solution for microcontrollers (MCUs) to execute neural networks (NNs) inference out-of-core without losing accuracy, enabling new use cases and significantly expanding the scope of NN inference on tiny MCUs. </p><br><p>We report the system designs, system implementations, and experimental results. Based on our experience in building above systems, we provide general guidance on designing runtime systems across hardware/software stack for a wider range of new applications on future hardware platforms.</p><div><br></div>

This master's thesis focuses on design and implementation of a flow allocator policy which supports bandwidth reservation for the Recursive InterNetwork Architecture (RINA). Each flow has some dedicated bandwidth, which is guaranteed to be available during the whole lifetime of the flow. The allocator, which operates as a distributed system, attempts to find a suitable path in the network graph. To achieve this goal, it must keep the information about link utilization up to date. The proposed allocator has been implemented in the open source project rlite. The first half of the thesis is concerned with congestion control theory, and also studies a number of algorithms used in TCP. Additionally, it contains an overview of the structure of RINA and the Raft consensus algorithm.

Due to a signifikant development of computer networks in the last few years the demands for measurement of network metrics increased. This master´s thesis is dealing with standards for testing and test techniques used for the measurement of network metrics. Further it compares the existing free accessible measurement tools and metrics measures them. Finaly the thesis draws up a web application for measurement of the basic network metrics.

Master‘s thesis is focused on basic characteristics of data transmissions in computer networks. In forepart I digestedly described basic characteristics computer networks, metods of their measurement and valuation. Computer networks type LAN as well as wireless computer networks Wifi, are mentioned. In second part I realized my own proposition of basic characteristics measurement. This proposition is used for SW applications development. Main goal of these applications is measurement of packet delay, packet jitter, paket loss and available bandwidth. I described details of SW applications and realized measurement of real computer networks in the end.

Circular polarization is valuable for many electromagnetic radiation applications such as wireless and satellite communication, radars, RFID, global positioning systems, etc. Many efforts have been made to manipulate and control polarization by using linear-to-linear or linear-to-circular transmission or reflection polarization converters. Most of the existing linear-to-circular single-layer polarizers have been found to be narrowband. Some attempts have been made to improve the bandwidth of these polarizers including using multiple layered structures at the expense of a bulkier device. There was, however, still a requirement for thin single-layer linear-to-circular polarizers with enhanced bandwidth. The purpose of this research was to design two thin single-layer linear-to-circular polarizers, one for transmission and the other for reflection, with enhanced bandwidth. A thin single-layer linear-to-circular transmission polarizer with a 3 dB axial ratio bandwidth of 34% is presented. The bandwidth of this polarizer is significantly better than that of previously published polarizers of the same type. The unit cell of the polarizer consists of an I-shaped strip and a perpendicular linear strip printed on the one side of a thin dielectric substrate and two additional capacitive coupling strips printed on the other side of the substrate. Experimental results were found to agree well with the simulated results. A thin single-layer reflective linear-to-circular polarizer with a 3 dB axial ratio bandwidth of 57% is also presented. The unit cell of the polarizer consists of an I-shaped strip and a perpendicular linear strip printed on the one side of a substrate and a ground plane on the other side of the substrate. Experimental results for this polarizer were also found to agree well with the simulated results. / Dissertation (MEng)--University of Pretoria, 2018. / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

Outsourcing, consolidation and cost savings of IT services, are increasingly becoming an imperative source of competitive advantage and a great challenge for most local and global businesses. These challenges not only affect consumers, but also the service providers’ community. As IT is slowly becoming commoditised, consumers, such as business organisations, are increasingly expecting IT services that will mimic other utility services such as water, electricity, and telecommunications.To this end, no one model has been able to emulate these utilities in the computing arena.Cloud Computing is the recent computing phenomenon that attempts to be the answer to most business IT requirements. This phenomenon is gaining traction in the IT industry, with a promise of advantages such as cost reduction, elimination of upfront capital outlay, pay per use models, shared infrastructure, and high flexibility allowing users and providers to handle high elasticity of demand. The critical success factor that remains unanswered for most IT organisations and its management is: What is the effect of the communication network factors on Internet based technology such as Cloud Computing, given the emerging market context.This study therefore, investigates the effect of four communication network factors (price, availability, reliability and security) in the adoption of Cloud Computing by IT managers in a South African context, including their propensity to adopt the technology. The study investigates numerous technology adoption theories, in which Technology, Organisation and Environment (TOE) framework is selected due to it having an organisational focus as opposed to an individual focus.Based on the results, this study proposes that Bandwidth (Pricing and Security) should be included into any adoption model that involves services running on the Internet. The study makes an attempt to contribute to the emerging literature of Cloud Computing, Internet in South Africa, in addition to offering organisations considering adoption and Cloud Providers’ significant ideas to consider for Cloud Computing adoption. / Dissertation (MBA)--University of Pretoria, 2012. / Gordon Institute of Business Science (GIBS) / unrestricted