• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 92
  • 12
  • 9
  • 9
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • Tagged with
  • 162
  • 162
  • 105
  • 43
  • 42
  • 32
  • 29
  • 21
  • 21
  • 20
  • 19
  • 19
  • 18
  • 18
  • 17
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
161

Crest Factor Reduction using High Level Synthesis

Mahmood, Hassan January 2017 (has links)
Modern wireless mobile communication technology has made noticeable improvements from the technologies in the past but is still plagued by poor power efficiency of power amplifiers found in today’s base stations. One of the factors that affect the power efficiency adversely comes from modern modulation techniques like orthogonal frequency division multiplexing which result in signals with high peak to average power ratio, also known as the crest factor. Crest factor reduction algorithms are used to solve this problem. However, the dominant method of hardware description for synthesis has been to start with writing register transfer level code which gives a very fixed implementation that may not be the optimal solution. This thesis project is focused on developing a peak cancellation crest factor reduction system, using a high-level language as the system design language, and synthesizing it using high-level synthesis. The aim is to find out if highlevel synthesis design methodology can yield increased productivity and improved quality of results for such designs as compared to the design methodology that requires the system to be implemented at the register transfer level. Design space exploration is performed to find an optimal design with respect to area. Finally, a few parameters are presented to measure the performance of the system, which helps in tuning it. The results of design space exploration helped in choosing the best possible implementation out of four different configurations. The final implementation that resulted from high-level synthesis had an area comparable to the previous register transfer level implementation. It was also concluded that, for this design, the high-level synthesis design methodology increased productivity and decreased design time. / Användning av högnivåsyntes för reduktion av toppfaktor Det har gjorts noterbara framsteg inom modern trådlös kommunikationsteknik för mobiltelefoni, men tekniken plågas fortfarande av dålig energieffektivitet hos förstärkarna i dagens basstationer. En faktor som påverkar energieffektiviteten negativt är om signaler har en stor skillnad mellan maximal effekt och medeleffekt. Kvoten mellan maximal effekt och medeleffekt kallas för toppfaktor, och en egenskap hos moderna moduleringstekniker, såsom ortogonal frekvensdelningsmodulering, är att de har en hög toppfaktor. Algoritmer för reducering av toppfaktor kan lösa det problemet. Den dominerande metoden för design av hårdvara är att skriva kod i ett hårdvarubeskrivande språk med abstraktionsnivån Register Transfer Level och sedan använda verktyg för att syntetisera hårdvara från koden. Resultatet är en specifik implementation som inte nödvändigtvis är den optimala lösningen. Det här examensarbetet är inriktat på att utveckla ett system för reducering av toppfaktor, baserat på algoritmen Peak Cancellation, genom att skriva kod i ett högnivåspråk och använda verktyg för högnivåsyntes för att syntetisera designen. Syftet är att ta reda på om högnivåsyntes som designmetod kan ge ökad produktivitet och ökad kvalitet, för den här typen av design, jämfört med den klassiska designmetoden med abstraktionsnivån Register Transfer Level. Verktyget för högnivåsyntes användes för att på ett effektivt sätt undersöka olika designalternativ för att optimera kretsytan. I rapporten presenteras ett antal parametrar för att mäta prestandan hos systemet, vilket ger information som kan användas för finjustering. Resultatet av undersökningen av designalternativ gjorde det möjligt att välja den bästa implementationen bland fyra olika konfigurationer. Den slutgiltiga implementationen hade en kretsyta som är jämförbar med en tidigare design som implementerats med hårdvarubeskrivande språk med abstraktionsnivån Register Transfer Level. En annan slutsats är att, för den här designen, så gav designmetoden med högnivåsyntes ökad produktivitet och minskad designtid.
162

Model-Based Early Validation and Verification of Design Decisions for Cross-Disciplinary Stakeholders

Stenlund, Alexander January 2024 (has links)
Systems engineering becomes more challenging as system engineers must tackle increasingly growing and complex systems while balancing stakeholder needs, dependability, costs, and much more. Engineering decisions at early stages become important as their impact can affect the product's entire life cycle. Model-Based System Engineering (MBSE) is a way to capture this system complexity and explore how different engineering decisions affect a system. With this in mind, this thesis will explore how MBSE can capture engineering decisions and their impact at early stages of development and what insight can be gained into a system.  This is done by exploratively creating a concept, using it on a toy example to get direct feedback on its performance, and refining the concept before evaluating it in a case study. The case study evaluates the concept in an industrial context through a focus group and receives direct industry feedback. The modelling concept is implemented in Papyrus Eclipse as a Unified Modeling Language (UML) profile to capture concept-specific details and an Eclipse plugin. This implementation is to retrieve and process data from the modelled system. This is used to answer impact questions of choices, such as: "How much does this parameter change if component X is exchanged for another?".The evaluation of the concept was generally positive and gave valuable feedback and possible future directions of the concept. This includes the opinion that a spread sheet gives a better overview of the system and ways to expand the tool though parameterised relations and other suggestions.  The thesis in the end presents, one way that MBSE can be used to support the analysis of early cross-disciplinary models, and two major insights which could be gained, namely risk and constraints.

Page generated in 0.076 seconds