This master thesis deals with measurement and modeling of MEMS inertial sensors. This paper describes basic principles of inertial sensors along with their most often errors. The next part shows results from inertial sensor market analysis, which enabling a selection of sensors to be measured. The following two chapters present methods for inertial sensor modeling and testing. The biggest part of text is dedicated to presentation of measurement results showing us static measurement of Allan variance, Earth rotation, temperature dependent bias and dynamic measurement of gyroscope sensitivity testing over temperature. In the last part of the thesis is presented a design of sensor error model by autocorrelation function and Allan variance and also an evaluation of achieved results.

Speech processing methods were studied to calculate parameters of pacient with Parkinon's disease. Main focus of this work is to examine diadochokinetic (DDK) tests. Algorithm for parameters extraction was proposed. It works in more parts. DC is removed from speech signal, preemphasis aplicated. Envelope of input signal is calculated, peaks of syllables are detected. Parameters and statistical results of Mann-Whitney U~test are calculated from detected peaks. Proposed algorithm is implemented in Matlab.

Orthogonal frequency division multiplexing (OFDM) has become a very popular method for high-data-rate communication. However, it is well known that OFDM is plagued by a large peak-to-average power ratio (PAPR) problem. This high PAPR results in overdesigned power amplifiers, which amongst other things leads to inefficient amplifier usage, which is undesirable. Various methods have been recommended to reduce the PAPR of an OFDM transmission; however, all these methods result in a number of drawbacks. In this thesis, a novel method called offset modulation (OM-OFDM) is proposed to control the PAPR of an OFDM signal. The proposed OM-OFDM method does not result in a number of the drawbacks being experienced by current methods in the field. The theoretical bandwidth occupancy and theoretical bit error rate (BER) expression for an OM-OFDM transmission is derived. A newly applied power performance decision metric is also introduced, which can be utilised throughout the PAPR field, in order to compare various methods. The proposed OM-OFDM method appears to be similar to a well-known constant envelope OFDM (CE-OFDM) transmission. The modulation, structural and performance differences between an OM-OFDM and a CE-OFDM method are discussed. By applying the power performance decision metric, the OM-OFDM method is shown to offer significant performance gains when compared to CE-OFDM and traditional OFDM transmissions. In addition, the OM-OFDM method is able to accurately control the PAPR of a transmission for a targeted BER. By applying the power performance decision metric and complementary cumulative distribution function (CCDF), the proposed OM-OFDM method is shown to offer further performance gains when compared to existing PAPR methods, under frequency selective fading conditions. In this thesis, the OM-OFDM method has been combined with an existing active constellation extended (ACE) PAPR reduction method. To introduce a novel method called offset modulation with active constellation extension (OM-ACE), to control the PAPR of an OFDM signal. The theoretical BER expression for an OM-ACE transmission is presented and validated. Thereafter, by applying the decision metric and CCDF, the OM-ACE method is shown to offer performance improvements when compared to various PAPR methods. The use of OM-OFDM for cognitive radio applications is also investigated. Cognitive radio applications require transmissions that are easily detectable. The detection characteristics of an OM-OFDM and OFDM transmission are studied by using receiver operating characteristic curves. A derivation of a simplified theoretical closed-form expression, which relates the probability of a missed detection to the probability of a false alarm, for an unknown deterministic signal, at various signal-to-noise ratio (SNR) values is derived and validated. Previous expressions have been derived, which relate the probability of a missed detection to the probability of a false alarm. However, they have not been presented in such a generic closed-form expression that can be used for any unknown deterministic signal (for instance OFDM and OM-OFDM). Thereafter, an examination of the spectrum characteristics of an OM-OFDM transmission indicates its attractive detection characteristics. The proposed OM-OFDM method is further shown to operate at a significantly lower SNR value than an OFDM transmission, while still offering better detection characteristics than that of an OFDM transmission under Rician, Rayleigh and frequency selective fading channel conditions. In addition to its attractive PAPR properties, OM-OFDM also offers good detection characteristics for cognitive radio applications. These aspects make OM-OFDM a promising candidate for future deployment. / Thesis (PhD)--University of Pretoria, 2012. / Electrical, Electronic and Computer Engineering / unrestricted

<p>Истраживање старења средства за влажење у реалним условима током 8-часовног радног времена у процесу офсет штампе је омогућило сагледавање критичних фактора деактивације средства за влажење и представља основ за праћење кинетике структурних и физичко-хемијских промена активних компоненти средства за влажење. Такође, истраживање обезбеђује и дефинисање могућих механизама хемијских процеса који условљавају пад активности присутних компоненти средства за влажење током процеса старења.</p> / <p>Istraživanje starenja sredstva za vlaženje u realnim uslovima tokom 8-časovnog radnog vremena u procesu ofset štampe je omogućilo sagledavanje kritičnih faktora deaktivacije sredstva za vlaženje i predstavlja osnov za praćenje kinetike strukturnih i fizičko-hemijskih promena aktivnih komponenti sredstva za vlaženje. Takođe, istraživanje obezbeđuje i definisanje mogućih mehanizama hemijskih procesa koji uslovljavaju pad aktivnosti prisutnih komponenti sredstva za vlaženje tokom procesa starenja.</p> / <p>The investigation of fountain solution aging in real condition during 8-hr<br />working shift of offset printing process allows an overview of the critical<br />factors of fountain solution deactivation and represents the basis for<br />monitoring the structural and physico-chemical changes of active<br />components of fountain solution. Also, the investigation defines the possible<br />mechanism of chemical processes which cause the decreasing of activity of<br />present components in fountain solution during aging process.</p>

Pile groups are often used to provide support for structures. Capping a pile group further adds to the system's resistance due to the passive earth pressure from surrounding backfill. While ultimate passive earth pressure under static loading conditions can be readily calculated using several different theories, the effects of cyclic and dynamic loading on the passive earth pressure response are less understood. Data derived from the full-scale testing of a pile cap system with a densely compacted sand backfill under static, cyclic, and dynamic loadings was analyzed with particular focus on soil pressures measured directly using pressure plates. Based on the testing and analyses, it was observed that under slow, cyclic loading, the backfill stiffness was relatively constant. Under faster, dynamic loading, the observed backfill stiffness decreased in a relatively linear fashion. During cyclic and dynamic loading, the pile cap gradually developed a residual offset from its initial position, accompanied by a reduction in backfill force. While the pile cap and backfill appeared to move integrally during static and cyclic loadings, during dynamic loading the backfill exhibited out-of-phase movement relative to the pile cap. Observed losses in backfill contact force were associated with both cyclic softening and dynamic out-of-phase effects. Force losses due to dynamic loading increased with increasing frequency (which corresponded to larger displacements). Losses due to dynamic loading were offset somewhat by increases in peak force due to damping. The increase in contact force due to damping was observed to be relatively proportional to increasing frequency. When quantifying passive earth forces with cyclic/dynamic losses without damping, the Mononobe-Okabe (M-O) equation with a 0.75 or 0.8 multiplier applied to the peak ground acceleration can be used to obtain a reasonable estimate of the force. When including increases in resistance due to damping, a 0.6 multiplier can similarly be used.

In the field of robotics and self-navigation, Simultaneous Localization and Mapping (SLAM) is a technique crucial for estimating poses while concurrently creating a map of the environment. Robotics applications often rely on various sensors for pose estimation, including cameras, inertial measurement units (IMUs), and more. Traditional discrete SLAM, utilizing stereo camera pairs and inertial measurement units, faces challenges such as time offsets between sensors. A solution to this issue is the utilization of continuous-time models for pose estimation. This thesis delves into the exploration and implementation of a continuous-time SLAM system, investigating the advantages of multi-modal sensor fusion over discrete stereo vision models. The findings indicate that incorporating an IMU into the system enhances pose estimation, providing greater robustness and accuracy compared to relying solely on visual SLAM. Furthermore, leveraging the continuous model's derivative and smoothness allows for decent pose estimation with fewer measurements, reducing the required quantity of measurements and computational resources.

Welds play a crucial role in the product portfolio of GKN Aerospace. For ease of manufacturing and repairability, butt welds are preferred in the aerospace industry. Weld regions undergo local distortions during manufacturing due to the external heat-input, and the resulting local geometries are complex and stochastic in nature. Under operational loads, the distorted geometry affects the local stress field around the weld region, and this has a significant impact on the fatigue life. Traditional design calculations of welds resort to idealization of the local weld geometry. In this thesis, the influence of real weld geometry on the computed fatigue life is investigated. Linear elastic fracture mechanics principles are utilized to calculate the fatigue life of a weld starting from a pre-defined initial crack. The influence of important weld geometric parameters, namely, – edge offset and weld toe radii, on the fatigue life is investigated in detail. A statistical analysis approach, using transfer functions and Monte Carlo simulation, is devised to study the effect of variation in the different weld geometric parameters. Different edge offset measures from a real geometry are identified and these measures are investigated as potential candidates to obtain conservative life estimates using the idealized geometry. Investigations in this thesis show that the real geometry has a significant effect on the weld fatigue life. In all cases of local weld geometry, the edge offset has the largest influence on life. For a real weld geometry, the root toe radius has a significant influence on life. In an idealized geometry, for the same normalized edge offset, an increase in the plate thickness leads to a decrease in life. A new edge offset measure is proposed that leads to conservative life estimates when used in conjunction with the idealized geometry, thus enabling computationally efficient design calculations. / Svetsar spelar en avgörande roll i GKN Aerospace produktportfölj. För att underlätta vid tillverkning och reparationer föredras stumsvetsar inom flygindustrin. Svetsområden genomgår lokala deformationer under tillverkningen på grund av extern värmepåverkan, och de resulterande lokala geometrierna är komplexa och stokastiska till sin natur. Vid termisk och mekanisk belastning i drift påverkar den deformerade geometrin det lokala spänningsfältet runt svetsområdet vilket har en betydande inverkan på utmattningslivslängden. Vid traditionell dimensionering av svetsar idealiseras den lokala svetsgeometrin. Denna avhandling undersöker inverkan av verklig svetsgeometri på den beräknade utmattningslivslängden. Linjärelastisk brottmekanik används för att beräkna utmattningslivslängden för en svets med utgångspunkt från en fördefinierad initial spricka. Inverkan av viktiga svetsgeometriska parametrar, nämligen - kantförskjutning och svetsradie, på utmattningslivslängden undersöks i detalj. En statistisk analysmetod, med hjälp av överföringsfunktioner och Monte Carlo simulering, är framtagen för att studera effekten av variation i de olika svetsgeometriska parametrarna. Olika kantförskjutningsmått från en verklig geometri identifieras och dessa mått undersöks som potentiella kandidater för att erhålla konservativa livslängdsuppskattningar med hjälp av den idealiserade geometrin. Undersökningar i denna avhandling visar att den verkliga geometrin har en signifikant effekt på svetsutmattningslivslängden. I alla fall av lokal svetsgeometri har kantförskjutningen den största inverkan på livslängden. För en riktig svetsgeometri har svetsradien vid svetsens rotsida en betydande inverkan på livet. I en idealiserad geometri, för samma normaliserade kantförskjutning, leder en ökning av plåttjockleken till en minskning av livslängden. Ett nytt kantförskjutningsmått föreslås som leder till konservativa livslängdsuppskattningar när de används i kombination med den idealiserade geometrin, vilket möjliggör beräkningseffektiva designberäkningar.

The human influence of global climate is an issue currently assessed in various mitigation strategies. Stockholm County has committed to becoming carbon neutral by 2040 and negative by 2045 according to the Paris agreement. The strategy includes cutting of various sectors emissions as well as compensating for remaining emissions with carbon sequestration methods. Accounting for ecosystems ability to sequester carbon at local level in green urban areas is an important in carbon offset efforts. It has emerged from previous research that the sequestration rate may differ depending on vegetation age and thus time passing, which is not assessed on a regional level, which is important for carbon offset efforts to accurately account for the sequestration potential in long-term mitigation strategies. This study therefore aims to fill the knowledge gap of how the temporal aspect affects the current sequestration potential and future predictions, as well as assessing how it can aid in reaching carbon neutrality by 2040. The study are thus aiming to answer the research questions (1) how the carbon sequestration potential of the existing green areas change over time in Stockholm County, (2) if additional measures need to be taken to preserve or increase carbon sequestration to maintain carbon neutrality until 2040 and (3) how the knowledge of sequestration dynamics aid in reaching a carbon neutral city by 2040. A weighing of which Corine Land Cover categories was performed and concluded in the including; discontinuous structures, green urban areas, forests and wetlands, due to their contribution to sequestration potential, estimated change through time, and relevance for Stockholm County. The spatial analysis was made based on calculations with information obtained from processing of obtained data on land cover and species distribution as well as scientific literature on sequestration rates of each vegetation across all life stages, where Net Ecosystem Production was the main measurement used. The estimated results were computed in a Geographic Information System to simulate and visualize the sequestration rates of current and future predictions of 2040 sequestration potential as well as locating areas of interest. The findings show that by including temporal aspects to the assessment of carbon sequestration potential in Stockholm County, the current and future sequestration potential increased from previous research estimations. The total current sequestration potential was 2,8 MtCO2-eq annually and the predictions were estimated to 3,3 MtCO2-eq per year in 2040. As the current emissions in Stockholm County are currently 6 MtCO2-eq per year, the natural sequestration potential provided by the green areas is compensating for 46% of the current emissions. As the estimated future emissions are 0,95 MtCO2-eq annually, the natural sequestration potential more than compensates for the emissions in the county, if the predicted emission reductions are realized. Although further measurements are not seemingly required to achieve carbon neutrality in 2040, the findings further locate areas and species where management practices or protection is beneficial to further add to the sequestration potential of Stockholm County. / Den mänskliga påverkan på det globala klimatet är ett problem som för närvarande bedöms i olika klimatåtgärder. Stockholms län har förbundit sig till ett mål att bli koldioxidneutralt till 2040 och koldioxidnegativt till 2045 enligt Parisavtalets överenskommelser. Strategin innefattar att minska utsläppen från olika sektorer samt att kompensera de återstående utsläppen med olika metoder för koldioxidlagring. Ekosystemens förmåga att binda kol på lokal nivå i gröna stadsområden är en viktig del av ansträngningarna för att kompensera för utsläppen. Det har varit uppenbart från tidigare forskning att potentialen av koldioxidlagring kan skilja sig åt beroende på vegetationens ålder samt passerande tid, även om aktuell forskning inte omfattar frågan på regional nivå, vilket är viktigt för insatser som omfattar koldioxidkompensation så att lagringspotentialen kan redovisas korrekt i de långsiktiga klimatåtgärderna. Denna studie syftar därför till att fylla kunskapsluckan gällande hur den tidsmässiga aspekten påverkar den befintliga lagringspotentialen och framtida prognoser samt hur det kan bidra till att nå koldioxidneutralitet fram till 2040. Studien avser därmed till att svara på forskningsfrågorna (1) hur kolbindningspotentialen för de befintliga grönområdena förändras över tid i Stockholms län, (2) ifall ytterligare åtgärder behöver vidtas för att bevara eller öka koldioxidbindningen för att uppnå eller bibehålla koldioxidneutralitet fram till 2040 och (3) hur kunskapen om koldioxidlagringsdynamiken underlättar för att nå en koldioxidneutral region år 2040. En avvägning av vilka Corina marktäckeskategorier utfördes och resulterade i inkludering av; diskontinuerliga strukturer, gröna stadsområden, skogar och våtmarker, där koldynamiken och omfattningen av lagringspotentialen var relevant för Stockholmsregionen. Den rumsliga analysen gjordes baserad på beräkningar med information erhållen genom bearbetning av införskaffad data om marktäcke och artfördelning samt vetenskaplig litteratur om kolbindningshastighet för varje vegetation över alla livsstadier, där Net Ecosystem Production var det huvudsakliga måttet. De uppskattade resultaten beräknades i ett Geografiskt Informationssystem för att simulera och visualisera lagringshastigheten för nuvarande och framtida förutsägelser om sekvestreringspotentialen år 2040 samt att lokalisera intressanta. Resultaten visar att genom att inkludera tidsmässiga aspekter i bedömningen av kolbindningspotentialen i Stockholms län ökade den nuvarande och framtida bindningspotentialen från tidigare forsknings uppskattningar. Den totala nuvarande lagringspotentialen var 2,8 MtCO2-ekv årligen och de framtida prognoserna uppskattades till 3,3 MtCO2-ekv årligen år 2040. Eftersom de nuvarande utsläppen i Stockholms län för närvarande är 6 MtCO2-ekv årligen, kunde man se att den naturliga lagringspotentialen som grönområdena avsåg, kompenserar för 46 % av de nuvarande utsläppen. Eftersom de beräknade framtida utsläppen är 0,95 MtCO2-ekv per år, mer än kompenserar de gröna områdena för de utsläpp som sker i länet, om den förutsedda reduceringen av utsläppen sker. Även om ytterligare mätningar inte tycks behövas för att uppnå koldioxidneutralitet till 2040, lokaliseras ytterligare områden samt arter i resultatet där förvaltning eller skydd är fördelaktiga för att ytterligare förbättra lagringspotentialen i Stockholms län.

<p dir="ltr">Optical frequency combs have revolutionized metrology and advanced other fields such as RF photonics and astronomy. While powerful, they can be bulky, expensive, and difficult to manufacture. This tends to limit uses in real-world scenarios. Within the last decade or so, coherent frequency combs have begun to be generated in millimeter-scale, CMOS fabrication-compatible nonlinear crystals. These so-called “microcombs” have led to hopes of overcoming deployability constraints of more traditional bulk combs.</p><p dir="ltr">One of the first applications for \textit{bulk} frequency combs after their explosion in 2000 was the optical atomic clock. It promised extreme long-term time stability better than that of the Cesium clock that currently defines the SI second. More recently, interest in a fully portable optical atomic clock has grown. Such a device could reliably keep time even without the aid of GPS references, and potentially with greater accuracy than current GPS synchronization can provide.</p><p dir="ltr">Frequency combs have also been used to sample electrical signals more rapidly than traditional electronics can accomplish. This has been used to achieve dramatically increased effective frequency bandwidths for signal detection architectures. One can imagine how this capability would be beneficial in a portable (microcomb-driven) form: a lightweight, comb-enhanced receiver able to capture a broadband snapshot of its surrounding electromagnetic environment could be a powerful tool.</p><p dir="ltr">Timekeeping and RF photonics are the primary applications of microcombs focused upon here. I will attempt to roughly summarize important concepts and highlight relevant work in both subjects in the Introduction. Then I will move a step closer to the hands-on lab work that has largely kept me preoccupied over the last several years and describe important or commonly-employed Methods for experiments. A collection of three journal manuscripts (two published, and the third recently submitted) will follow in the Publications chapter, highlighting some experimental results. Finally, I will conclude with a brief Outlook.</p>

A wide range of services and applications become possible when accurate position information for a radio terminal is available. These include: location-based services; navigation; safety and security applications. The commercial, industrial and military value of radio-location is such that considerable research effort has been directed towards developing related technologies, using satellite, cellular or local area network infrastructures or stand-alone equipment. This work studies and investigates two location techniques. The first one presents an implementation scheme for a wideband transmission and direction finding system using OFDM multi-carrier communications systems. This approach takes advantage of delay discrimination to improve angle-of-arrival estimation in a multipath channel with high levels of additive white Gaussian noise. A new methodology is interpreted over the multi carrier modulation scheme in which the simulation results of the estimated channel improves the performance of OFDM signal by mitigating the effect of frequency offset synchronization to give error-free data at the receiver, good angle of arrival accuracy and improved SNR performance. The full system simulation to explore optimum values such as channel estimation and AoA including the antenna array model and prove the operational performance of the OFDM system as implemented in MATLAB. The second technique proposes a low cost-effective method of tracking and monitoring objects (examples: patient, device, medicine, document) by employing passive radio frequency identification (RFID) systems. A multi-tag, (totalling fifty-six tags) with known ID values are attached to the whole patient’s body to achieve better tracking and monitoring precision and higher accuracy. Several tests with different positions and movements are implemented on six patients. The aim is to be able to track the patient if he/she is walking or sitting; therefore, the tests considered six possible movements for the patient including walking, standing, sitting, resting, laying on the floor and laying on the bed, these placements are important to monitor the status of the patient like if he collapsed and fall on the ground so that the help will be quick. The collected data from the RFID Reader in terms of Time Stamp, RSS, Tag ID, and a number of channels are processed using the MATLAB code.