Mismatches between the analog to digital converters (ADCs) in time-interleaved sampling causes spurious signals, limiting the performance of the architecture. This thesis introduces the most commonly modeled mismatches, analyses their e ects and reviews and evaluates state-of-the art compensation methods published in recent papers. A novel method for choosing calibration signal for non-blind compensation is suggested, and how it can be used for current blind compensation methods (relaxing input signal requirements). Application of some common compensation methods is demonstrated with simulated data as well as data captured by a development board with two time-interleaved ADCs sampling at a high rate. / Skillnader mellan analog-till-digital-omvandlarna (AD-omvandlarna) vid tids-interleavad sampling orsakar falska signaler, vilket kan bergransa arkitekturens prestanda. Den har masteruppsatsen introducerar de skillnader som vanligtvis modelleras, analyserar dess e ekter och granskar och utvarderar de framsta metoderna i nyligen publicerade artiklar. En ny metod for val av kalibreringssignal till icke-blinda kompenseringsmetoder foreslas, och hur den kan anvandas for blinda kompenseringsmetoder (vilket mildrar kraven pa insignalen). Nagra kompenseringsmetoder tillampas pa simulerad data saval som data insamlad av ett utvecklingskort med tva tids-interleavade AD-omvandlare med hog sampelfrekvens.

Video conferencing services are dependent on many other underlying devices, network servicesand infrastructure and TCP/IP services before they can provide seamless, reliable and good qualityvideo meeting services to end users. Providing fully automated video conferencing services atSkiptrip AB requires engagement of even more variant and complex set of TCP/IP services anddevices that has made its network a heterogeneous one consisting of hundreds of modern andlegacy systems along with the high definition and bandwidth sensitive video conferencingsystems. In this thesis the process of designing and implementing a secure network module forseparating and transferring non-production (management) network traffic flow of all networkequipment via establishing and fine-tuning virtual IP-sec tunnels among edge routers or firewallsof each video station in this enterprise-scale network has been conducted in order to make surethat the network traffic flow belonging to the management module is treated separately andsecurely thanks to the encryption mechanisms of IPsec protocol on the header and payload of IPpackets.After getting inspired by studying some well-known network design and architecturemethodologies and industry best practices like Cisco SAFE, characterizing the existing network isdone in the early stages of this thesis with a focus on security measures such as the utilization ofAccess Control Lists on different router interfaces which were utilized to provide perimeternetwork security to some extent. Afterwards, a new network design is proposed where themanagement flow is separated from the production traffic flow and is transferred through thesecure IPsec tunnels in a semi-mesh topology which form a virtual network module for themanagement traffic of the whole internetwork. The new network module is then given a new IPaddressing scheme based on the private range of IPv4 addresses and, after relevant discussions, acertain way of implementation of static routing in combination with classless interdomain routingand variable length subnetmasking is introduced to provide, implemented and tested in order toprovide route-redundancy in IP connectivity level of management network module in a similar-todynamicrouting protocol manner.Innate sensitivity of high definition video conferencing protocols like H.323 and SIP to quality ofthe underlying network infrastructure which is usually defined in terms of packet loss and jitter aswell as the bandwidth limitation of costly Internet links in each video station and theheterogeneity of the internetwork were amongst the main technical challenges of this thesis andshaped the outcome of proposed design and also the evaluation mechanisms which are done atthe end of this project.

It is fascinating, and yet often neglected, that a user’s privacy can be invaded notonly by the absence of security measures and mechanisms, but also by improperor inadequate usage of security services and mechanisms. When designingsecure systems, we must consider what services are needed and what is not.The work in this thesis revolves around privacy-friendly instant messaging (IM)systems. In such a system, an inadequate usage of security measures leads tohaving IM servers being able to intercept or gather users’ private conversations.An improper usage of security measures could bring about non-repudiationwhich is desirable when signing contracts, but unwelcome in IM and privateconversations.We will look into requirements of the desired IM system, study the currentstate-of-the-art solutions, deploy an IM server, and briefly extend an existingmodern privacy-friendly IM protocol and an open source mobile application tomeet our security and privacy requirements. This extended IM application iscalled Guidepal-IM and is available as open source1The thesis work is introduced and carried out at Guidepal, a startup companyin Stockholm. It is therefore supervised partly at Guidepal and partly at KTH.Since Guidepal is also looking into possibilities of integrating an IM featureto its current social media apps, our contribution would also briefly extend tostudying the limitations and recommendations for Guidepal’s social media appto help user privacy preservation.

This is a report of a Master’s thesis done in behalf of the local network in Stockholm, under Fortum Distribution AB. The thesis was to conduct a review of a distribution station named Fs Myrängen located in Täby. A model was created for the 11 kV network linked to the secondary side of Fs Myrängen. Based on this model and using symmetrical components, theoretical fault currents (short circuit and ground fault) could be calculated. The results of these fault currents are used to investigate and develop new settings for relay protection and ground equipment in the station. The results of the calculated fault currents were also used to investigate unwanted voltages that occurs for grounded parts due to earth faults. The value of these unwanted voltages was then compared to heavy current regulations developed by the Electrical Safety Authority in Sweden. Another element of the work was to create a rehabilitation plan for the structure on the 11 kV side of Fs Myrängen, the main goal regarding the rehabilitation plan was to investigate the advantages and disadvantages that exist at a possible transition to numerical protection relays. This was done by making a market overview of the different types of numerical protection relays and compare these with the existing analog protective device in Fs Myrängen. The study showed that protective relay settings should be revised according to the recommendations in the report, ground equipment should be replaced with a self-regulating type, transformers should be replaced due to age and capacity limitations. Regarding a switch to numerical protection relays the type REF615 from ABB was recommended, in that case the control board in the station should be replaced to a station computer. Furthermore should current transformers be replaced to recive a desired sensitivity regarding detection of ground faults. The report also showed that Fs Myrängen meets the regulations of unwanted voltages over grounded parts in distribution stations due to earth faults. / Detta är en rapport av ett examensarbete utfört i uppdrag av lokalnät Stockholm under Fortum Distribution AB. Examensarbetet gick ut på att göra en översyn av en fördelningsstation vid namn Fs Myrängen som är belägen i Täby. En modell skapades av 11 kV nätet kopplat till Fs Myrängen, utifrån denna modell samt med hjälp av symmetriska komponenter beräknades teoretiska felströmmar både för kortslutning och jordslutning. Resultaten av dessa felströmmar användes för att utreda och ta fram inställningar på reläskydd och nollpunktsutrustning i stationen. Resultaten på framräknade jordfelsströmmar användes även för att utreda vilken spänningssättning som uppstår vid jordfel, i de nätstationer som matas av Fs Myrängen. Spänningssättningen på nätstationerna jämfördes sedan med de starkströmsföreskrifter som är framtagna av elsäkerhetsverket. Ett annat moment i arbetet var att skapa en upprustningsplan för skyddsanordningen på 11 kV sidan i Fs Myrängen, huvudmålet för den biten var att utreda vilka fördelar och nackdelar som finns vid en eventuell övergång till numeriska skydd. Detta gjordes genom att göra en marknadsöversikt över olika typer av numeriska reläskydd, samt jämföra dessa med den befintliga skyddsanordningen i Myrängen som är av analog typ. Studien visade att reläskyddsinställningar bör ses över enligt rekommendationer i rapporten, nollpunktsutrustningen bör bytas ut till självreglerande typ, transformatorerna bör bytas ut p.g.a. ålder och kapacitetsbegränsning. Gällande skyddsanordningen rekommenderas en övergång till numeriska skydd av typen REF615, och i samband med detta byta ut den befintliga kontrolltavlan i stationen till en stationsdator. Vidare bör även strömtransformatorer för jordfelsskydden bytas ut för att uppnå önskad känslighet gällande detektering av jordfel. För spänningssättning av nätstationer vid jordfel, visar resultaten att stationen uppfyller kraven enligt starkströmsföreskrifterna från elsäkerhetsverket.

The use of the power converters in railway traction systems introduces high frequency electromagnetic interference (EMI) in the propulsion system, which causes electromagnetic compatibility (EMC) problems. These high frequency phenomena come from fast variations of current and voltage during the switching operations in the power converter. The high frequency currents generate Electromagnetic (EM) disturbances that could distort the smooth functionality of the electrical drive system. In fact, power and audio frequency emissions could disturb track signaling and the control systems, while high frequency currents injected into cable screens could damage the cables. In order to ensure compatibility to conducted and radiated EMC requirements, and related infrastructure signaling specications, it is necessary to perform 3D modelling of the drive system to predict the EM emission during the design phase of the propulsion system. CST, an electromagnetic analysis tool, is used to create the 3D model of the converter module and the cables. The model allows for the inclusion of the parasitic characteristics of the IGBTs, the bus-bars, and the motor cables. Inuence of dierent grounding schemes is analyzed. The model predicts the EM eld distribution at points inside the converter module and in the vicinity. / Anvandningen av kraftomvandlare i jarnvagstraktionssystem introducerar hogfrekvens elektromagnetisk interferens (EMI) i framdrivningssystemet, vilket orsakar elektromagnetiska kompatibilitetsproblem (EMC). Dessa hogfrekvensfenomen orsakas av snabba variationer i strom och spanning under omkopplingsoperationer i kraftomvandlare. Hogfrekvensstrommarna alstrar elektromagnetiska (EM) storningar, som kan paverka funktionaliteten hos det elektriska drivsystemet. Storningar vid kraft- och ljudfrekvenser kan paverka signal- och kontrollsystemen, medan hogfrekventa strommar injiceras i kabelskarmar kan skada kablarna. For att sakerstalla kompatibiliteten mellan EMC-kraven, vad galler ledningsbundna och utsanda storningar, och specikationerna for signalsystemets infrastruktur ar det nodvandigt att utfora 3D-modellering av drivsystemet, for att redan under designfasen av framdrivningssystemet kunna forutsaga de elektromagnetiska storningarna. CST, som ar ett elektromagnetiskt analysverktyg, anvands for att skapa 3D-modellen av omriktarmodulen och kablarna. Modellen gor det mojligt att ta med de parasitiska egenskaperna hos IGBT, ledningsmoduler och motorkablar. Inverkan av olika jordningssystemen analyseras. Modellen forutsager det elektromagnetiska faltet vid olika punkter inuti omriktarmodulen och i dess narhet.

This master thesis describes the design simulation of a conformal antenna for the 88-108MHz Very high frequency (VHF) broadcast audio frequency-modulation (FM) band. The antenna is intended to t in the 　at area inside the head-band of an over ear hearing-protector headset. The space for the antenna is limited by an existing head-band design, where the unused internal area is the space studied in this thesis. A genetic algorithm is described for the multiple objective optimization of the antenna matching and radiation pattern optimization. The results of multiple genetic algorithm evaluations are described, and possible further improvements outlined. Progress is made on the development of the antenna. The antenna radiation pattern is evolved in desirable way, but a di-culty in solving the antenna matching problem is identied. Proposals for resolving the antenna matching problem is described in the final section.

Compressive Sensing has attracted a lot of attention over the last decade within the areas of applied mathematics, computer science and electrical engineering because of it suggesting that we can sample a signal under the limit that traditional sampling theory provides. By then using dierent recovery algorithms we are able to, theoretically, recover the complete original signal even though we have taken very few samples to begin with. It has been proven that these recovery algorithms work best on signals that are highly compressible, meaning that the signals can have a sparse representation where the majority of the signal elements are close to zero. In this thesis we implement some of these recovery algorithms and investigate how these perform practically on a real video signal consisting of 300 sequential image frames. The video signal will be under sampled, using compressive sensing, and then recovered using two types of strategies, - One where no time correlation between successive frames is assumed, using the classical greedy algorithm Orthogonal Matching Pursuit (OMP) and a more robust, modied OMP called Predictive Orthogonal Matching Pursuit (PrOMP). - One newly developed algorithm, Dynamic Iterative Pursuit (DIP), which assumes and utilizes time correlation between successive frames. We then performance evaluate and compare these two strategies using the Peak Signal to Noise Ratio (PSNR) as a metric. We also provide visual results. Based on investigation of the data in the video signal, using a simple model for the time correlation and transition probabilities between dierent signal coecients in time, the DIP algorithm showed good recovery performance. The main results showed that DIP performed better and better over time and outperformed the PrOMP up to a maximum of 6 dB gain at half of the original sampling rate but performed slightly below the PrOMP in a smaller part of the video sequence where the correlation in time between successive frames in the original video sequence suddenly became weaker. / Compressive sensing har blivit mer och mer uppmarksammat under det senaste decenniet inom forskningsomraden sasom tillampad matematik, datavetenskap och elektroteknik. En stor anledning till detta ar att dess teori innebar att det blir mojligt att sampla en signal under gransen som traditionell samplingsteori innebar. Genom att sen anvanda olika aterskapningsalgoritmer ar det anda teoretiskt mojligt att aterskapa den ursprungliga signalen. Det har visats sig att dessaaterskapningsalgoritmer funkar bast pa signaler som ar mycket kompressiva, vilket innebar att dessa signaler kan representeras glest i nagon doman dar merparten av signalens koecienter ar nara 0 i varde. I denna uppsats implementeras vissa av dessaaterskapningsalgoritmer och vi undersoker sedan hur dessa presterar i praktiken pa en riktig videosignal bestaende av 300 sekventiella bilder. Videosignalen kommer att undersamplas med compressive sensing och sen aterskapas genom att anvanda 2 typer av strategier, - En dar ingen tidskorrelation mellan successiva bilder i videosignalen antas genom att anvanda klassiska algoritmer sasom Orthogonal Matching Pursuit (OMP) och en mer robust, modierad OMP : Predictive Orthogonal Matching Pursuit (PrOMP). - En nyligen utvecklad algoritm, Dynamic Iterative Pursuit (DIP), som antar och nyttjar en tidskorrelation mellan successiva bilder i videosignalen. Vi utvarderar och jamfor prestandan i dessa tva olika typer av strategier genom att anvanda Peak Signal to Noise Ratio (PSNR) som jamforelseparameter. Vi ger ocksa visuella resultat fran videosekvensen. Baserat pa undersokning av data i videosignalen visade det sig, genom att anvanda enkla modeller, bade for tidskorrelationen och sannolikhetsfunktioner for vilka koecienter som ar aktiva vid varje tidpunkt, att DIP algoritmen visade battre prestanda an de tva andra tidsoberoende algoritmerna under visa tidsekvenser. Framforallt de sekvenser dar videosignalen inneholl starkare korrelation i tid. Som mest presterade DIP upp till 6 dB battre an OMP och PrOMP .

In telecommunication baseband signal processing systems, thousands of tasks are executed every millisecond. These tasks take in dierent parameters and cause heavy load to the system. The aim of the thesis is to build proper mathematical models for these tasks, enabling the prediction of their load given the corresponding parameters. For each task, data samples of task load measure and corresponding parameters are provided. No prior knowledge on the task load and its parameters is available. By studying the data samples, an explicit, accurate and simple model is expected. Graphical skills like scatter plots are used as a preliminary analysis of the data. Then rst-order and second-order linear models, piecewise-linear models and tree-based models are taken as prototypes for the task modeling. Methods like stepwise linear regression and partial correlation analysis are applied to select proper parameters from many available parameters to simplify the models. An automatic tool is further developed to automate the whole modeling process. There are 17 tasks in total. For 15 tasks, acceptable models are built with a RMSE lower than 2 times of the estimated noise standard deviation with the assumption of a Gaussian noise, while for the other 2, no adequate models are given. Reasons for not getting acceptable models are discussed and suggestions on future work are proposed. / I telekommunikationssystem utförs tusentals uppgifter varje millisekund. Dessa uppgifter tar in olika parametrar och orsakar stor belastning på systemet. Syftet med avhandlingen är att bygga riktiga matematiska modeller för dessa uppgifter som gör det möjligt att förutsäga deras last givet motsvarande parametrar. För varje uppgift har datasampel med lastutnyttjande och motsvarande parametrar tillhandahållits. En explicit, exakt och enkel modell önskas. Spridningsdiagram används som en preliminär analys av datat. Sedan används första ordningens och andra ordningens linjära modeller, styckvis-linjära modeller och trädbaserade modeller som prototyper för uppgiftsmodellering. Metoder som styckvis linjär regression och partiell korrelationsanalys tillämpas för att välja rätt parametrar från många tillgängliga parametrar för att förenkla modellerna. Ett automatisk verktyg har utvecklats för att automatisera hela modelleringsprocessen. Det finns 17 uppgifter totalt. För 15 av 17 uppgifter hittades acceptabla modeller byggda med ett RMSE lägre än 2 gånger standardavvikelsen av det uppskattade bruset med antagandet om ett gaussiskt brus. För de andra två uppgifterna hittades inga adekvata modeller. Skäl till att inte få acceptabla modeller diskuteras och förslag på framtida arbete föreslås.

Nowadays wireless networks are used widely in industry due to the advantagesthat they provide. However, these networks are not as reliable as wired networks.This unreliability is the result of problems like delays, packet lossesand limited bandwidth. One should take care of these problems in advance,to get the best performance of the systems.A packet loss, which is the main focus of this thesis, occurs when a packet ofdata gets lost, either from the sensors to the controller or from the controllerto the actuators in the process. There are many methods to deal with packetlosses. In some methods it is tried to improve the connection, to get lesspacket losses. In other methods, a predictive outage compensator (POC) canbe used before the actuator or the controller to compensate for the eects of thepacket losses on the network control system's performance. The POC sends areplacement signal if it does not receive the sent data in the appropriate time.In this thesis, we consider a system with an unreliable wireless network connectionbetween the controller and the actuator. We study the eect of threePOC methods on the performance of this system: Zero (where a signal witha value equal to zero is sent to the system during the time the connection islost), Hold (where the last received signal is used instead of the lost signal)and optimal method. In our proposed method, knowing some parameters ofthe system together with the monitoring of the control history, we estimatethe lost signal. To nd the systems parameters (or to initialize the system), weuse two dierent system identication methods, oine (or batch) and online(or recursive).To model the connection, we consider two packet loss models, outages and theGilbert Elliot model. An outage is dened as a period when several consecutivepackets are lost.  We show that the optimal method improves the performance of the networkedcontrol system remarkably, compared to the other two methods. The conclusionsare veried by simulation results.

Mobile Visual Search (MVS) is a research field which focuses on the recognition of real-world objects by using mobile devices such as smart phones or robots. Current mobile visual search solutions achieve search results based on the appearance of the objects in images captured by mobile devices. It is suitable for planar structured objects such as CD cover images, magazines and art works. However, these solutions fail if different real objects appear similar in the captured images. To solve this problem, the novel solution captures not only the visual appearance of the query object, but uses also the underlying 3D geometry. Vocabulary Tree (VT) methods have been widely used to efficiently find the match for a query in the database with a large volume of data. In this thesis, we study the vocabulary tree in the scenario of multi-view imagery for mobile visual search. We use hierarchically structured multi-view features to construct a multi-view vocabulary trees which represent the 3D geometric information of the objects. Relevant aspects of vocabulary trees such as the shaping of trees, tf-idf weighting and scoring functions have been studied and incorporated in the multi-view scenario. The experimental results show that our multi-view vocabulary trees improve the matching and ranking performance of mobile visual search.