We present Spitzer infrared, Galaxy Evolution Explorer UV, and Sloan Digitized Sky Survey and Southeastern Association for Research in Astronomy optical images of the peculiar interacting galaxy pair Arp 285 (NGC 2856/4), and compare with a new numerical model of the interaction. We estimate the ages of clumps of star formation in these galaxies using population synthesis models, carefully considering the uncertainties on these ages. This system contains a striking example of "beads on a string": a series of star-formation complexes 1 kpc apart. These "beads" are found in a tail-like feature that is perpendicular to the disk of NGC 2856, which implies that it was formed from material accreted from the companion NGC 2854. The extreme blueness of the optical/UV colors and redness of the mid-infrared colors implies very young stellar ages (4-20 Myr) for these star-forming regions. Spectral decomposition of these "beads" shows excess emission above the modeled stellar continuum in the 3.6 μm and 4.5 μm bands, indicating either contributions from interstellar matter to these fluxes or a second older stellar population. These clumps have -12.0 < M B< -10.6, thus they are less luminous than most dwarf galaxies. Our model suggests that bridge material falling into the potential of the companion overshoots the companion. The gas then piles up at apogalacticon before falling back onto the companion, and star formation occurs in the pile-up. There was a time delay of 500 Myr between the point of closest approach between the two galaxies and the initiation of star formation in this feature. A luminous (M B -13.6) extended (FWHM 1.3 kpc) "bright spot" is visible at the northwestern edge of the NGC 2856 disk, with an intermediate stellar population (400-1500 Myr). Our model suggests that this feature is part of a expanding ripple-like "arc" created by an off-center ring-galaxy-like collision between the two disks.

We present Spitzer infrared imaging of the peculiar galaxy pair Arp 107 and compare with an optical Hα map and a numerical model of the interaction. The [3.6] - [4.5] colors of clumps in the galaxy do not vary around the ringlike primary spiral arm and are consistent with those of stars; thus, these bands are dominated by starlight. In contrast, the [5.8 μm] - [8.0 μm] colors are consistent with those of interstellar dust and vary by about 0.2 mag around the ring/spiral, with redder colors associated with regions with stronger star formation as indicated by Ha and mid-infrared luminosity. The [4.5 μm] - [5.8 μm] colors for clumps in this arm are bluer than dust and redder than stars and vary by 1.3 mag around the arm. This color is therefore a measure of the relative number of young stars to old stars, with a redder color indicating a higher proportion of young stars. There is a clear azimuthal sequence in the [4.5] - [5.8] color around the arm, indicating a sequence in average stellar age. The L HQ/L 8.0 μm ratio varies around the arm by a factor of ≈7; this variation may be due to extinction or to polycyclic aromatic hydrocarbon excitation by nonionizing photons. Our model of Arp 107 accounts for the general morphology of the system and explains the age variation along the arm as the result of differences in the time of maximum compression in the arm. Using Spitzer colors, we are able to distinguish background quasars and foreground stars from star-forming regions associated with Arp 107.

Les systèmes de téléphonie mobile de 4ème et 5ème générations ont adopté comme techniques de codage de canal les turbocodes, les codes LDPC et les codes polaires binaires. Cependant, ces codes ne permettent pas de répondre aux exigences, en termes d’efficacité spectrale et de fiabilité, pour les réseaux de communications futurs (2030 et au-delà), qui devront supporter de nouvelles applications telles que les communications holographiques, les véhicules autonomes, l’internet tactile … Un premier pas a été fait il y a quelques années vers la définition de codes correcteurs d’erreurs plus puissants avec l’étude de codes LDPC non binaires, qui ont montré une meilleure performance que leurs équivalents binaires pour de petites tailles de code et/ou lorsqu'ils sont utilisés sur des canaux non binaires. En contrepartie, les codes LDPC non binaires présentent une complexité de décodage plus importante que leur équivalent binaire. Des études similaires ont commencé à émerger du côté des turbocodes. Tout comme pour leurs homologues LDPC, les turbocodes non binaires présentent d’excellentes performances pour de petites tailles de blocs. Du point de vue du décodage, les turbocodes non binaires sont confrontés au même problème d’augmentation de la complexité de traitement que les codes LDPC non binaire. Dans cette thèse nous avons proposé une nouvelle structure de turbocodes non binaires en optimisant les différents blocs qui la constituent. Nous avons réduit la complexité de ces codes grâce à la définition d’un algorithme de décodage simplifié. Les codes obtenus ont montré des performances intéressantes en comparaison avec les codes correcteur d’erreur de la littérature. / Nowadays communication standards have adopted different binary forward error correction codes. Turbo codes were adopted for the long term evolution standard, while binary LDPC codes were standardized for the fifth generation of mobile communication (5G) along side with the polar codes. Meanwhile, the focus of the communication community is shifted towards the requirement of beyond 5G standards. Networks for the year 2030 and beyond are expected to support novel forward-looking scenarios, such as holographic communications, autonomous vehicles, massive machine-type communications, tactile Internet… To respond to the expected requirements of new communication systems, non-binary LDPC codes were defined, and they are shown to achieve better error correcting performance than the binary LDPC codes. This performance gain was followed by a high decoding complexity, depending on the field order.Similar studies emerged in the context of turbo codes, where the non-binary turbo codes were defined, and have shown promising error correcting performance, while imposing high complexity. The aim of this thesis is to propose a new low-complex structure of non-binary turbocodes. The constituent blocks of this structure were optimized in this work, and a new low complexity decoding algorithm was proposed targeting a future hardware implementation. The obtained results are promising, where the proposed codes are shown to outperform existing binary and non-binary codes from the literature.

Syftet med uppsatsen är att undersöka varför få svenska kommunerna valt att installera DNSSEC på sina domäner. DNS är en av de viktigaste protokollen på Internet och behövs för att sammanlänka IP-adresser med mer lättförståeliga adresser för oss människor. DNS skapades utan att tänka på säkerheten, för att kunna göra DNS säkrare utvecklades ett säkerhetstillägg till DNS detta fick namnet DNSSEC.Vi har använt oss av litteraturstudie, experiment och intervjuer för att skapa en djupare kunskap och förståelse om hur DNS och DNSSEC fungerar samt besvara varför få kommuner har valt att installera DNSSEC.Under vår litteraturstudie läste vi om flera sårbarheter i DNS och hur dessa kan utnyttjas för att utsätta en organisation för attacker såsom cacheförgiftning och MITM. Vi testade dessa sårbarheter och bekräftade det. Efter installationen av DNSSEC kunde inte angreppen längre genomföras i vår testmiljö.Under intervjuerna kom vi fram till att den vanligaste orsaken att kommuner inte väljer att installera DNSSEC är okunskap om tillvägagångsättet för en installation och att de tycker deras nuvarande DNS fungerar bra, det blir då ingen prioriterad fråga. Kommunerna som installerat DNSSEC är nöjda med sin installation och bara en kommun har upplevt problem vid införandet.För att vi ska kunna fortsätta utveckla Internet är en kontroll av säkerheten en nödvändighet och då är DNSSEC en vägvisare. Kommunerna borde föregå med gott exempel och vara bland de första som inför DNSSEC så besökarna till deras hemsidor kan känna sig säkra att informationen på deras sidor är korrekt. / The purpose of this paper is to investigate why few Swedish municipalities have chosen to install DNSSEC on their domains. DNS is one of the most important protocols on the Internet and used to link IP-addresses to understandable addresses for users. DNS was created without thinking about security, to make DNS more secure a security extension was developed to DNS, named DNSSEC.We have used literature review, experiments and interviews to create a deeper knowledge and understanding about DNS and DNSSEC, how it works and why few municipalities have chosen to install DNSSEC.In the literature we read about several vulnerabilities in DNS and it can easily be exposed to attacks such as cache poisoning and MITM. We tested these vulnerabilities and confirmed them. After installation of DNSSEC we could not expose our implemented DNS anymore in our test environment.During the interviews, we concluded that the most common reason why municipalities do not choose to install DNSSEC is ignorance of an installation and they think that their current DNS works well and it does not become a priority. The municipalities that have installed DNSSEC are satisfied with its installation and only one municipality has experienced difficulties during the implementation.In order for us to continue developing the Internet a control of security is a necessity and DNSSEC is a good example. Local authorities should lead by good example and be among the first to implement DNSSEC, so users of their websites can be assured that the information on their pages is accurate.

This MSc Thesis focuses on principles, practical performability and security against four attacks used in contemporary local-area networks: CAM Table Overflow capable of capturing traffic in switched networks, ARP Man-in-the-Middle, whose target is to redirect or modify traffic and against two variants of VLAN Hopping attack allowing a hacker to send and capture data from VLANs he has no access to.

Leaf primordia initiation takes place at the flanks of SAM and then passes through common developmental stages. Very different final leaf shapes and sizes result from varying the timing and further patterning events within these developmental stages. Similar final leaf shapes may also result from very distinct early events. Begonia section Gireoudia is a recently radiated group of species with highly divergent leaf forms. I have used a classical genetic approach and candidate gene approach to explain the evolution of leaf form in this genus. These results suggest that convergent evolution of peltate leaves may be through changes at different loci. Key developmental regulators KNOX and ARP genes are reported to be involved in the evolution of leaf form in different species. I have shown that in at least one species ARP is linked to the evolution of peltate leaf form. In a second species there is no link between STM-like KNOX genes and leaf dissection. Estimates of the rate of evolution of ARP CDS showed that different domains of the genes are under different selection pressures. Myb domain2 of ARP genes is under positive selection and variable between two copies of ARP genes in Begonia. Results of complementation tests with Begonia ARP genes in Arabidopsis show that ARPs from Begonia are functionally equivalent to Arabidopsis AS1 genes and one of the two ARP genes in Begonia may be a dominant negative. Expression analysis based on insitu hybridization in compound, peltate and simple leaved Begonias is described. There is no variation in expression patterns between peltate, non peltate or compound leaved Begonia species for BARP1 and KNB1 genes.

<p>The use of microcontrollers and FPGAs is getting more and more wide spread in electronic designs. A recent developmenthas been to implement microcontrollers onboard the FPGA, there are a lot of benefits but also disadvantages with this. Often the microcontroler requires a lot of resources in the expensive FPGA. This is where PicoBlaze, a microcontroller provided by Xilinx, fits in. It is designed with one main object, keep it as small and powerful as possible. </p><p>In this report PicoBlaze is evaluated and documented. Two implementations have been done. One smaller to show how to use PicoBlaze and one larger implementation of an Ethernet network interface. The function of the implementations have been verified on a experiment board utilizing a Virtex-II FPGA. </p><p>The conclusion is that PicoBlaze is a very powerful microcontroller in comparison to the resources it uses on the FPGA. It uses only a little more than 80 slices on a Virtex II FPGA. This is its main advantage, the disadvantages of PicoBlaze is its limited program memory and the limited address space.</p>

The use of microcontrollers and FPGAs is getting more and more wide spread in electronic designs. A recent developmenthas been to implement microcontrollers onboard the FPGA, there are a lot of benefits but also disadvantages with this. Often the microcontroler requires a lot of resources in the expensive FPGA. This is where PicoBlaze, a microcontroller provided by Xilinx, fits in. It is designed with one main object, keep it as small and powerful as possible. In this report PicoBlaze is evaluated and documented. Two implementations have been done. One smaller to show how to use PicoBlaze and one larger implementation of an Ethernet network interface. The function of the implementations have been verified on a experiment board utilizing a Virtex-II FPGA. The conclusion is that PicoBlaze is a very powerful microcontroller in comparison to the resources it uses on the FPGA. It uses only a little more than 80 slices on a Virtex II FPGA. This is its main advantage, the disadvantages of PicoBlaze is its limited program memory and the limited address space.

Die vorliegende Arbeit entwickelt und implementiert ein System für die Bereitstellung von Internet-Diensten an öffentlichen Orten. Besonderer Wert wurde auf eine einfache Handhabung für den Nutzer, eine breite Unterstützung von Klienten und ein breites Spektrum möglicher Privilegien (von der vollen Internetkonnektivität bis zu einer Art Infoterminalstatus) gelegt.

Diese Diplomarbeit beschreibt eine Software zur dreidimensionalen Visualisierung der Vorgänge in Computernetzwerken. Dabei werden die Protokolle TCP, UDP, IP, ARP und Ethernet grafisch veranschaulicht. Das Programm stellt außerdem Routing und NAT (Network Address Translation) dar. Zusätzlich gibt es drei von der Software erstellte Videos (je 2 - 15 MB).