<p>Offline Web applications are increasingly popular. The possibility to have both the advantages of Web applications and traditional desktop applications is exiting. An offline Web application can be accessed from all computers, with any operating system, as well as offering to store information locally, giving the user the opportunity to use the application when the user does not have Internet access. The concept of offline Web applications is tempting, but it is important to integrate security in the process of making them. The users rely on a high level of security. In this thesis I have looked specifically on how the persistent client-side storage needed for offline storage for the offline Web application can be compromised due to security vulnerabilities on the Web server. I have performed a literature review to gather information on the topic of security in offline Web applications, and it was found that there has not been much previous research in this area. Two technologies for realization of offline Web applications were reviewed: HTML5 and Google Gears. Following, a Web server was set up, and two test applications with offline capabilities, representing the two chosen technologies, were put on the Web server. A set of security tests were performed on these test applications to reveal possible vulnerabilities in having persistent client-side storage. The results of the security testing demonstrate the consequences of having security weaknesses in Web servers hosting offline Web applications. If there is one cross-site scripting vulnerability on the Web server, an attacker can attack the persistent client-side storage: steal, change, delete or add information related to the offline Web application. Some thoughts on possible consequences of attacks on the hosting Web server are also given. A comparison between Google Gears and HTML5 was performed, and it was found that some of the design choices in Google Gears help provide a higher level of security in offline Web applications. Some strategies for testing the security of offline Web applications are suggested, focused on cross-site scripting vulnerabilities. The work in this thesis underlines the importance of including security in the process of developing and deploying offline Web applications. It shows the large consequences that can result from small security vulnerabilities present in the hosting Web server. Introductorily, the advantages of offline Web applications were discussed. The work presented here shows that the increasing use of offline Web applications relies on a high focus on security in order to keep the users' information safe.</p>

<p>In this paper we present the first efficient point counting algorithm due to Schoof, before giving a significant improvement due to Elkies. In the final section I give Satoh's algorithm which is even faster and has paved the way for the field of $p$-adic point counting.</p>

<p>The Cross-Entropy Ant System (CEAS) has been implemented in ns-3. The implementation has been validated against the latest ns-2 implementation and some extra simulations have been performed. Apart from that, experiences with ns-3 have been described as part of the thesis.</p>

<p>This thesis describes an electronic voting system based on Damgaa rd, Jurik and Nielsen's generalization of Paillier's probabilistic public key system. A threshold variant of this homomorphic cryptosystem is used to provide universally verifiable elections, where zero-knowledge proofs are used for proving correctness of votes. Using this cryptosystem, an electronic voting system that supports voting for 1 out of $L$ candidates is described. Two types of encoding may be used to prove the validity of the votes. The number of proofs needed using normal encoding is linear in $L$, while the number of proofs needed using binary encoding is logarithmic in $L$. It is shown how to extend the system to allow casting a vote for $t$ out of $L$ candidates. This method may easily be used to carry out elections with weighted votes without any added complexity to the system. The system is shown to satisfy the requirements for elections at The Norwegian University of Science and Technology (NTNU). A fully functional implementation of the electronic voting system as a distributed system, using Java Remote Method Invocation, is presented. The implementation is used to analyze the feasibility of using this voting system for future elections at NTNU. The implementation is tested using various keylengths and various election parameters. With a keylength of 1024 bits, the simulated time for verification of complex elections is small enough to be considered universally verifiable.</p>

<p>Pervasive environments are characterized by ubiquitous, mobile and embedded computing devices and wireless networking. The vision is an environment where the technology resides in the background ready to be used when it is needed. An essential part of such environments is context aware applications and context information. A context aware system exploits context information to provide relevant services or information to an entity, where relevancy depends on the entity’s task. As a user, the employment of such systems involves revealing a lot of personal data. Context information can divulge a lot of sensitive information which represents a threat to a person’s privacy. This master thesis looks into privacy handling in pervasive computing environments. The object is to propose a solution on how a user can control the extent of access to his or her context information. In order to identify the most important privacy concerns in the implementation of a context management system, privacy principles are looked into and privacy challenges in consequence of pervasive computing are evaluated. The different strategies for handling privacy are pointed out, such as legislation, self-regulation and technology. Among these, technology is further looked into, first through an evaluation of existing solutions and research projects, then through design an implementation of a possible solution. The principles which are identified to be most important to handle are a user’s awareness of data collection and the possibility to restrict this collection. In addition it is pointed out the importance of making the system convenient to use. The heterogeneity of different users’ privacy preferences implies that some kind of personalization of the system should be present. A design and an implementation are presented which propose a solution where a user can constrain the access to his or her personal data, based on other users’ identities, his or her present situation and the type of context information the other users want to receive. The system also includes functionality to abstract details away from the context information which is disseminated to other system users/entities.</p>

<p>The Distributed Multimedia Plays (DMP) System Architecture provides combined adaptive scene resolution and traffic control in packet networks, see http://www.item.ntnu.no/~leifarne. This project focuses on adaptive scene composition declaration, specification and realisation, and comprises the following: * Review of 3D multiview, autostereoscopic object oriented audiovisual scenes theory and practice * Propose extensions to SMIL and SIP to handle adaptive composition of scenes consisting of distributed objects * Propose and demonstrate extensions to SMIL enabling 3D, transparency and custom shapes</p>

<p>a</p>

<p>This thesis focuses on context-aware services that make decisions based on the situation (i.e. context) of the involved entities. Context may in general be based on user input, sensed or derived (e.g. combination of multiple context entities). The type of such services is vast, but in this thesis the system is aimed towards the aquaculture industry. During the last years, aquaculture quality has become more and more important in the fish farming industry. But this importance has not been reflected yet in using information and communication technologies (ICT). The main problem in a fish farm is that most of them are without supervision for a long time while they are exposed to changing weather conditions. This problem gets even bigger when fish farms are established far from land and often becomes exposed to extreme weather conditions. In order to minimize the consequences caused by lack of information on the fish farm (such as weather conditions and other variables) when there is no workers around, ICT surveillance systems should be used. Context-aware services are perfectly suited for this type of application, and the task of this thesis is to specify, design and implement a context-aware application for the aquaculture industry. This includes a context source application, a context consumer application and a service to be deployed on a context management system. Our solution is named FiFaMoS (Fish Farm Monitoring System) and is based on the APMS context manager. This is a context management system that provides easy service deployment due to built-in support of multiple binding types as well as persistent storing of context. As a context source, an application for an M2M module is developed. There will be one module situated at each sea cage, which collects information like feed level, temperature, pH, oxygen level from the sensors. In addition, the module gets the positioning information for the sea cage from a connected GPS receiver. This information is periodically sent to the context manager that interprets the context and makes it available to users of the system (context consumers). In addition alarms will be triggered if values are out of bounds. It has been developed two different context consumers; one for personal computers and one for mobile phones. In these applications it is possible to view both current and historical sensor values, and receive alarms. It is also possible to alter the fish farm configuration via the PC client. Aspects that will be discussed are the use of different binding types, representation of data when transmitting and storing, hardware choices and various implementation choices. The implemented FiFaMoS system uses web services as binding to get a loosely coupled system, and objects are represented in XML which makes the system easy to alter. Detailed testing has been performed, and the system works as intended.</p>

<p>As increasingly more services and devices become integrated into pervasive systems, future network topologies will be vastly more sophisticated with numerous heterogeneous devices interconnected. To integrate a new service into this already complex network topology and traffic can give unwanted results if the functional blocks (applets) of a service are not placed at the best suited locations (devices). This thesis will look into the performance and scalability issues when confronted with options of multiple locations in which to run an applet. We will define a modelling framework taking into consideration system usage, network loads, device loads, overloads, timing requirements and propagation delays to mention some factors. In this framework we are able to set up our own scenarios with user patterns and the amount of users in the system. This framework will be written in Simula. From the output gained from this framework we can improve the system or the applets to improve overall traffic flow and resource usage. The framework will be run on a total of 8 different scenarios based on an airport usage model. We will have 6 static applets residing in their own devices and one dynamic applet which we will try to find the best location for within a predefined network topology. The amount of users can be set to a static amount or it can be a dynamic amount changing from hour to hour. The results produced give a better picture of the whole system working together. Based on these results it is possible to come to a conclusion of best suited applet location.</p>

<p>This master's thesis will investigate some of the technological solutions available for the construction of a backbone for a wireless citywide network (CWN), such as Wireless Trondheim, and a special attention is given to the possible advantages of using infrastructure already in place. There are several different technologies which can be used as a backbone for the wireless CWN, and in this thesis strengths and weaknesses of the various solutions will be discussed. Some requirements, such as scalability to handle increased capacity demand from future access technologies, may be more important to satisfy than others, and the discussion will see how this might rule out or favor some technology choices. In addition to demands from regular traffic to be carried in the network, also research traffic should be allowed. It is discussed how different requirements as a consequence of this research traffic will set limitations to which technology to use. While the performance of each of the technologies are relatively constant for all cases and easy to determine, the cost of using the different technologies is not. For this reason most effort is put into the analysis of this decision factor. Different possible scenarios are introduced, and the costs of using various technologies in the different scenarios are calculated. The results from the calculations are then discussed in more detail, and an important part of this discussion is sensitivity analysis. The values for several of the parameters used in the calculations will be changed, and the various analyses show how the change of one parameter will affect the result of the calculations. Based on the calculation results and the discussion of the results, some principles which can be used as guidelines in planning and deployment of a backbone for a wireless CWN will be suggested.</p>