Dessy : desktop search and synchronization

Current smartphones have a storage capacity of several gigabytes. More and more information is stored on mobile devices. To meet the challenge of information organization, we turn to desktop search. Users often possess multiple devices, and synchronize (subsets of) information between them. This makes file synchronization more important. This thesis presents Dessy, a desktop search and synchronization framework for mobile devices. Dessy uses desktop search techniques, such as indexing, query and index term stemming, and search relevance ranking. Dessy finds files by their content, metadata, and context information. For example, PDF files may be found by their author, subject, title, or text. EXIF data of JPEG files may be used in finding them. User–defined tags can be added to files to organize and retrieve them later. Retrieved files are ranked according to their relevance to the search query. The Dessy prototype uses the BM25 ranking function, used widely in information retrieval. Dessy provides an interface for locating files for both users and applications. Dessy is closely integrated with the Syxaw file synchronizer, which provides efficient file and metadata synchronization, optimizing network usage. Dessy supports synchronization of search results, individual files, and directory trees. It allows finding and synchronizing files that reside on remote computers, or the Internet. Dessy is designed to solve the problem of efficient mobile desktop search and synchronization, also supporting remote and Internet search. Remote searches may be carried out offline using a downloaded index, or while connected to the remote machine on a weak network. To secure user data, transmissions between the Dessy client and server are encrypted using symmetric encryption. Symmetric encryption keys are exchanged with RSA key exchange. Dessy emphasizes extensibility. Also the cryptography can be extended. Users may tag their files with context tags and control custom file metadata. Adding new indexed file types, metadata fields, ranking methods, and index types is easy. Finding files is done with virtual directories, which are views into the user’s files, browseable by regular file managers. On mobile devices, the Dessy GUI provides easy access to the search and synchronization system. This thesis includes results of Dessy synchronization and search experiments, including power usage measurements. Finally, Dessy has been designed with mobility and device constraints in mind. It requires only MIDP 2.0 Mobile Java with FileConnection support, and Java 1.5 on desktop machines.

Compression of Weighted Graphs

We propose to compress weighted graphs (networks), motivated by the observation that large networks of social, biological, or other relations can be complex to handle and visualize. In the process also known as graph simplication, nodes and (unweighted) edges are grouped to supernodes and superedges, respectively, to obtain a smaller graph. We propose models and algorithms for weighted graphs. The interpretation (i.e. decompression) of a compressed, weighted graph is that a pair of original nodes is connected by an edge if their supernodes are connected by one, and that the weight of an edge is approximated to be the weight of the superedge. The compression problem now consists of choosing supernodes, superedges, and superedge weights so that the approximation error is minimized while the amount of compression is maximized. In this paper, we formulate this task as the 'simple weighted graph compression problem'. We then propose a much wider class of tasks under the name of 'generalized weighted graph compression problem'. The generalized task extends the optimization to preserve longer-range connectivities between nodes, not just individual edge weights. We study the properties of these problems and propose a range of algorithms to solve them, with dierent balances between complexity and quality of the result. We evaluate the problems and algorithms experimentally on real networks. The results indicate that weighted graphs can be compressed efficiently with relatively little compression error.

Middleware for Mobile Sensing Applications in Urban Environments

Sensor networks represent an attractive tool to observe the physical world. Networks of tiny sensors can be used to detect a fire in a forest, to monitor the level of pollution in a river, or to check on the structural integrity of a bridge. Application-specific deployments of static-sensor networks have been widely investigated. Commonly, these networks involve a centralized data-collection point and no sharing of data outside the organization that owns it. Although this approach can accommodate many application scenarios, it significantly deviates from the pervasive computing vision of ubiquitous sensing where user applications seamlessly access anytime, anywhere data produced by sensors embedded in the surroundings. With the ubiquity and ever-increasing capabilities of mobile devices, urban environments can help give substance to the ubiquitous sensing vision through Urbanets, spontaneously created urban networks. Urbanets consist of mobile multi-sensor devices, such as smart phones and vehicular systems, public sensor networks deployed by municipalities, and individual sensors incorporated in buildings, roads, or daily artifacts. My thesis is that "multi-sensor mobile devices can be successfully programmed to become the underpinning elements of an open, infrastructure-less, distributed sensing platform that can bring sensor data out of their traditional close-loop networks into everyday urban applications". Urbanets can support a variety of services ranging from emergency and surveillance to tourist guidance and entertainment. For instance, cars can be used to provide traffic information services to alert drivers to upcoming traffic jams, and phones to provide shopping recommender services to inform users of special offers at the mall. Urbanets cannot be programmed using traditional distributed computing models, which assume underlying networks with functionally homogeneous nodes, stable configurations, and known delays. Conversely, Urbanets have functionally heterogeneous nodes, volatile configurations, and unknown delays. Instead, solutions developed for sensor networks and mobile ad hoc networks can be leveraged to provide novel architectures that address Urbanet-specific requirements, while providing useful abstractions that hide the network complexity from the programmer. This dissertation presents two middleware architectures that can support mobile sensing applications in Urbanets. Contory offers a declarative programming model that views Urbanets as a distributed sensor database and exposes an SQL-like interface to developers. Context-aware Migratory Services provides a client-server paradigm, where services are capable of migrating to different nodes in the network in order to maintain a continuous and semantically correct interaction with clients. Compared to previous approaches to supporting mobile sensing urban applications, our architectures are entirely distributed and do not assume constant availability of Internet connectivity. In addition, they allow on-demand collection of sensor data with the accuracy and at the frequency required by every application. These architectures have been implemented in Java and tested on smart phones. They have proved successful in supporting several prototype applications and experimental results obtained in ad hoc networks of phones have demonstrated their feasibility with reasonable performance in terms of latency, memory, and energy consumption.

Modeling Efficient Classification as a Process of Confidence Assessment and Delegation

In visual object detection and recognition, classifiers have two interesting characteristics: accuracy and speed. Accuracy depends on the complexity of the image features and classifier decision surfaces. Speed depends on the hardware and the computational effort required to use the features and decision surfaces. When attempts to increase accuracy lead to increases in complexity and effort, it is necessary to ask how much are we willing to pay for increased accuracy. For example, if increased computational effort implies quickly diminishing returns in accuracy, then those designing inexpensive surveillance applications cannot aim for maximum accuracy at any cost. It becomes necessary to find trade-offs between accuracy and effort. We study efficient classification of images depicting real-world objects and scenes. Classification is efficient when a classifier can be controlled so that the desired trade-off between accuracy and effort (speed) is achieved and unnecessary computations are avoided on a per input basis. A framework is proposed for understanding and modeling efficient classification of images. Classification is modeled as a tree-like process. In designing the framework, it is important to recognize what is essential and to avoid structures that are narrow in applicability. Earlier frameworks are lacking in this regard. The overall contribution is two-fold. First, the framework is presented, subjected to experiments, and shown to be satisfactory. Second, certain unconventional approaches are experimented with. This allows the separation of the essential from the conventional. To determine if the framework is satisfactory, three categories of questions are identified: trade-off optimization, classifier tree organization, and rules for delegation and confidence modeling. Questions and problems related to each category are addressed and empirical results are presented. For example, related to trade-off optimization, we address the problem of computational bottlenecks that limit the range of trade-offs. We also ask if accuracy versus effort trade-offs can be controlled after training. For another example, regarding classifier tree organization, we first consider the task of organizing a tree in a problem-specific manner. We then ask if problem-specific organization is necessary.

Digitaalisten teosten maailmanlaajuinen rekisteröinti ja käyttömaksujen kerääminen internetissä

The vastly increased popularity of the Internet as an effective publication and distribution channel of digital works has created serious challenges to enforcing intellectual property rights. Works are widely disseminated on the Internet, with and without permission. This thesis examines the current problems with licence management and copy protection and outlines a new method and system that solve these problems. The WARP system (Works, Authors, Royalties, and Payments) is based on global registration and transfer monitoring of digital works, and accounting and collection of Internet levy funded usage fees payable to the authors and right holders of the works. The detection and counting of downloads is implemented with origrams, short and original parts picked from the contents of the digital work. The origrams are used to create digests, digital fingerprints that identify the piece of work transmitted over the Internet without the need to embed ID tags or any other easily removable metadata in the file.

Developing Peer-To-Peer Web Applications

As the virtual world grows more complex, finding a standard way for storing data becomes increasingly important. Ideally, each data item would be brought into the computer system only once. References for data items need to be cryptographically verifiable, so the data can maintain its identity while being passed around. This way there will be only one copy of the users family photo album, while the user can use multiple tools to show or manipulate the album. Copies of users data could be stored on some of his family members computer, some of his computers, but also at some online services which he uses. When all actors operate over one replicated copy of the data, the system automatically avoids a single point of failure. Thus the data will not disappear with one computer breaking, or one service provider going out of business. One shared copy also makes it possible to delete a piece of data from all systems at once, on users request. In our research we tried to find a model that would make data manageable to users, and make it possible to have the same data stored at various locations. We studied three systems, Persona, Freenet, and GNUnet, that suggest different models for protecting user data. The main application areas of the systems studied include securing online social networks, providing anonymous web, and preventing censorship in file-sharing. Each of the systems studied store user data on machines belonging to third parties. The systems differ in measures they take to protect their users from data loss, forged information, censorship, and being monitored. All of the systems use cryptography to secure names used for the content, and to protect the data from outsiders. Based on the gained knowledge, we built a prototype platform called Peerscape, which stores user data in a synchronized, protected database. Data items themselves are protected with cryptography against forgery, but not encrypted as the focus has been disseminating the data directly among family and friends instead of letting third parties store the information. We turned the synchronizing database into peer-to-peer web by revealing its contents through an integrated http server. The REST-like http API supports development of applications in javascript. To evaluate the platform s suitability for application development we wrote some simple applications, including a public chat room, bittorrent site, and a flower growing game. During our early tests we came to the conclusion that using the platform for simple applications works well. As web standards develop further, writing applications for the platform should become easier. Any system this complex will have its problems, and we are not expecting our platform to replace the existing web, but are fairly impressed with the results and consider our work important from the perspective of managing user data.