DATA TRANSMISSION DEVICES

Author: PEIG OLIVE ENRIC.
Year: 2003.
University: POMPEU FABRA.
Place of defense: DEPARTAMENTO DE TECNOLOGIA.
Place of preparation: DEPARTAMENTO DE TECNOLOGIA.

Summary: Using metadata to reference, label and describe resources is becoming increasingly common. This allows for better mechanisms for detecting and locating material posted in public repositories accessible by the Internet. We are pursuing various initiatives to establish standards for metadata models, but each focuses on its own requirements for the purpose of defining attributes, its meaning, the possible values that can take and the relationship between them. From the user's perspective that seeks to find resources in different environments, this is a real problem, because it forces him to have the appropriate tools for each case, or be aware of the differences between them. In this thesis work is presented and validated a model for interoperability between different schemes metadata at the semantic level suitable for environments search of material, whose main contribution is that it offers such interoperability in a transparent manner. Neither content providers, who offer metadata, or for the users, who are those who consult, they must be aware of its existence, thus allowing all players in the system continue to use its own scheme metadata . To that end, the model requires the presence of an external agent that implements it.

Author: HIGUERO APERRIBAI M. VICTORIA.
Year: 2004.
University: PAÍS VASCO.
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA DE BILBAO.
Place of preparation: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA DE BILBAO (UPV/EHU).

Summary: The digitilización of information has opened up a great number of opportunities in various economic sectors, through new technologies have largely succeeded in lowering various processes of their business. At the same time the deployment and development of the Internet has given them new schemes for distributing materials faster and cheaper than their predecessors, as well as a potential market on a global scale. However, this situation has led kicker ease of illegally accessing their products by millions of users, which also can intercambiárselo easily via the Internet, thus violating the copyrights associated with a large part of these materials. These so-called "piracy" is now one of the greatest concerns you of almost all the players in this sector, which consistently promoted the development of solutions to the problem. In this context the watermarks are the basis for most of the various solutions proposed so far, but still has developed a universal solution to resolve the problem completely. The overall objective of this thesis is the definifición a distribution model that will allow the establishment of new schemes to carry out transactions with digital content, respecting all the rights involved, as are the copyright associated with the authors, rights of clients like anonymity , and so on. With simplicity in terms of features and performance to ensure the viability of this model. To study and evaluation of the security model as a key aspect for its viability, defines a pattern of Risk Analysis based on the application of the methodology of the "attack trees." This process gives us, in a systematic way, information on various issues related to the level of security supported by a distribution model as proposed, as well as significant comparative data between different distribution patterns.

Author: MARTINEZ BALLESTE ANTONI.
Year: 2004.
University: POLITÉCNICA DE CATALUÑA.
Place of defense: Edifici B-3 1a. planta. Aula Teleensenyament.
Place of preparation: EDIFICI C3 Campus NORD.

Summary: This thesis covers different aspects of e-commerce. More precisely, it deals with both unicast and multicast pay-per-view distribution of protected multimedia streams. From the customer's point of view, content should be paid as it is being accessed. From the merchant's point of view, the intellectual property of his content must be protected and/or illegal redistributors must be traceable. Finally, the merchant must be able to use web statistics to improve his revenue and/or customer-merchant relationship. Moreover, the customer must be sure that her privacy is protected when web statistics are data mined by third parties. After an overview of the technology related to e-commerce of video streams towards personal devices, this thesis has presented as results prototypes, security protocols and data processing algorithms. Several results have been presented in this thesis: - We have presented a design for a pay-as-you-watch service, in which content is being paid for as it is being received. A prototype has been implemented and tested over WLAN, GPRS and UMTS networks. Moreover, we have proposed a set of protocols for secure fingerprinting based on a TTP.From experimentation with that prototype, sending payment tokens through wireless networks has proven to be feasible, whereas generation of payment tokens by handheld devices has proven to be a heavy task due to the low computing power of such devices. - We have addressed pay-as-you-watch and fingerprinting copyright protection in multicast environments. Two problems must be overcome. On one hand, payment implosion arises at the source; on the other hand it seems unnatural that a single stream that is multicast can be fingerprinted differently for each customer. We have overcome the first problem with two contributions: a subscription-based payment scheme using our protocol for securely aggregating information from the receivers to the multicast source (this protocol can also be used for security in sensor data collecting or even for deploying a secure large-scale bingo) and a protocol for aggregation of verifiable payments. - Regarding the second problem (multicast fingerprinting), we have attacked and repaired the attractive Parvainien approach of encryption-based multicast fingeprinting. - Finally, customer privacy in web statistics has been addressed. We have contributed a method for protecting web statistics which allows data mining without disclosure of customer identities. The method is based on generation of synthetic data whose statistical properties are similar to original customer data. The advantage is that our method computes synthetic data in a single step, rather than using iterative refinement as most synthetic data generation methods in the literature.

Author: LLORENTE VIEJO SILVIA.
Year: 2004.
University: POMPEU FABRA.
Place of defense: DEPARTAMENTO DE TECNOLOGIA.
Place of preparation: UNIVERSIDAD POLITÉCNICA DE CATALUNYA.

Summary: The research work done in this thesis has centered around two main concepts: electronic commerce and services, in order to specify the e-commerce services (abbreviated and services, and services, in English). To meet this goal, has been developed and validated a methodology that allows the description of the e servicios.El starting point for this research was the development and implementation of a system for the provision of legal and administrative services under the project European TRADE (TRiAls in the Domain of Electronic Commerce). This system was developed using proprietary tools, which hampered its expansion and application to other types of services. However, the concepts learned in this project are applicable to other areas, so we are continuing our research in this line. The thesis is divided into paragraphs of the State of the art, Contribution, Conclusions and future lines and references and apéndices.En paragraph of the state of the art, are some initiatives that uses the XML language for describing processes and, in general, for describing the structure of the same. It also describes patterns expressed in XML metadata. The selection of these two items was motivated by the need to describe the structure of services to be provided in electronic form and its associated information. This imposed the additional requirement that such a description is made using language XML.En point contribution, describing the contributions made by this research work in relation to the description, definition and implementation of e-commerce services. They have used some of the elements described in the state of the art. Specifically, in the various sections of this section introduces the concept of electronic commerce, focusing on the case of services, compared with e-commerce products. It also provides a definition of and service, which leads to the classification of services and in terms of different characteristics studied. Furthermore, it explains the description of the structure of services and methods of control over workflow (workflow, in English). Finally, it presents the proposed methodology for defining and services, with the description of each component, and is validated by this methodology with the help of two distinct types of services: legal services, administrative and joint editing of documents. In terms of conclusions and future lines, details the findings of the investigation conducted and describes some of the lines of research that are opening up with this trabajo.En paragraphs references and appendices detailing the literature used for the realization of this labor and develops in detail the modeling of some of the services studied.

Author: GIL IRANZO ROSA MARÍA.
Year: 2004.
University: POMPEU FABRA.
Place of defense: DEPARTAMENTO DE TECNOLOGÍA.
Place of preparation: DEPARTAMENTO DE TECNOLOGÍA.

Summary: The semantic architecture designed has been tested in the environment for the management of intellectual property rights, specifically within projects NewMars and AgentWeb. The most significant features of this architecture is that the use of ontologies related technologies, systems experts and agents to achieve a high degree of flexibility to customize addition to innately possess a great integration with other environments. Despite develop ontologies and new multi-agent systems, this architecture allows new modules, as well as changes in the structure, this factor is relevant because it continuously for example will develop new versions of multi-agent systems. Regarding the ontologies have been specifically designed its own: IPROnto, which is completely interoperable with other standards, and proposes a new way of managing the intellectual property rights that allows the coexistence of two aspects that might seem a priori antagonists: customization and integration with diversity formatos.Así way, ontologies have been used to model the knowledge that they have the players. Specifically IPROnto was sent to the first international conference on Semantic Web and was also nominated for the standard MPEG-21. Specifically, we made a proposal for the dictionary terms of rights of expression and language rights. The proposal was offered as pioneering a unique way to be able to represent terms and language of a highly compact: IPROnto, which allows you to customize and at the same time meet the standards. In addition, we performed a thorough analysis systems Multi-Agente fulfill two basic characteristics, high integration and high semantic interoperability, we find a system that is very well suited to these features: JADE-LEAP. We studied extensively integrating players and customization in the process of negociaciónPara provide the architecture model of the Semantic Web as real as possible so that when agents did not have information could remove it selectively and not using 'force Gross' conducted a statistical analysis of the Semantic Web as well as the elements that we were using, IPROnto. This study revealed that the Semantic Web behaved as a complex system, ie autonomous, which owned properties for which one could characterize microscopic behavior. This offers perspectives unimaginable to devise new applications within this network, thus taking advantage a key feature of the Semantic Web: the meaning of relationships between different elementos.Así which to validate the escabilidad and applicability of this study started working in different activities with applications-oriented ad-hoc Web semánticaLa last part that is still developing manages the possibility of giving officers the ability to 'learn how to learn', as in the negotiating process are many obstacles to power independently increase awareness of the actors. In order to achieve some ontologies are discussed as a paradigm of 'ultimate knowledge' as DOLCE or FrameNet, besides processes to interiorizarlas near cognitive science using 'image-schemas' for attaching to the semantic web, in addition to using geometric algebra to connect the symbolism (algebra) and semantics (geometry understood as meaning geometric) as a last step. Finally, this section shows a methodology to discover new knowledge from the physical experience up to the abstract knowledge using a model human cognitive. Although this work is still in development meeting presents some results applied to the negotiation process

Author: SOBRADO DELGADO IGOR.
Year: 2005.
University: OVIEDO.
Place of defense: DEPARTAMENTO DE INFORMATICA.
Place of preparation: DEPARTAMENTO DE INFORMATICA.

Summary: The author of this Doctoral Thesis proposed three strategies aimed at routing traffic in communication networks based on IPv6 protocol. The basic objective is to increase safety, both in protecting the payload of packets as in the fault-tolerance of the communication infrastructure. As a complementary objective, it seeks to minimize the impact of these techniques on the yield of intermediate systems to allow its introduction in the new data transmission networks of ultra-alta speed. The three strategies presented converge on an innovative model of network autoorganizativa we call NETWORK ADAPTATIVA. This network has the capacity to restructure if partially destroyed. As a side effect, we have found a deficiency in the model addressing anycasting, which suggest a possible solution which we call ANYCASTING SELECTIVE based on the use of a new heading for the optional protocol IPv6.

Author: LIBERAL MALAINA FIDEL.
Year: 2005.
University: PAÍS VASCO.
Place of defense: ETSI DE BILBAO.
Place of preparation: ETSI BILBAO.

Summary: In this thesis work is intended to address the problems of quality of service from a global point of view. In addition to analyzing the performance objective of the different actors involved in the provision of service, it is intended to represent what effect this performance in the satisfaction of end users, seen as the ultimate goal of systems analysis and quality management. It has conducted a focused state of the art in 4 different areas: the QoS technique, models for assessing the subjective quality, systems of quality management and business approaches integrators. Based on that analysis of the state of the art, we have identified a number of deficiencies that will mark the characteristics due to the model proposed. This justifies the need for a model will mark the characteristics due to the proposed model. This justifies the need for a model that allows us to represent the different facets of quality with a global perspective. The proposal consists of a theoretical model and an implementation methodology that facilitates their use by researchers outside this work. Specifically, we propose a matrix model that represents one hand, the services they want to analyze, which will depend in turn on the user preference. In the same vertical axis is collected for each of genetic services concerned, the most significant subjective perceptions, which will mark how users evaluate quality. At the other axis is the facet objective, represented through the various actors involved in the provision of services and features that contribute to the value chain. Finally, matches or crossing points reveal the existence of relations between the two realities. From that abstract model generally defines a number of elements and interactions intermediate, and a mathematical formulation which provides a framework for the development of subsequent mathematical analysis. To complete the proposal, develops a methodology for implementation, which defines a series of steps to be carried out by the investigator to build the model. It is a methodology "back to front", as long as part of the subjective dimension and is retreating to identify objective parameters on which it depends. Finally, the methodology incorporates a series of diagnostic tools mathematical analysis expensive to extract useful information from this model built. To test the usefulness of the model and methodology proposed, apply them to illustrate one of the cases that led to precisely the entire thesis work. Thus, it shall be ensured that the proposal allows represent the variation of user satisfaction due to the behavior of one of the agents, as well as identify the most responsible for this satisfaction every time.

Author: ORTUÑO PÉREZ MIGUEL ÁNGEL.
Year: 2005.
University: REY JUAN CARLOS.
Place of defense: ESCUELA SUPERIOR DE CIENCIAS EXPERIMENTALES Y TECNOLOGÍA.
Place of preparation: ESCUELA SUPERIOR DE CIENCIAS EXPERIMENTALES Y TECNOLOGÍAS.

Summary: If we want any devices can connect to a global network, it is highly desirable to have a unique identifier. For teams with limited resources (sensors, toys, disposable devices, micro-robots, etc..), The size of that identifier can make an overload intractable because conventional techniques are handled in full addresses transmission, processing and storage of data needed for routing. This is felt especially in DSR, a source routing protocol for ad hoc networks, where each data packet must store the addresses of all stations through which must pass to reach their destination. In this type of device the typical solution is to limit the address space to one or two octets, thus sacrificing the uniqueness global identifiers. Or more dramatically, dispense with the routing. We present in this thesis a proposal that keeps addresses identified globally as the seasons, because the tasks of routing employ what we call leadership discriminant for routing, address or abbreviated). They will be not only, ie two or more different stations can share the same shorthand, we call made to the collision. Do not avoid collisions, but will develop techniques to tolerate the consequences. We develop this idea in a particular protocol which call ADSR, amending the protocol for handling DSR abbreviated addresses. On the other hand, the Ad-Hoc networks typically require lso layer protocols link the reliability of the physical environment that lacks wireless. Using abbreviated addresses as SARD ago, a unique network address corresponds to link several directions, so it is necessary to send to multiple recipients reliable characteristics of the protocols available to us liaison usual. To meet this requirement, we have developed the protocol to which we call LLRB Link Layer Reliable Broadcasta, based on IEEE 802.11. LLRB consists of three modules, each of which develops a novel algorithm: Extension of MACA mechanism, which is located in the frame sequence control RTS-CTS-dato-ACK by a fabric RTS, CTS several frames, a frame data and several raster ACK. Algorithm contest limited by masks choice slot LCSEN mechanism by which the issuer node coordinates the sending of raster CTS and ACK by the recipients of the data. This uses a single octet, which comprise the target with its own address. Algorithm slight estimation algorithm neighborhood or master blind LANE, which provides an estimate of the nodes adjacent to the node issuer, without employing more frames that cTS and ACL extension of MACA. This algorithm detects the absence of a reply CTS and ACL and attached either collisional access to the media, either to changes in the network topology. This thesis presents the design of protocols ADSR and LLRB as an implementation of the simulator on the same network ns-2. It also analyzes the performance of both simulator under various conditions, obtained some conditions and will be introduced lines for future work.

Author: RODRIGUEZ-MORCILLO GARCIA CARLOS.
Year: 2006.
University: PONTIFICIA COMILLAS.
Place of defense: Alberto Aguilera 23.
Place of preparation: Escuela Técnica Superior de Ingenierias.

Summary: This thesis proposes a digital communications system that per-mite convey a great deal of information between teams embar-cados in a transport vehicle, thus increasing the capacity for transmitting systems currently installed. In addition, the system propues-to is capable of functioning sharing the same physical environment that siste-mas current to minimize the cost and resources involved in setting up a new system in a vehicle and manufactured. To carry out the development of proposed digital system has been rea-lizado a study theoretical and experimental analysis in several phases. In pri-mer place, has been obtained electromagnetic model of a transmission line from a transport vehicle, including a train. This has analyzed the feasibility of the thesis. Secondly, it has developed a communications system pro-puesto basing its operation in the technical Orthogonal Frequency Divi-sion Multiplexing (OFDM), for his ability to convey a great deal of information with low bandwidth. This technique is used in digital systems such as WLAN, DSL, PLC and DVB-T. In this part of the investigation have been defined all the parameters necessary for the operation of the system and is configured with appropriate values. Then there has been a simulator system software previously developed, which can analyze the operation of the same to get the optimum configuration. This simulator has been taken into account both the model of the transmission line analyzed in the prime-ra part of the thesis, like the ability to run concurrently with other systems. Subsequently, it has built a platform of evidence based FPGA with which it has been verified the correct functioning of the system di-gital developed and with which it has validated the simulator before. This has enabled transmit over 30 meters of cable, a wealth of data from 8 Mbps on a bandwidth of 3 MHz. Lastly, there has been a testing laboratory with which has been compared simultaneous operation of the system developed in the thesis with other system currently used in railway vehicles, specifically with the standard IEC 61375-1 on the Train Communica-tion Network (TCN). This test validated the work done in the Te-sis and has shown that the system developed, it can increase the flow of information from a communications system currently insta-lado in a transport vehicle, with a minimum cost computers and instala-ción.