Catálogo de publicaciones - libros

Similiar to a hard disk, NAND flash memory must be accessed in sector unit, and cannot be used for code storage without copying its contents to RAM. A virtual memory technique is promising as a RAM saving solution. However, it can not be easily used without the operating system supports, and it is not suitable for real time systems because it causes unpredicted execution delays. Virtual-ROM is a light-weight demand paging solution designed for RTOS based mobile devices. It is OS-independent, easy-to-use, and well optimized for NAND flash memory. Because it occupies only a restricted address space, real time tasks can be free from unpredictable execution delays by being excluded from Virtual-ROM. Our trace driven simulation showed that its performance is similar to 70ns NOR flash memory, and our real taget adaptation for a CDMA mobile phone showed that it saved about 30% RAM usage.

There is a growing interest in the model-driven software development as a viable solution to the increasing complexity of embedded software. An important characteristic of a model-driven approach is that it allows a developer to focus on high-level models rather than low-level details, thereby raising the level of abstraction at which a developer can work. Therefore, the model-driven approach can provide a significant productivity gain and improved maintainability. One of the most noticeable approaches is the OMG’s MDA (Model-Driven Architecture), which is a de facto standard. Unfortunately, MDA provides little support for the development of RTOS-based embedded software since it focuses primarily on middleware target platforms such as EJB, Web Services, .NET, and CORBA. In this paper, we present a model-driven approach to RTOS-based embedded software development and an automated tool that can produce RTOS-specific code. Our contributions are two fold. First, we propose generic RTOS APIs (Application Programming Interfaces) that can capture most of typical RTOS services that can be used as a means for describing application’s RTOS-related behavior at an early design stage. We then present a transformation tool that can generate fully functional code by transforming generic RTOS APIs into RTOS-specific APIs. Our tool, called TransPI, is able to produce ’C’ code for POSIX-compliant RTOSs, and is also configurable to target other RTOSs that do not conform to the POSIX standard.

Internet and chat mediums provide important and quite useful information about human life in different societies such as their current interests, habits, social behaviors and criminal tendency. In this study, we have presented an intelligent identification system that is designed to identify the sex of a person in a Turkish chat medium. To do this task, a target oriented chat agent is implemented. A simple discrimination function is proposed for sex identification. The system also uses a semantic analysis method to determine the sex of the chatters. Here, the sex identification is taken as an example in the information extraction in chat mediums. This proposed identification system employs the agent and acquires data from a chat medium, and then automatically detects the chatter’s sex from the information exchanged between the agent and chatter. The system has achieved accuracy about 90% in the sex identification in a real chat medium.

Information retrieval over clustered document collections has two successive stages: first identifying the best-clusters and then the best-documents in these clusters that are most similar to the user query. In this paper, we assume that an inverted file over the entire document collection is used for the latter stage. We propose and evaluate algorithms for within-cluster searches, i.e., to integrate the best-clusters with the best-documents to obtain the final output including the highest ranked documents only from the best-clusters. Our experiments on a TREC collection including 210,158 documents with several query sets show that an appropriately selected integration algorithm based on the query length and system resources can significantly improve the query evaluation efficiency.

Shared-nothing, parallel text retrieval systems require an inverted index, representing a document collection, to be partitioned among a number of processors. In general, the index can be partitioned based on either the terms or documents in the collection, and the way the partitioning is done greatly affects the query processing performance of the parallel system. In this work, we investigate the effect of these two index partitioning schemes on query processing. We conduct experiments on a 32-node PC cluster, considering the case where index is completely stored in disk. Performance results are reported for a large (30 GB) document collection using an MPI-based parallel query processing implementation.

In this paper we present a formal description of PROSA , a P2P resource management system heavily inspired by social networks. Social networks have been deeply studied in the last two decades in order to understand how communities of people arise and grow. It is a widely known result that networks of social relationships usually evolves to small–worlds, i.e. networks where nodes are strongly connected to neighbours and separated from all other nodes by a small amount of hops. This work shows that algorithms implemented into PROSA allow to obtain an efficient small–world P2P network.

E-cognocracy [1,2,3] is a cognitive democracy that combines both representative and participative democracy with two main objectives: (i) to overcome some of the problems associated with the traditional democratic system (participation, transparency and control) and (ii) to extract and socially share knowledge related with the scientific resolution of high complexity problems associated with the governance of society. E-cognocracy uses the democratic system as a catalyst for the vital or cognitive process characteristic of living systems, the multicriteria techniques as the methodological aid and the Internet as the communication and participation support. To capture the decision makers’ preferences and to share the knowledge related with the scientific resolution of the public decision making problems, the e-voting process of e-cognocracy has some characteristics that have not been covered by traditional e-voting systems. In this paper, we present the necessary modifications in the classical e-voting methods to deal with one of these properties: the linkage among rounds.

There are many challenges when providing multicast services to mobile hosts (MHs) since host’s mobility may cause tree reconstruction, non-optimal routing and the disruption of multicast services. Generally, frequent modifications to multicast tree are avoided as the expense of multicast routing inefficiency and vice versa. Two basic mechanisms, proposed by the IETF in Mobile IP to support multicasting, are two typical cases. This paper introduces a new mobile multicast approach with dynamic service range, which eliminates unnecessary tree reconstruction overhead by utilizing the state of multicast agent, and does not induce significant multicast delivery overhead at the same time. In addition, we present a previous subscription mechanism, which can resume multicast delivery quickly after a MH performs handoff. Simulation results show that the proposed protocol gets an improved performance over the existing protocols in terms of signaling overhead and delay.

Pervasive computing has emerged as a new computing par- adigm with a great deal with appeal in our everyday environment. However, the benefits offered by this new computing paradigm are relatively below our expected standard. This paper discusses the major challenges for the next generation pervasive computing and the difficulties in developing a promising system to meet these challenges. Then, we present a survey that covers relevant existing approaches addressed to overcome the challenges. Finally, we highlight future research directions and present a new intriguing exploration, aiming to broaden the appeal and bridge the gap for the fulfillment of the challenges.

Mobile virtual environments, with real-time 3D and 2D graphics, are now possible on smart phone and other camera-enabled devices. Using computer vision, the camera sensor can be treated as an input modality in applications by analyzing the incoming live video. We present our tracking algorithm and several mobile virtual environment and gaming prototypes including: a 3D first person shooter, a 2D puzzle game and a simple action game. Camera-based interaction provides a user experience that is not possible through traditional means, and maximizes the use of the limited display size.