Vol. 30 · No. 6

Nowadays information systems are being shifted to distributed architectures, i.e. Grid and Peer-to-peer (P2P) models to obtain the benefits like scalability, autonomy, and fault-tolerance. We consider the P2P model as a fully distributed, scalable system, which is composed of peer processes (peers). Here, a group of multiple peers cooperate with each other. Peers have to efficiently and flexibly deliver messages to every peer of the group in P2P overlay networks. In order to efficiently and reliably broadcast messages in a scalable group, we take advantage of the multipoint relaying (MPR) mechanism. Here, each peer sends messages to only a subset of its acquaintances. However, if a peer which forwards messages to other peers is faulty, the peers cannot receive messages. In this paper, we newly discuss a trustworthiness-based broadcast (TBB) algorithm where only trustworthy peers forward messages. That is, untrustworthy peers are peers which cannot forward the messages due to some faults. Here, the transmission fault implied by faults of untrustworthy peers can be reduced. We evaluate the TBB algorithm in terms of the number of messages transmitted.

Smartphone technologies have enabled sophisticated pervasive applications for mobile users. Still, many intensive applications perform poorly on smartphones due to the shortage of resources for computation, data storage, network bandwidth, and battery capacity. While such applications can be re-designed with client-server models to benefit from subscribed cloud services, the users are no longer in full control of the entire application execution, which has raised a serious concern. Meanwhile, privacy and security are also important issues, and it is an ongoing debate if public cloud services could be trusted with sensitive data. For mobile users to take full advantage of cloud services, these issues need to be resolved. In this paper, we propose an innovative framework for mobile users to execute existing Android applications on a personal virtual phone safely in the cloud. Instead of using a client-server model, the entire virtual phone is mostly controlled by the user to minimize the intervention from the service provider. Virtualization and encryption are employed to protect against eavesdropping from cloud providers and network attackers. To quickly migrate an Android application between the physical phone and the virtual phone, we use a new application-level checkpointing mechanism and minimize the state of the application.

This paper introduces two new classes of specialised vision systems called UBIAS and E-UBIAS. Such systems belong to the group of cognitive reasoning systems and are designed for the semantic analysis of visual patterns in the form of medical images. Proposed systems are predecessors of a new generation of intelligent systems dedicated for understanding medical diagnostic visualization and using this data as biometric characteristics.

This paper presents a novel method that automatically indexes searches for relevant images using a combination of geo-coded information and content-based visual features. Photographic images are labeled with their corresponding GPS (Global Positioning System) coordinates and UTC time (Coordinated Universal Time) information at the moment of capture, which are then utilized to create spatial and temporal indexes for photograph retrieval. Assessing the performance in terms of average precision and F-score with real-world image collections revealed that the proposed approach significantly improved and enhanced the retrieval process compared to searches based on visual content alone. Combining content and context information thus offers a useful and meaningful new approach to searching and managing large image collections.

Pervasive computing is a user-centric mobile computing paradigm, in which tasks should be migrated over different platforms in a shadow-like way when users move around. In this paper, we propose a context-sensitive task migration model that recovers program states and rebinds resources for task migrations based on context semantics through inserting resource description and state description sections in source programs. Based on our model, we design and develop a task migration framework xMozart which extends the Mozart platform in terms of context awareness. Our approach can recover task states and rebind resources in the context-aware way, as well as support multi-modality I/O interactions. The extensive experiments demonstrate that our approach can migrate tasks by resuming them from the last broken points like shadows moving along with the users.

A Mobile Ad hoc Network (MANET) is a collection of wireless mobile terminals that are able to dynamically form a temporary network without any aid from fixed infrastructure or centralized administration. In this paper, we present the implementation and analysis of our implemented MANET testbed and simulation system considering Ad-hoc On demand Distance Vector (AODV) and Optimized Link State Routing (OLSR) protocols for wireless multi-hop networking. We investigate the effect of mobility and topology changing in MANET. We evaluate and compare the performance by simulation (using ns-2 simulator) and experiments in a real environment. In this work, we consider two models: stationary and mobile. We assess the performance of our testbed and simulation in terms of throughput, number of received packets and hop distance. From the results, we found that the AODV protocol has a good performance when the relay node is moving. Also, the AODV protocol provides a flexible and effective routing for indoor environments.

The growing applications of mobile ad hoc networks (MANETs) have made related security issues much more important. B. Zhu et al. proposed a key management scheme using Shamir's secret sharing scheme to construct an Autonomous Key Management (AKM) hierarchy structure. However, Shamir's secret sharing in AKM to control key hierarchy incurs high message transmission costs. This paper modifies the secret sharing scheme and applies it to AKM to reduce communication and computation costs.

Context data is important information for catching the behaviors of applications in a pervasive space. To effectively store huge amount of data, tree-like layered storage architecture is proposed, where the leaf nodes collect data from sensing devices. In order to integrate data from mobile devices, the related leaf nodes that get data from the same device should upload and store the data to the host node. This paper presents a deep study of the data storage problem and proposes a global algorithm GHS and an online algorithm DHS to dynamically select the host node, which reduces the communication cost significantly. This paper also gives the theoretical and experimental analysis of these algorithms, which shows both GHS and DHS are correct and effective.

The paradigm of distributed computation in heterogeneous resources, grid computing, has given rise to a large amount of research on resource scheduling. This paper presents a Meta-Scheduler for grid computing that does not need any given information about tasks length or tasks arrival time unlike traditional dynamic heuristics. Our Meta-Scheduler is of multi-criteria type, because it solves two conflicting objectives: minimize the makespan of a set of tasks and distribute these tasks in a balanced way among the resources of the Grid. Experimental results using fuzzy scheduler show that, through our proposal, we achieve these two objectives and improve dynamic heuristics presented in prior literature.

In this paper a multiagent system for intelligent virtual environments using the Magentix Multiagent Platform is presented. It is based on a previous framework which has been improved with agent groups, security and efficiency concerns. Therefore, the framework presented can be used in common Intelligent Virtual Environment domains such as education, commercial games and simulation.

With the recent development of high-performance graphical processing units (GPUs), capable of performing general-purpose computation (GPGPU: general-purpose computation on the GPU), a new platform is emerging. It consists of a central processing unit (CPU), which is very fast in sequential execution, and a GPU, which exhibits high degree of parallelism and thus very high performance on certain types of computations. Optimally leveraging the advantages of this platform is challenging in practice. We spotlight the analogy between GPGPU and hardware/software co-design (HSCD), a more mature design paradigm, to derive a design process for GPGPU. This process, with appropriate tool support and automation, will ease GPGPU design significantly. Identifying the challenges associated with establishing this process can serve as a roadmap for the future development of the GPGPU field.

Exploiting Grids intuitively requires developers to alter their applications, which calls for expertise on Grid programming. Gridification tools address this problem by semi-automatically making user applications to be Grid-aware. However, most of these tools produce monolithic Grid applications in which common tuning mechanisms (e.g. parallelism) are not applicable, and do not reuse existing Grid middleware services. We propose BYG (BYtecode Gridifier), a gridification tool that relies on novel bytecode rewriting techniques to parallelize and easily execute existing applications via Grid middlewares. Experiments performed by using several computing intensive applications on a cluster and a simulated wide-area Grid suggest that our techniques are effective while staying competitive compared to programmatically using such services for gridifying applications.

We present numerical studies of the positional word web citeCanso, based on an ancient text (The Bible). Adopting methods from the graph theory we show that such a web has small world and scale free properties citeWatts,Albert. Degree distribution and hierarchy of nodes in various word webs are also studied. Complete solution of the mathematical model, which explains the differences between the data and the previous model of Dorogovtsev and Mendes citeDogmen1 is presented.

Mathematical modelling of forest fire propagation in time plays a key role in existing fire propagation predicting systems. Such systems are capable of simulating the growth of fire front in time and space and predicting spatial and temporal fire behaviour. In our previous papers, we studied mathematical foundations of the elliptical fire propagation model implemented in several advanced simulation systems. The model is based on Huygens' principle applied on fire propagation assuming locally elliptical fire spread. However, in the literature various other mathematical representations of local fire propagation have been reported, such as double ellipse, lemniskata, oval shape, tear shape, and others. Such types of local fire propagation have been experimentally observed in certain conditions during laboratory and field fires. In this paper, we demonstrate several simple examples of fire propagation corresponding both to the elliptical and non-elliptical local fire propagation in homogeneous conditions.