Saul E. Pomares Hernandez

A Neural Network Scheme for Long-Term Forecasting of Chaotic Time Series

The accuracy of a model to forecast a time series diminishes as the prediction horizon increases, in particular when the prediction is carried out recursively. Such decay is faster when the model is built using data generated by highly dynamic or chaotic systems. This paper presents a topology and training scheme for a novel artificial neural network, named “Hybrid-connected Complex Neural Network” (HCNN), which is able to capture the dynamics embedded in chaotic time series and to predict long horizons of such series. HCNN is composed of small recurrent neural networks, inserted in a structure made of feed-forward and recurrent connections and trained in several stages using the algorithm back-propagation through time (BPTT). In experiments using a Mackey-Glass time series and an electrocardiogram (ECG) as training signals, HCNN was able to output stable chaotic signals, oscillating for periods as long as four times the size of the training signals. The largest local Lyapunov Exponent (LE) of predicted signals was positive (an evidence of chaos), and similar to the LE calculated over the training signals. The magnitudes of peaks in the ECG signal were not accurately predicted, but the predicted signal was similar to the ECG in the rest of its structure.

Abstract channels as connectors for software components in group communication services

Building new services by assembling software components, when adopted at the communication level, would allow developers to build powerful group communication services by assembling proven and efficient algorithms for ordered group communication. The integration of communication protocols with different delivery policies has not been addressed in past research on group communication. This integration cannot be considered as concatenation of protocols and needs delivery policies to be redefined in the context of multi-channel communication. Our paper deals with this issue. We define communication channels as software connectors of the communication level for coordinated group communication. We cover the three standard delivery policies (FIFO order, causal order, and total order).

GMTE: A Tool for Graph Transformation and Exact/Inexact Graph Matching

Multi-labelled graphs are a powerful and versatile tool for modelling real applications in diverse domains such as communication networks, social networks, and autonomic systems, among others. Due to dynamic nature of such kind of systems the structure of entities is continuously changing along the time, this because, it is possible that new entities join the system, some of them leave it or simply because the entities’ relations change. Here is where graph transformation takes an important role in order to model systems with dynamic and/or evolutive configurations. Graph transformation consists of two main tasks: graph matching and graph rewriting. At present, few graph transformation tools support multi-labelled graphs. To our knowledge, there is no tool that support inexact graph matching for the purpose of graph transformation. Also, the main problem of these tools lies on the limited expressiveness of rewriting rules used, that negatively reduces the range of application scenarios to be modelling and/or negatively increase the number of rewriting rules to be used. In this paper, we present the tool GMTE - Graph Matching and Transformation Engine. GMTE handles directed and multi-labelled graphs. In addition, to the exact graph matching, GMTE handles the inexact graph matching. The approach of rewriting rules used by GMTE combines Single PushOut rewriting rules with edNCE grammar. This combination enriches and extends the expressiveness of the graph rewriting rules. In addition, for the graph matching, GMTE uses a conditional rule schemata that supports complex comparison functions over labels. To our knowledge, GMTE is the first graph transformation tool that offers such capabilities.

Logical Mapping: An Intermedia Synchronization Model for Multimedia Distributed Systems

The preservation of temporal dependencies among different media data, such as text, still images, video and audio, and which have simultaneous distributed sources as origin, is an open research area and an important issue for emerging distributed multimedia systems, such as Teleimmersion, Telemedicine, and IPTV. Although there are several works oriented to satisfy temporal dependencies in distributed multimedia systems, they are far from resolving the problem. In this paper we propose a logical synchroniza- tion model able to specify at runtime any kind of temporal relationship among the distributed multimedia data involved in a temporal scenario. The synchronization model is based on a new concept that we call logical mapping. A logical mapping, in general terms, translates a temporal relation based on a timeline to be expressed according to its causal dependencies. The logical mappings allow us to avoid the use of global references, such as a wall clock and shared memory. We note that the proposed intermedia synchronization model does not require previous knowledge of when, nor of how long, the media involved of a temporal scenario is executed. Finally, in order to show the viability of the proposed model, a syncrhonization approach is presented.

Multi-Session Key Management Scheme for Multimedia Group Communications

The Internet 2 deployment introduces new capabilities, such as multi-party collaboration, high-scale multimedia assembly and multicast communication. For this reason, the research concerning security is facing new challenges. One such challenge is to create secure multi-session frameworks to ensure the confidentiality of exchanged information. In a multi-session environment, several users are joined at two or more work sessions simultaneously. The confidentiality in these environments can be achieved using cryptographic methods. Unfortunately, the key management, necessary for such environments, creates two main problems: a high complexity in key distribution and a high storage cost. In this paper, we propose an efficient multi-session key management mechanism for dynamic multimedia group communication. Our solution proposes a functional architecture that exploits the overlapping of the user sessions to reduce the redundancy in key distribution. The proposed key management makes use of two key generation strategies: a key derivation technique to reduce the rekey overhead and a pseudorandom number generator that allows the users to generate an independent key per cipher packet.

An efficient multi-channel distributed coordination protocol for collaborative engineering activities

The advances in several computing fields, such as: networking (topology, bandwidth, etc), powerful data processing, and storage, have enabled in the last years, a great progress in collaborative technology. This collaborative technology is used as a support for advanced collaborative applications [1] such as distributed engineering. The distributed engineering applications consider activities that involve geographically distributed working groups, which interact into work sessions according to a preset or improvised planning. Such activities have been analyzed during the European Distributed System Engineering (DSE) project [2] where we made use of collaboration scenarios, which involved participants distributed in three sites, at Turin, Munich and Paris. During these sessions, participants needed to review design issues in collaboration with remote partners. One of the main problems that we found is to synchronously use, in a consistent manner, distributed engineering components and Conferencing GroupWare, such as the collaborative scenario depicted in Fig. 1a. This figure shows a scenario involving four participants using an Engineering CAD tool to share design data, a Conferencing GroupWare to discuss and comment design diagrams, and a Collaborative authoring tool to produce and review design documents. In this scenario, the participants that interact through the Conferencing Groupware component and Engineering CAD tool must have the same consistent view with respect to the shared design data in order to have a coherent discussion with the ensemble of participants. Otherwise, the comments about the design data would not make sense. This is a simple example of inter-component dependencies that must be satisfied in a consistent manner (Fig. 1b). In this paper we propose a new fully distributed coordination protocol that avoids erroneous collaboration scenarios in distributed engineering components and applications. To achieve global coherence, our approach maintains two consistency levels: at a channel level and at an interchannel level. In this way, we guarantee consistency for monolithic applications or components that use a unique communication channel, and consistency between components and applications that use multichannel communication. This approach has been adopted as a result of the experience of the DSE project.

An efficient validation approach for quasi-synchronous checkpointing oriented to distributed diagnosability

Quasi-synchronous checkpointing algorithms are classified into: SZPF, ZPF and ZCF.We propose a validation approach to detect the previously mentioned properties.We model an algorithm execution into the HBR graph and IDR graph.We designed two sets of validation rules to work over the HBR graph and IDR graph.A lower cost is achieved using the IDR graph which is the minimal causal graph. The autonomic computing paradigm is oriented towards enabling complex distributed systems to manage themselves, even in faulty situations. The diagnosability analysis is a priori a study through which a system can be self-aware about its current state. It is from the determination of a consistent state that a system can take some action to repair or reconfigure itself. Nevertheless, in a distributed system it is hard to determine consistent states since we cannot observe simultaneously all the local variables of different processes. In this context, the challenge is to efficiently monitor the system execution over time to capture trace information in order to determine if the system accomplishes both functional and non-functional requirements. Quasi-synchronous checkpointing is a technique that collects information from which a system can establish consistent snapshots. Based on this technique, several checkpointing algorithms have been developed. According to the checkpoint properties detected and ensured, they are classified into: Strictly Z-Path Free (SZPF), Z-Path Free (ZPF) and Z-Cycle Free (ZCF). Generally, the method adopted for the performance evaluation of checkpointing algorithms involves simulation. However, few works have been designed to validate their correctness. In this paper, we propose an efficient validation approach based on a graph transformation oriented towards the automatic detection of the previously mentioned properties. To achieve this, we took the vector clocks resulting from an algorithm execution, and we modeled them into the happened-before graph and the immediate dependency graph (which is the minimal causal graph). Then, we designed a set of transformation rules to verify if in these graphs, the algorithm is exempt from non-desirable patterns, such as Z-paths or Z-cycles, according to the case.

A Graph Transformation-Based Approach for the Validation of Checkpointing Algorithms in Distributed Systems

Autonomic Computing Systems are oriented to prevent the human intervention and to enable distributed systems to manage themselves. One of their challenges is the efficient monitoring at runtime oriented to collect information from which the system can automatically repair itself in case of failure. Quasi-Synchronous Check pointing is a well-known technique, which allows processes to recover in spite of failures. Based on this technique, several check pointing algorithms have been developed. According to the checkpoint properties detected and ensured, they are classified into: Strictly Z-Path Free (SZPF), Z-Path Free (ZPF) and Z-Cycle Free (ZCF). In the literature, the simulation has been the method adopted for the performance evaluation of check pointing algorithms. However, few works have been designed to validate their correctness. In this paper, we propose a validation approach based on graph transformation oriented to automatically detect the previous mentioned check pointing properties. To achieve this, we take the vector clocks resulting from the algorithm execution, and we model it into a causal graph. Then, we design and use transformation rules oriented to verify if in such a causal graph, the algorithm is exempt from non desirable patterns, such as Z-paths or Z-cycles, according to the case.

A Delayed Checkpoint Approach for Communication-Induced Checkpointing in Autonomic Computing

Although the initiative of Autonomic Computing was introduced a dozen years ago, several challenges remain open. One of these challenges is the efficient monitoring at runtime oriented to the detection, diagnosis, and repair of problems that result from failures or bugs in software and/or hardware components. For this purpose, Communication-induced Checkpointing (CIC) can be a useful tool. Communication-induced Checkpointing has been used to attack a wide range of problems that arise in distributed systems, such as rollback recovery, software debugging and software verification, among others. In CIC algorithms, an autonomic component (process) asynchronously cooperates by exchanging information on the application messages about saved local states called checkpoints. CIC aims to form global consistent snapshots by grouping checkpoints (one by each component) in a non-coordinated way. To achieve this, CIC solutions continuously monitor the exchanged control information to identify possible dangerous checkpointing patterns. When a dangerous pattern is identified, it is broken by locally triggering a forced checkpoint. Nevertheless, as we will show, not all forced checkpoints triggered by current solutions are necessary. In this paper, we present a delayed checkpoint approach suitable for autonomic computing that reduces forced checkpoints by establishing certain triggering rules that we call safe checkpoint conditions. Finally, some results are presented which show that our proposal is more efficient than other current solutions.

Data Alignment for Data Fusion in Wireless Multimedia Sensor Networks Based on M2M

Advances in MEMS and CMOS technologies have motivated the development of low cost/power sensors and wireless multimedia sensor networks (WMSN). The WMSNs were created to ubiquitously harvest multimedia content. Such networks have allowed researchers and engineers to glimpse at new Machine-to-Machine (M2M) Systems, such as remote monitoring of biosignals for telemedicine networks. These systems require the acquisition of a large number of data streams that are simultaneously generated by multiple distributed devices. This paradigm of data generation and transmission is known as event-streaming. In order to be useful to the application, the collected data requires a preprocessing called data fusion, which entails the temporal alignment task of multimedia data. A practical way to perform this task is in a centralized manner, assuming that the network nodes only function as collector entities. However, by following this scheme, a considerable amount of redundant information is transmitted to the central entity. To decrease such redundancy, data fusion must be performed in a collaborative way. In this paper, we propose a collaborative data alignment approach for event-streaming. Our approach identifies temporal relationships by translating temporal dependencies based on a timeline to causal dependencies of the media involved.