Imene Lahyani

QoS Monitoring and Analysis Approach for Publish/Subscribe Systems Deployed on MANET

In this paper, we propose a decentralized and adaptive approach for QoS aware publish/subscribe systems deployed on MANET. It provides monitoring and analysis modules acting at the middleware layer to detect QoS degradations between brokers. The paper presents a statistical method allowing for a permanent measurement of QoS parameters and an instantaneous detection of link failures. The statistical method used resorts to the Extreme Values Theory (EVT) and particularly uses the Gaussian and Gumbel distributions to approximate empirical distributions of the measured QoS parameter. Simulation results show the efficiency of the proposed approach.

Analytical Framework for QoS Aware Publish/subscribe System Deployed on MANET

In this paper, we propose an analytical framework for monitoring and analyzing QoS of publish/subscribe systems on MANET. Our framework copes with intermittent connectivity frequently occurring in MANET and provides statistical methods allowing detecting QoS degradations affecting links between brokers at the middleware layer. Besides, our analytical framework identifies QoS degradation source which enables to repair the system. The proposed framework is extensively evaluated with simulation experiments. Simulations results show its efficiency.

Predictive Schemes for QoS Awareness of Publish/Subscribe Systems on MANET

In this paper, we propose a failure prediction methodology for quality of service (QoS) degradation prediction for publish/subscribe systems on MANET. Our propose is to use the Auto Regressive Integrated Moving Average (ARIMA) method to predict failure occurrence in the system and to provide optimal QoS provision of applications. Besides, our forecasting algorithm looks for the source behind QoS degradation using the Correlation method. Simulations results are performed to prove the efficiency of the proposed approach. A comparison is done proving that our proposal outperforms the Auto Regression (AR) based prediction approach.

Towards Self Healing Publish/Subscribe System on MANET

This paper presents a novel approach for Quality of Service aware publish/subscribe systems on Mobile Adhoc Network. Crisis management application seems to be a relevant case study to our approach. On the one hand, our proposed solution monitor the system and analyzes his state towards predicting Quality of Service degradations. Real time prediction is based on the Auto Regressive Integrated Moving Average Formula. On the other hand, our solution defines new reconfiguration actions to be applied once Quality of Service degradation are predicted. In order to evaluate the proposed reparation actions, we define a protocol describing different transformation rules for adaptation using graphs and graph grammar. The Graph Matching and Transformation Engine(GMTE) is used to apply rules to the publish/subscribe system.

Experimentations for QoS Evaluation of Publish/Subscribe Systems Deployed on Ad-hoc Network

Publish/subscribe (P/S) systems have been widely studied and applied in wired networks. However, their deployment on mobile ad hoc networks still presents a lot of challenges especially when there are quality of service (QoS) requirements. In this paper, we approach these challenges by looking first at network routing protocols. We present experimentations that allow to study the reaction of these protocols against QoS degradation that occur at the middleware layer. Based on an empirical study, we show that the routing layer is generally not aware about QoS degradations and so it does not react when they occur. Thus, a reaction in the middleware layer have to be developed in order to achieve a reasonable performance.

Towards Accurate Medical Data in Mobile Health Applications

In this paper, we propose to employ a statistical prediction model to assess medical data accuracy. To this end, the purpose of this paper is to perform medical data prediction using the Auto Regressive (AR) model. The experimental results which prove the efficiency of the proposed approach are reported based on three performance criteria namely Root Mean Square Error, Mean Absolute Error and the Theil Inequality Coefficient.

Analytical decisional model for latency aware publish/subscribe systems on MANET

We propose an analytical model for latency aware publish/subscribe systems on MANET.We combine both proactive and reactive statistical analysis.Reactive analysis detects failures by approximating latency series with Gumbel distribution.Proactive analysis forecasts failures relying on AR or ARIMA statistical methods.Simulations scenarios proved the efficiency of the proposed scheme. This paper proposes an analytical model for latency aware publish/subscribe systems on mobile ad hoc networks. The proposed approach combines both proactive and reactive statistical analysis. On the one hand, the reactive analysis, suitable for multimedia applications, detects failures by approximating latency series with the Gumbel distribution. On the other hand, the proactive analysis, suitable for crisis management applications, forecasts failures occurrence relying on Auto Regression or Auto Regressive Integrated Moving Average Formulas. Finally, a hybrid analysis was proposed by dynamically switching from reactive to predictive forms of analysis whenever quality of service violations are noticed. In order to extract failure cause, we refer to the correlation method once failure was detected or predicted. Simulations done under different scenarios proved the efficiency and accuracy of the proposed scheme.

Failure Prediction for Publish/Subscribe System on MANET

In this paper, we treat the quality of service challenge for publish/subscribe system deployed in MANET. The proposed approach provides monitoring and proactive analysis at the middleware layer for predicting QoS degradations of logical links between brokers. Failure prediction is performed using a statistical method called Auto Regression (AR) which allows to analyze the monitored data in order to avoid failures occurrence. The simulation results which prove the efficiency of the proposed approach are reported.

PCA-based multivariate anomaly detection in mobile healthcare applications

Real time mobile Health applications highly depend on sensor readings to provide high-quality health services. However, real-time sensor readings may be inaccurate and cause abnormal physiological measurements due to internal and external factors. Thus, abnormal readings have a significant impact on the reliability of such applications and consequently affect the patients life. This paper addresses the following issue by proposing a robust approach for online detection of abnormal medical measurements. The proposed approach is based on robust Principal Component Analysis (PCA) to analyze collected physiological measurements from sensors and detect the occurrence of multivariate anomalies based on squared prediction error at runtime. We apply our proposed approach on real medical dataset. Our simulation results prove the effectiveness of our approach in achieving good recall with a low false alarm rate. The reduced time and space complexity of our approach make it useful and efficient for real time settings.

Qos-Driven Architectural Mining for Publish/Subscribe Systems Deployed on MANET

Abstract An important issue in distributed systems is to improve the information dissemination, especially in the Publish/Subscribe systems deployed in mobile ad hoc networks (MANET). In fact, the achievement of this goal needs to take into account various problems such as the failure of the communication system and the degradation of the quality of service (QoS). We focus on QoS problem, we address possible solutions to ensure efficient and reliable communication systems. We propose reconfiguration actions rules which are applicable to a network having QoS degradation. We use graphs to model the distributed system and graph transformation engine (GMTE) to execute rules of reconfiguration. The generated solution is selected according to various evaluation criteria in order to satisfy the requirements of system in terms of QoS.