Mohamed Khalgui

Reconfigurable Coordination of Distributed Discrete Event Control Systems

Dynamic reconfigurability is receiving more and more attention from both academy and industry, which means the ability to flexibly modify system functions by adding/removing hardware/software components, modifying logic relation between components, or updating particular system data at runtime without sacrificing the system performance. A distributed reconfigurable discrete event control system (DRDECS) is composed of several networked reconfigurable subsystems. In order to realize system functions, these reconfigurable subsystems communicate and coordinate with each other, since any casually reconfiguration applied to a subsystem may cause risks to others, or even to the safety of the whole system. This brief proposes a new coordination method for a DRDECS, where each subsystem is modeled by a reconfigurable timed net condition/event system. A virtual coordinator together with a communication protocol between it and subsystems is developed in order to achieve two aims: 1) to coordinate subsystems with an optimal coordination solution using judgement matrices while multiple subsystems require global reconfigurations and 2) to reduce exchanged messages between the coordinator and these subsystems. Furthermore, for the purpose of checking functional and temporal properties of a DRDECS with this virtual coordinator, a computation tree logic-based model checking method is applied. Finally, a hypothetic manufacturing plant is used as a running example to illustrate this brief.

Dynamic Low-Power Reconfiguration of Real-Time Systems With Periodic and Probabilistic Tasks

This paper deals with the dynamic low-power reconfiguration of a real-time system. It processes periodic and probabilistic tasks that have hard/soft deadlines corresponding to internal/external events. A runtime event-based reconfiguration scenario is a dynamic operation allowing the addition/removal of the assumed periodic/probabilistic tasks. Thereafter, some tasks may miss their hard deadlines and the power consumption may increase. In order to reconfigure the system to be feasible, i.e., satisfying its real-time constraints with low-power consumption, this research presents a software-agent-based architecture. An intelligent agent is developed, which provides four solutions to reconfigure the system at runtime. For these solutions, in order to reconfigure the probabilistic tasks to be feasible, the agent modifies their temporal parameters dynamically; moreover, in order to feasibly serve the probabilistic tasks and reduce the systems power consumption, the agent provides three virtual processors by dynamically extending the periods of the periodic tasks. A simulation study verifies the effectiveness of the agent.

Deadlock Prevention for a Class of Petri Nets With Uncontrollable and Unobservable Transitions

This study deals with deadlock problems in a system with uncontrollable and unobservable events, which is modeled with a subclass of Petri nets. We design monitors to enforce constraints for minimal siphons in a Petri net model. A key point of this paper is how to select the input and output arcs of a monitor when there exist uncontrollable and unobservable transitions. After the enforcement of constraints, an iterative algorithm is developed to obtain a controlled system with liveness. A sufficient and necessary condition is established to decide the existence of a monitor enforcing a constraint in a plant with uncontrollable and unobservable transitions.

An optimised simulation of component-based embedded systems in manufacturing industry

This paper deals with an optimised simulation of embedded systems following the international standard IEC61499 in which a Function Block is an event triggered component owning data and an application is a network of blocks that have to satisfy properties described in user requirements. We design these components in different abstraction levels of a hierarchy to control the development complexity. To achieve a feasible simulation, we formalise at first-time the FB concept by defining the Control Function Module (CFM) and the Virtual Plant Module (VPM). We propose thereafter a technique processing temporal deadlines for control modules and we define in the system behaviour critical scenarios where properties can be violated. We apply the simulation around these scenarios by applying a fault injection technique to optimise the system validation and to consider its hierarchy, we propose a simulator architecture based on the master-slave model where a slave injects faults in each level of the hierarchy, whereas the master chooses the best values of these errors to generate critical scenarios.

R-Node: New Pipelined Approach for an Effective Reconfigurable Wireless Sensor Node

The pipeline of reconfiguration is a novel approach that gives a decision making mechanism for any system of tasks when several reconfiguration scenarios are required. These scenarios represent the run-time operations that allow automatic functional modification of the system behavior. In the case of their frequent income, the pipeline follows a specific set of steps in order to figure out the scenarios that need to be applied without altering its own performance. The steps are represented by a sequence of modules. The first module checks the availability of the resources needed by the tasks. The second maintains the stability by rejecting any insupportable flow of software reconfigurations that can destabilize a system. The third module checks the availability of memory for each software reconfiguration request, whereas the fourth verifies the availability of energy for the execution of a new system software configuration. The last module checks the schedulability of this configuration by changing the priorities of the finally accepted tasks in order to reduce their response and blocking times, and to meet their deadlines. The feasibility of the reconfiguration through this approach allows its implementation in wireless sensor nodes. Known by their important requirements in terms of memory and energy, this proposition is adequate when it comes to guaranteeing a maximum lifespan of the nodes. Therefore, the pipeline of reconfiguration is integrated as a middleware within the software architecture of a node. The functionality of this middleware is coordinated by the mean of a set of threads that manage the communication between the pipeline of reconfiguration and the operating system. A tool has been developed within LISI Laboratory to encode this proposition and show its impact on a large range of real devices under diversified scenarios of reconfigurations.

New Power-Oriented Methodology for Dynamic Resizing and Mobility of Reconfigurable Wireless Sensor Networks

This paper deals with reconfigurable wireless sensor networks (RWSNs) that should be adapted to their environment in order to minimize the energy consumption during the communication among nodes and consequently to maximize the lifetime of the network as much as possible. An RWSN is assumed to be composed of a set of nodes located in distributed zones such that each node executes reconfigurable software tasks to control local sensors. We propose in a previous work a zone-based multiagent architecture for an RWSN where a communication protocol is well-defined to optimize distributed reconfigurations. This architecture combines all possible reconfiguration forms of the network to be adapted to its environment under energy constraints. With the multiagent architecture we gain in term of energy to be consumed by each node in the network. After applying a set of reconfiguration scenarios, the total charge of the network decreases. In order to preserve this charge as much as possible before the next recharging operation, in this particular study, we tend to resize geographically the zones in an RWSN. We propose also to move the mobile nodes in order to increase the network lifetime. A new run-time power oriented methodology is reported, which manages the zones and their mobile nodes in order to control the energy consumption. Two dynamic solutions are applied: 1) the resizing of zones and 2) the mobility of nodes. For the new proposed methodology, we present a set of equation systems to model the resizing of zones and the mobility of nodes. The solution of these systems allows us to conclude that we can conserve more energy during the communication among network elements (nodes, stations, and agents). The contributions are applied to a case study that we simulate with the reconfigurable wireless network environment1 to analyze the research originality.1http://lisi-lab.wix.com/rwinproject

Multiagent Framework for Smart Grids Recovery

Failure propagation in smart grids (SGs) complicates and prolongs the recovery time as the faults to be resolved increase. This paper presents the design and implementation of a framework for SGs modeling, simulation, and recovery. The proposed approach is based on a multiagent system composed of static and mobile agents to ensure local and remote resolutions. The efficiency and completeness of the power system recovery protocol are proved as all the existing solutions are analyzed. The use of fault classification and a performing communication process as well as the deployment of local distributed databases updated at run-time ensure the effectiveness of the proposed fault recovery strategy. An experimental study confirms and validates the expected results.

New Optimal Solutions for Real-Time Reconfigurable Periodic Asynchronous Operating System Tasks with Minimizations of Response Time

Scheduling tasks is an essential requirement in most real-time and embedded systems, but leads to unwanted central processing unit (CPU) overheads. The authors present a real-time schedulability algorithm for preemptable, asynchronous and periodic reconfigurable task systems with arbitrary relative deadlines, scheduled on a uniprocessor by an optimal scheduling algorithm based on the earliest deadline first (EDF) principles and on the dynamic reconfiguration. A reconfiguration scenario is assumed to be a dynamic automatic operation allowing addition, removal or update of operating system’s (OS) functional asynchronous tasks. When such a scenario is applied to save the system at the occurrence of hardware-software faults, or to improve its performance, some real-time properties can be violated. The authors propose an intelligent agent-based architecture where a software agent is used to satisfy the user requirements and to respect time constraints. The agent dynamically provides precious technical solutions for users when these constraints are not verified, by removing tasks according to predefined heuristic, or by modifying the worst case execution times (WCETs), periods, and deadlines of tasks in order to meet deadlines and to minimize their response time. They implement the agent to support these services which are applied to a Blackberry Bold 9700 and to a Volvo system and present and discuss the results of experiments.

The embedded control system through real-time task

This paper deals with the study of Embedded Control System through Real-Time Task. In the embedded control system, we decompose a component into several real-time tasks. To ensure safety control of tasks at run-time, we define service and reconfiguration processes for tasks and use the semaphore concept to obtain safety mutual exclusions. The Priority Ceiling Protocol (PCP) is applied as an approach to ensure the scheduling between periodic tasks with precedence and mutual exclusion constraints. We simulate the scheduling of Real-time tasks with PCP through the Cheddar tool.

New Pack Oriented Solutions for Energy-Aware Feasible Adaptive Real-Time Systems

This paper addresses the management of tasks execution for real-time reconfigurable systems powered by battery. In this context, one of major problem concerns the management of battery life between two different recharges. For this type of systems, a reconfiguration scenario means the addition, removal or update of tasks in order to manage the whole system at the occurrence of hardware/software faults, or also to improve its performance at run-time. When such a scenario is applied, the system risks a fatal increase in energy consumption, a violation of real time constraints or a memory saturation. To prevent this type of problems during the execution, a new scheduling strategy is necessary. Our proposal is based on the definition of packs of tasks and the management of different parameters of these packs. For each reconfiguration scenario, modifications will be performed on packs/tasks parameters in order to respect the memory, real-time and energy constraints.