Thierry Divoux

Strict priority versus weighted fair queueing in switched Ethernet networks for time critical applications

In an industrial or technical context, some of the communications are strongly time-constrained. Our objective is to evaluate the service offered by a switched Ethernet architecture. We have chosen the network calculus theory because it gives a good representation of the exchanges and enables to determine the maximum end to end delays. Previous models implementing a strict priority policy presented in (J. -P. Georges et al., 2004) are improved in order to take into account the weighted fair queueing mechanism, which enables to balance the allocation of the network to the different traffics regarding the time contraints they have to respect. This paper also presents the new analytical formulas which upper bound the maximum end-to-end delays of the time-critical messages over the whole network. This enables to determine the optimal bandwith dedicated to both critical and best effort traffics. On an industrial scenario, strict priority and weighted fair queueing policies are finally compared.

A design process of switched Ethernet architectures according to real-time application constraints

Ethernet networks are based on a medium access method which is not deterministic. The use of such networks in factory environments (which are strongly time constraints) can absolutely not guarantee that the applications requirements will be respected. This paper presents a method based on genetic algorithms to minimize end-to-end delays by providing a good distribution of the devices on the network switches. The objective function is defined by using the network calculus which is a deterministic theory and enables to determine bounded delays. In this paper, a case study is described: theoretical results are verified by a real experimentation and compared with results obtained with a network simulator.

IEEE 802.11 Load Balancing: An Approach for QoS Enhancement

With the 802.11 WLAN multimedia applications (Video, Audio, real-time voice over IP,…) increasing, providing Quality of Service (QoS) support becomes very important since the original standard doesn’t take QoS into account. The standard offers access to the wireless users only regarding physical considerations. This can lead to overloaded access points (AP) and considerable degradation of the QoS. This paper deals with this problem. It focuses on the presentation of a QoS management solution for wireless communication systems. It mainly defends that a balanced distribution of mobile stations among the available access points leads to better performances of the Wireless LAN. Some OPNET simulations of the proposed approach are presented to show a better resources allocation and efficiency on QoS metrics. A protocol structure between mobiles and APs is also specified for the implementation of this approach. An SDL description and MSC simulation of this protocol is provided as a first step in its development.

Comparison of switched Ethernet architectures models

The objective of this work is to model a switched Ethernet architecture in order to be able to evaluate the end to end delays. The major interest is to apply these results in an industrial or technical context, where the communications are strongly time-constrained. In this paper, we have chosen the network calculus theory, because it enables a good representation of the exchanges between the equipment which are connected to the network, and also to determine the maximum end to end delays. This paper describes several models of the switching function, in order to identify the better one, regarding some tests achieved by using a network simulation tool.

An approach for end-to-end QoS and network resources management

Abstract In this paper, we address the problem of the end-to-end QoS management for multimedia and real-time applications. We would focus on the presentation of a multi-profile communication environment and its reliability in ensuring end-to-end QoS management in the network communication processes. This communication environment, called Virtual Communication Support (VCS), provides a dynamic assignment of application requirements to the network resources reservation and protocol layers invocation. The advantage of this model is that it performs the capabilities of the network by abstracting the details of all communication stacks-layers that must be invoked in the communication process from application specificities and needs. A great attention is devoted to the end-to-end QoS management in the VCS communication model. We would present our approach for the development of a transport middleware for end-to-end QoS control. This middleware performs the QoS management with an adaptive and predictive algorithm based on the network jitter measurement. The main contribution of this paper is that it provides a solution to predict congestion during communication process and then it can avoid network overloading. This is of several importances when we are streaming multimedia real-time packets. In fact, when real-time application losses some packets it may lead to unfavorable situations. Our approach for the specification and the design of the middleware is based on the use of the Specification and Description Language (SDL). This is a formal description technique that enables model validation extension and reusability. The dynamic behavior of this model is being checked with the Message Sequence Chart (MSC). The simulation and validation of this middleware was performed for the VCS. It was simulated by The Network Simulator (NS). Some profiles with the JAVA language on the Fast-Ethernet Local Area Network. Its testing was performed for a real-time manufacturing process and multimedia supervision for control processes.

Design of the distributed architecture of a machine-tool using FIP fieldbus

In this paper we propose a distributed control system based on FIP fieldbus. It is applied to machine-tool as a replacement for the traditional CNC (Computerized Numerical Controller). The system is composed of a set of microprocessor-based modules (PCs, motion controllers, I/Os, ...) interconnected by FIP real-time network. The main idea is to enable each module to be intelligent improving thus the flexibility and the fault tolerant capability of the whole system. Each module being a sub-control system, accomplishes its own control task some of them for motion control and others for evaluating sensors and regulating actuators. The communication (information exchanges and synchronization) among these modules is ensured by FIP. This system allows both task distribution as well as equipment topological distribution. We discuss some distribution criteria and describe an experimental implementation.

WSN dynamic clustering for oil slicks monitoring

Clustering algorithms for wireless sensor networks are known as application-dependent. In the framework of the oil slicks monitoring in marine environment, this paper describes the modeling of a node implementing a dedicated dynamic partitioning algorithm, using a specific metric for which simulation provides optimal weights for combining various parameters (energy, density, mobility).

Integrating multimedia communications into an MMS environment

This paper presents an approach of network resource management in a heterogeneous manufacturing LANs environment in order to integrate multimedia services. The primary objective is to define a uniform communication interface with which the multimedia functionality can be integrated into the distributed manufacturing applications. The proposed model extends the Virtual Manufacturing Device (VMD) concept defined in the international standard Manufacturing Message Specification (MMS) in order to virtualise the communication systems. A virtual server, named Virtual Communication Support (VCS), is defined to provide the transparent services on the QoS (Quality of Service) parameters negotiation, the communication channel establishment/termination and QoS monitoring in a standardised constrained industrial communications framework. Multimedia application and traditional manufacturing applications have to co-exist in a standardised MMS environment. Furthermore the VCS model provides the support for the possible real-time extension of MMS. A platform of implementation, named PLACOMAP, is presented to illustrate a multimedia enhanced manufacturing application in a heterogeneous communication context where different types of networks, ATM, ETHERNET and MAP are involved.

QoS and network resources management for communication in distributed manufacturing processes

In this paper, we present a multimedia communication model based on the virtual manufacturing device (VMD) concept that has been defined in the international manufacturing message specification (MMS). This model called the virtual communication support (VCS), makes it possible to cohabit different communication techniques, such as ATM and ethernet, provides a multiprofile solution to introduce real-time multimedia applications and manages network resources. Its specification and design used to perform workstation capabilities to avoid network bottleneck. For these purposes, it defines a service set between the application layer and the other communication layers to manage network resources reservation negotiate the quality of service (QoS) and ensure real-time service execution. The design of this communication model is based on the idea of the specification and description language (SDL) use. It is a formal description technique that enables model validation, extension and reusability. The dynamic behavior of this model is being checked with the message sequence chart (MSC).The implementation and the experimentation of this VCS model are being achieved using the JAVA language. It is being done on the fast-ethernet local area network equipped with multimedia and manufacturing devices.

Designing Suitable Switched Ethernet Architectures Regarding Real-Time Application Constraints

Abstract Since the Ethernet network is not deterministic, its use in factory environments (which are strongly time constraints) can not absolutely guarantee that the applications requirements will be respected. This paper presents a method based on genetic algorithms to minimise end-to-end delays, by providing a good distribution of the devices on the network switches. The objective function is defined by using the network calculus, which is a deterministic theory and enables to bound these delays. A case study is described. The results are compared with those obtained by using a network simulator.