Christian Fraboul

Methods for bounding end-to-end delays on an AFDX network

Architectures of avionics networks, such as that of the Airbus A380, currently know important evolutions. This is principally due to the increase in the complexity of the embedded systems, in term of rise in number of integrated functions and their connectivity. The evolution of switched Ethernet technologies allows their implementation as an avionics architecture (AFDX: avionics full duplex switched Ethernet). The problem is then to prove that no frame is lost by the network (no switch queue will overflow) and to evaluate the end-to-end transfer delay through the network. The objective of this paper is to present and shortly compare three methods for the evaluation of end-to-end delays: network calculus, queuing networks simulation and model checking

Improving the Worst-Case Delay Analysis of an AFDX Network Using an Optimized Trajectory Approach

Avionics Full Duplex Switched Ethernet (AFDX) standardized as ARINC 664 is a major upgrade for avionics systems. The mandatory certification implies a worst-case delay analysis of all the flows transmitted on the AFDX network. Up to now, this analysis is done thanks to a tool based on a Network Calculus approach. The more recent Trajectory approach has been proposed for the computation of worst-case response time in distributed systems. This paper shows how the worst-case delay analysis of the AFDX can be improved using an optimized Trajectory approach. The Network Calculus and the Trajectory approaches are compared on a real avionics AFDX configuration. Moreover, an evaluation of an upper bound of the pessimism of each approach is proposed.

A Probabilistic Analysis of End-To-End Delays on an AFDX Avionic Network

AFDX (Avionics Full DupleX Switched Ethernet, ARINC 664) developed for the Airbus A380 represents a major upgrade in both bandwidth and capability. Its reliance on Ethernet technology helps to lower some implementation costs, but guaranteed service presents challenges for system designers. An analysis of end-to-end transfer delays through the network is required in order to determine upper bounds. In this paper, we propose to compute probabilistic upper bounds for end-to-end delays on avionic flows. Such upper bounds can be exceeded with a given probability p, and are relevant in the context of avionics, where functions are designed to give accurate results even if they miss some frames. The stochastic network calculus approach analytically determines a probabilistic upper bound, whereas the simulation approach gives an experimental upper bound. The former may be used for new certification needs since it assures that the probability of exceeding the computed upper bound is not greater than p. The latter closely approximates actual network behavior and can help to give some idea of the pessimism of the stochastic network calculus upper bound. The two approaches have been developed in the context of an industrial AFDX network configuration.

Applying and optimizing trajectory approach for performance evaluation of AFDX avionics network

AFDX (Avionics Full Duplex Switched Ethernet) standardized as ARINC 664 is a major upgrade for avionics systems. But network delay analysis is required to evaluate end-to-end delays upper bounds. The Network Calculus approach, that has been used to evaluate such end-to-end delay upper bounds for certification purposes, is shortly described. In this paper we present how the Trajectory approach can be applied to an AFDX avionics network. Moreover we explain how this approach can be optimized in this context. We show that, on an industrial configuration, it outperforms existing end-to-end delays upper bounds.

A method of computation for worst-case delay analysis on SpaceWire networks

SpaceWire is a standard for on-board satellite networks chosen by the ESA as the basis for future data-handling architectures. However, network designers need tools to ensure that the network is able to deliver critical messages on time. Current research only seek to determine probabilistic results for end-to-end delays on Wormhole networks like SpaceWire. This does not provide sufficient guarantee for critical traffic. Thus, in this paper, we propose a method to compute an upper-bound on the worst-case end-to-end delay of a packet in a SpaceWire network.

Applying Trajectory approach with static priority queuing for improving the use of available AFDX resources

AFDX (Avionics Full Duplex Switched Ethernet) standardized as ARINC 664 is a major upgrade for avionics systems. The mandatory certification implies a worst-case delay analysis of all the flows transmitted on the AFDX network. Up to now, this analysis is done thanks to a tool based on a Network Calculus approach. The more recent Trajectory approach has been proposed for the computation of worst-case response time in distributed systems. It has been shown that the worst-case delay analysis of an AFDX network can be improved using an optimized Trajectory approach. This paper extends this optimized approach with the integration of static priority QoS policies. This extension makes possible to compute the bounds needed for deterministic avionics flows (high priority) when (lower priority) non avionics flows are added. Moreover, the paper provides an analysis of the pessimism of the obtained bounds.

Performance Analysis of a Master/Slave Switched Ethernet for Military Embedded Applications

Current military communication network is a generation old and is no longer effective in meeting the emerging requirements imposed by the next-generation military embedded applications. A new communication network based upon Full Duplex Switched Ethernet is proposed in this paper to overcome these limitations. To allow existing military subsystems to be easily supported by a Switched Ethernet network, our proposal consists in keeping their current centralized communication scheme by using an optimized master/slave transmission control on Switched Ethernet thanks to the Flexible Time Triggered (FTT) paradigm. Our main objective is to assess the performance of such a proposal and estimate the quality-of-service we can expect in terms of latency. Using the Network Calculus formalism, schedulability analysis are determined. These analysis are illustrated in the case of a realistic military embedded application extracted from a real military aircraft network, to highlight the proposals ability to support the required time constrained communications.

Improving end-to-end delay upper bounds on an AFDX network by integrating offsets in worst-case analysis

AFDX (Avionics Full Duplex Switched Ethernet) standardized as ARINC 664 is a major upgrade for avionics systems. The mandatory certification implies a worst-case delay analysis of all the flows transmitted on the AFDX network. Up to now, this analysis is done thanks to a tool based on the Network Calculus approach. This existing approach considers that all the flows transmitted on the network are asynchronous and it does not take into account the scheduling of output flows done by each end system. The main contribution of this paper is to extend the existing Network Calculus approach by introducing the offsets associated to the different periodic flows into the computation. The resulting approach is evaluated on an industrial AFDX configuration with an existing offset assignment algorithm. The obtained upper bounds are significantly reduced.

CAN-Ethernet architectures for real-time applications

Embedded systems have specific real-time requirements that led to the development of dedicated communication protocols. Such systems must face increasing communication needs and the evolution of switched Ethernet architecture. But moving from existing dedicated field-busses architectures to new Ethernet based architectures is not always feasible, due to industrial constraints. In this paper, we compare different solutions for integrating existing data busses (such as CAN, which is an important standard in automotive context) on a global architecture that respects increasing bandwidth requirements. In a first step, we study classical CAN/CAN bridging strategies. In a second step, we propose CAN/Ethernet bridging strategies that respect the real time behavior of CAN End System when communicating through an Ethernet network that can be shared by (non CAN) applications

Modelling and simulation of an avionics full duplex switched Ethernet

Architectures of embedded avionics networks, such as that of the future Airbus A380, currently know important evolutions. This is principally due to the increase in the complexity of the embedded systems, in terms of the rise in the number of integrated functions and their connectivity. The evolution in the technologies of data transmission local area networks makes it possible to convey new answers to airframe manufacturers. In particular, switched Ethernet technologies are being implemented as an avionics architecture, even if the non-deterministic character of certain networks must be compensated by strong assumptions. The paper presents a simulation model for dimensioning an avionics full duplex switched Ethernet network (AFDX) and for studying the performance of this network. Comparison with other deterministic studies which use network calculus is also performed.