Marc Boyer

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

Comparison of the expressiveness of arc, place and transition time Petri nets

In this paper, we consider bounded Time Petri Nets where time intervals (strict and large) are associated with places (-TPN), arcs (A-TPN) or transitions (T-TPN). We give the formal strong and weak semantics of these models in terms of Timed Transition Systems. We compare the expressiveness of the six models w.r.t. (weak) timed bisimilarity (behavioral semantics). The main results of the paper are : (i) with strong semantics, A-TPN is strictly more expressive than P-TPN and T-TPN; (ii) with strong semantics P-TPN and T-TPN are incomparable; (iii) T-TPN with strong semantics and T-TPN with weak semantics are incomparable. Moreover, we give a classification by a set of 9 relations explained in Fig. 14 (p. 80).

The PEGASE project: precise and scalable temporal analysis for aerospace communication systems with network calculus

With the increase of critical data exchanges in embedded real-time systems, the computation of tight upper bounds on network traversal times is becoming a crucial industrial need especially in safety critical systems. To address this need, the French project PEGASE grouping academics and industrial partners from the aerospace field has been undertaken to improve some key aspects of the Network Calculus and its implementation.

Non preemptive static priority with network calculus

The paper addresses performance analysis of embedded real time networks. We use network calculus to assess the quality of service of a residual flow in a context of aggregation with non preemptive fixed priority scheduling. The main contribution concerns the evaluation of the residual service, given to the low priority flows. In previous works, the effect of the non-preemption seems to be not well taken into account. In particular, when a non preemptive flow is served, it benefits from the full speed from the server, even if, from a long term point of view, it gets only a fractional part. The modeling of this aspect is the starting point of this work1.

NC-maude: a rewriting tool to play with network calculus

Embedded real-time systems are more and more distributed communicating systems. Then, to ensure correctness of application, respect of task dead-line must be ensured, but communication delays must also be bounded. Network calculus is a theory designed to compute such bounds (it have been successfully applied on A380 AFDX backbone). In order to disseminate, and to experiment new results, a tool is needed. Unlike other tools, its purposes are to be open, to allow the user to see the class of function manipulated (sub-additive, star-shaped, concave), the theorems used to get results, etc. To get a code as close as possible to the mathematical context, we chose to use a rewriting language, Maude.

An efficient and simple class of functions to model arrival curve of packetised flows

Network Calculus is a generic theory conceived to compute upper bounds on network traversal times (WCTT -- Worst Case Traversal Time). This theory models traffic constraints and service contracts with arrival curves and service curves. As usual in modelling, the more realistic the model is, the more accurate the results are, however a detailed model implies large running times which may not be the best option at each stage of the design cycle. Sometimes, a trade-off must be found between result accuracy and computation time. This paper proposes a combined use of two simple class of curves in order to produce accurate results with a low computational complexity. Experiments are then conducted on a realistic AFDX case-study to benchmark the proposal against two existing approaches.

Non preemptive static priority with network calculus: enhancement

The paper addresses worst case performance analysis of non preemptive static scheduling priority scheduling within the network calculus theory. Previous studies have been done, each one generalizing some other [8,1,7,3,10], needing weaker hypotheses or improving accuracy of results. This paper presents a very general results, with an accuracy that appear, on preliminary examples, as good as all other one.

Synthesis and Verification of Constraints in the PGM Protocol

Specifications of protocols usually involve several parameters, for example the number of retransmissions or the timeout delays. The properties satisfied by the protocol depend often on the relation between these parameters. Automatic synthesis of such relations becomes a difficult problem when the constraints are too complex, e.g., non-linear expressions between integer and/or real parameters. This paper reports about modeling and constraint synthesis in the Pragmatic General Multicast (PGM) protocol. The property that we aim to satisfy is the full reliability property for data transmission. The complexity of the PGM prevents us from doing automatic synthesis of this constraint. Instead, we propose a methodology to deal with this problem using classical model-checking tools for timed and finite systems. Our methodology consists of several steps. First, we identify the sources of complexity and, for each source, we propose several abstractions preserving the full reliability property. Then, we build an abstract parameterized model on which we test, after instantiation of parameters, that the basic properties of the protocol (deadlock freedom, liveness) are preserved. By analyzing the scenario which invalidate the full reliability property, we find a non-linear constraint between the parameters of the protocol. We check the relation found by instantiating the parameters with relevant values and applying model-checking.

Interconnecting can Busses via an Ethernet Backbone

Abstract: 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 needed integration of switched Ethernet architecture. But moving from existing dedicated fieldbusses architectures to a new Ethernet based architectures is not always easily 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 and preserves the respect of real-time constraints. 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 behaviour of CAN End System when communicating through an Ethernet network that is also shared by (non CAN) applications.

TTCAN over mixed CAN/switched Ethernet architecture

Embedded systems have specific real-time requirements that led to the development of dedicated communication protocols. Such systems often face increasing communication needs and the integration of switched Ethernet architecture. But moving from existing dedicated field-buses architectures to new Ethernet based architectures is not always easily feasible, due to industrial constraints. In this paper, we propose a solution 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. We consider a time triggered CAN application and propose CAN/Ethernet bridging strategies that respect the time-triggered real time behaviour of CAN End System when communicating through a switched Ethernet network that can also be shared by (non CAN) applications