Thilo Streichert

Reliability-Aware System Synthesis

Increasing reliability is one of the most important design goals for current and future embedded systems. In this paper, we will put focus on the design phase in which reliability constitutes one of several competing design objectives. Existing approaches considered the simultaneous optimization of reliability with other objectives to be too extensive. Hence, they firstly design a system, secondly analyze the system for reliability and finally exchange critical parts or introduce redundancy in order to satisfy given reliability constraints or optimize reliability. Unfortunately, this may lead to suboptimal designs concerning other design objectives. Here, we presented: a) a novel approach that considers reliability with all other design objectives simultaneously, b) an evaluation technique that is able to perform a quantitative analysis in reasonable time even for real-world applications, and c) experimental results showing the effectiveness of our approach

Modeling and design of fault-tolerant and self-adaptive reconfigurable networked embedded systems

Automotive, avionic, or body-area networks are systems that consist of several communicating control units specialized for certain purposes. Typically, different constraints regarding fault tolerance, availability and also flexibility are imposed on these systems. In this article, we will present a novel framework for increasing fault tolerance and flexibility by solving the problem of hardware/software codesign online. Based on field-programmable gate arrays (FPGAs) in combination with CPUs, we allow migrating tasks implemented in hardware or software from one node to another. Moreover, if not enough hardware/software resources are available, the migration of functionality from hardware to software or vice versa is provided. Supporting such flexibility through services integrated in a distributed operating system for networked embedded systems is a substantial step towards self-adaptive systems. Beside the formal definition of methods and concepts, we describe in detail a first implementation of a reconfigurable networked embedded system running automotive applications.

Dynamic task binding for hardware/software reconfigurable networks

In this paper, a new methodology for tolerating link as well as node defects in self-adaptive reconfigurable networks will be presented. Currently, networked embedded systems need a certain level of redundancy for each node and link in order to tolerate defects and failures in a network. Due to monetary constraints as well as space and power limitations, the replication of each node and link is not an option in most embedded systems. Therefore, we will present a hardware/software partitioning algorithm for reconfigurable networks that optimizes the task binding onto resources at runtime such that node/link defects can be handled and data traffic on links between computational nodes will be minimized. This paper presents a new hardware/software partitioning algorithm, an experimental evaluation and for demonstrating the realizability, an implementation on a network of FPGA-based boards.

Gateway strategies for embedding of automotive CAN-frames into ethernet-packets and vice versa

Todays automotive communication architectures are composed of up to five special purpose communication technologies with dedicated features to interconnect hundreds of comfort-, safety-, and infotainment-related distributed functions. In the future, the number and complexity of such highly distributed functions will further increase and the outcome of this are stronger requirements to the underlying communication architectures. Ethernet and IP, both standardized and widely used in other industrial sectors, could be one solution to handle the upcoming requirements and could homogenize the variety of communication technologies. This paper focuses on a transformation concept to connect a CAN-based network to an Ethernet/IP-based network and vice versa. It highlights several variants which optimize different objectives, like the protocol header overhead or the latency of messages. Moreover, it presents measured results for different network utilizations and a currently deployed automotive CAN subnetwork.

Design space exploration of reliable networked embedded systems

In this paper, a new methodology is presented for topology optimization of networked embedded systems as they occur in automotive and avionic systems as well as wireless sensor networks. By introducing a model which is (1) suitable for heterogeneous networks with different communication bandwidths, (2) modeling of routing restrictions, and (3) flexible binding of tasks onto processors, current design issues of networked embedded systems can be investigated. On the basis of this model, the presented methodology firstly allocates the required resources which can be communication links as well as computational nodes and secondly binds the functionality onto the nodes and the data dependencies onto the links such that no routing restrictions will be violated or capacities on communication links will be exceeded. Due to the often error-prone communication in networks, we allow for routing each data dependency over multiple routes in the networks. With this strategy, our methodology is able to increase the reliability of the entire system. This reliability analysis is based on Binary Decision Diagrams (BDDs) and is integrated in our multi-objective design space exploration. By applying Evolutionary Algorithms, we are able to consider multiple objectives simultaneously during the optimization process and allow for a subsequent unbiased decision making. An experimental evaluation as well as a demonstration of a case study from the field of automotive electronics will show the applicability of the presented approach.

Accuracy of ethernet AVB time synchronization under varying temperature conditions for automotive networks

Todays premium vehicles are equipped with a multitude of Advanced Driver Assistance Systems (ADAS) which present additional driving information or even actively influence the vehicle behavior. These are normally highly distributed real-time systems which process information from distributed sensors like cameras or radar sensors. To fulfiill the high data rates and real-time requirements of these and future systems, it is necessary to provide a suitable underlying network architecture which can handle the communication requirements. Ethernet with its high data rates and Audio Video Bridging (AVB) with its mechanisms for time synchronization as well as traffic shaping could be one solution to interconnect such systems. An essential prerequisite for this is an extensive investigation of Ethernet AVB for automotive applications. This paper concentrates on the evaluation of the accuracy, stability, and robustness of the time synchronization mechanism of AVB under varying temperature conditions and presents measurement results obtained from a test setup with climate chambers.

Formal analysis of the startup delay of SOME/IP service discovery

An automotive network needs to start up within the millisecond range. This includes the physical startup, the software boot time, and the configuration of the network. The introduction of Ethernet into the automotive industry expanded the design space drastically and is increasing the complexity of configuring every element in the network. To add more flexibility to automotive Ethernet networks, the concept of Service Discovery was migrated from consumer electronics to AUTOSAR within the SOME/IP middleware. A network is not fully functional until every client has found its service. Consequently, this time interval adds to the startup time of a network. This work presents a formal analysis model to calculate the waiting time of every client to receive the first offer from its service. The model is able to determine the worst case of a given parameter set. Based on this, a method for calculating the total startup time of a system is derived. The model is implemented in a free-to-use octave program and validated by comparing the analytical results to a timing-accurate simulation and an experimental setup. In every case the worst-case assumption holds true - the gap between the maximum of the simulation and the presented method is less than 1.3%.

Dynamic reconfiguration of FlexRay schedules for response time reduction in asynchronous fault-tolerant networks

In this paper, we present fault-tolerance strategies for implementing passive replication techniques in networked embedded systems based on TDMA-communication such as FlexRay busses. In particular, we assume that processes are replicated at different nodes for tolerating node failures. Hence, if one node fails another node can execute the process and requires the bandwidth for transmitting those messages created by the process over the bus medium. Two concepts are introduced to solve this problem: 1.) to replicate not only the processes but also the messages and to reserve the required bandwidth a priori at design time or 2.) to reconfigure the TDMA-schedule and assign the bandwidth dynamically to the nodes. Obviously, reserving bandwidth for each failure case might lead to a huge overhead and to long response times. Therefore, we provide different reconfiguration strategies for the recently developed FlexRay bus. Moreover, the timing behavior as well as the implementation overhead are evaluated with the help of an experimental setup consisting of five FlexRay nodes.

ReCoNets—Design Methodology for Embedded Systems Consisting of Small Networks of Reconfigurable Nodes and Connections

Automotive, avionic or body-area networks are systems that consist of several communicating control units specialized for certain purposes. Typically, constraints regarding reliability, availability but also flexibility are imposed on these systems. In this chapter, we will present the ReCoNets approach for increasing reliability and flexibility of such systems by solving the hardware/software codesign problem online. A ReCoNet allows to migrate tasks implemented in hardware or software from one node to another. Typically, it consists of a network of communicating Field-Programmable Gate Arrays (FPGAs) and CPUs. Moreover, if a sufficient number of hardware/software resources is not available, the migration of functionality from hardware to software or vice versa is initiated by the system itself. For supporting such flexibility, new design methods as well as services integrated in a distributed operating system for networked embedded systems are revealed. Besides the formal definition of methods and concepts providing several self-x properties such as self-healing, self-adaptiveness and self-optimization, a ReCoNet demonstrator is presented hosting a driver assistance application.

Multi-Objective Topology Optimization for Networked Embedded Systems

In this paper, a new methodology is presented for topology optimization of networked embedded systems as they occur in automotive and avionic systems and partially in wireless sensor networks. By introducing a model which is (1.) suitable for heterogeneous networks with different communication bandwidths, (2.) modeling of routing restrictions and (3.) flexible binding of tasks onto processors, current design issues of networked embedded systems can be investigated. On the basis of this model, the presented methodology firstly allocates the required resources which can be communication links as well as computational nodes and secondly binds the functionality onto the nodes and the data dependencies onto the links such that no routing restrictions will be violated or capacities on communication links will be exceeded. By applying evolutionary algorithms, we are able to consider multiple objectives simultaneously during the optimization process and allow for a subsequent unbiased decision making. An experimental evaluation as well as a demonstration of a case study from the field of automotive electronics shows the applicability of the presented approach