In the context of satellite communication, unique challenges arise, such as long delay links that can impact the network performance also at the Transport Layer. With regard to the Transmission Control Protocol (TCP), congestion control plays a crucial role in managing network traffic and ensuring efficient data transmission over the Internet. This paper wants to give an overview of how TCP congestion control mechanisms adapt to mitigate the effects of long link delays, highlighting the importance of optimizing protocols for these specific environments. Within this context, where signals travel extended distances, traditional TCP algorithms designed for terrestrial networks with relatively low latency, may struggle to maintain optimal performance. As a result, specialized congestion control algorithms, like TCP CUBIC, can be employed to address these challenges. Through the integration of TCP CUBIC with HyStart within a simulation framework based on OMNeT++ and INET, this study sheds light on the importance of tailoring congestion control strategies to suit specific network environments including satellite communication.

Starlink is the largest and most powerful low Earth orbit satellite megaconstellation built to date. Since its debut in 2021, there has been an enormous interest in performance evaluations. https://starlinkstatus.space is a crowdsourced measurement platform based on the Ookla speed test. This paper presents its design, user statistics, and results: Almost 1.7 million measurements from 309 users and 29 countries. The median goodput rates are between 100 Mbit/s and 200 Mbit/s in the forward link (download), and between 10 Mbit/s and 20 Mbit/s in the return link (upload). Goodput rates in North America are lower than in other countries, and the time of day matters, with peak time performance down to a median of 50 Mbit/s. Median round trip times are between 40 ms and 50 ms. Latency under load seems to have improved recently. The hardware version of the standard user terminal does not seem to have an impact on the performance.

This paper aims to solve the challenge of quantifying the perfor- mance of Hardware-in-the-Loop (HIL) computer systems used for data re-injection. The system can be represented as a multiple queue and server system that operates on a First-In, First-Out (FIFO) basis. The task at hand involves establishing tight bounds on end-to-end delay and system backlog. This is necessary to optimise buffer and pre-buffer time configurations. Network Calculus (NC) is chosen as the basic analytical framework to achieve this. In the literature, there are different techniques for estimating arrival and service curves from measurement data which can be used for NC calcu- lations. We have selected four of these methods to be applied to datasets of industrial Timestamp Logging (TL). The problem arises because these conventional methods often produce bounds that are much larger (by a factor of 1000 or more) than the measured maximum values, resulting in inefficient design of HIL system pa- rameters and inefficient resource usage. The proposed approach, called TBASCEM, introduces a reverse engineering approach based on linear NC equations for estimating the parameters of arrival and service curves. By imposing constraints on the equation variables and employing non-linear optimization, TBASCEM searches for a burst parameter estimation which derives tight global delay bounds. In addition, TBASCEM simplifies the run-time measurement pro- cess, supporting real-time data acquisition to evaluate and optimise HIL system performance, and enhancing observability to adapt the HIL configuration to new sensor data. The benefits of TBASCEM are clearly that it enables an efficient performance logging of arrival and service curve parameters and with deriving tighter bounds in HIL systems, compared to evaluated state-of-the-art methods, mak- ing TBASCEM an invaluable tool for optimising and monitoring streaming applications in non-hard-real-time environments.

Paths with high delay and packet loss are problematic for transport protocols, especially when loss-based congestion control algorithms are used. Our topology of interest is a geostationary satellite path (round trip time approx. 600 ms) with additional packet loss, which could be caused by a lossy wireless local area network. In geostationary satellite networks, Performance Enhancing Proxies split TCP connections to improve performance, but this is not possible in case of encrypted transport protocols (VPNs, QUIC). This motivates the use of local loss recovery, i.e., avoiding packet loss on specific path segments. We briefly discuss related concepts, architectures, and protocols. We then evaluate local loss recovery using a TCP tunnel on the lossy link. By means of a simple emulation setup, we demonstrate that local loss recovery has great performance benefits if CUBIC is used. It is also shown that BBR works well for high-delay and lossy paths even without local loss recovery.

Ultra-reliable and low-latency communication has received significant research attention. A key part of this evolution are the Time-Sensitive Networking (TSN) standards, which extend Ethernet with real-time mechanisms. To guarantee high reliability, the standard IEEE 802.1CB-2017 Frame Replication and Elimination for Reliability enables redundant communication over disjoint paths. While this mechanism is essential for time-critical applications, the standard contains some fundamental limitations that can compromise safety. Although some of these limitations have been addressed, none of the previous works provide solutions to these problems. This paper presents solutions to four main limitations of the IEEE 802.1CB-2017 standard. These are 1) choosing match versus vector recovery algorithm, 2) defining the length of the sequence history, 3) setting a timer to reset the sequence history, and 4) dimensioning the burst size in case of link failures. We show how these challenges can be solved by using best- and worst-case path delays of the network. We have performed simulations to illustrate the impact of the limitations and prove the correctness of our solutions. Thereby, we demonstrate how our solutions can improve reliability in TSN networks and propose these methods as guidance for users of the IEEE 802.1CB standard.

Developing and testing automated driving functions on public roads or proving grounds is time-consuming and expensive. Driving in virtual environments is increasingly seen as an eligible approach to counteract these drawbacks. The virtual environment is provided by one or more simulation tools and is coupled with software modules of the automated driving function. This results in a feedback loop that controls a virtual vehicle. Since simulation runs are carried out on off-the-shelf computers, determinism issues may occur due to race conditions. However, determinism is a desirable property in order to be able to reasonably benchmark the automated driving function. This work identifies execution patterns of software modules and shows how they can be synchronized by means of mapping to discrete event semantics. The resulting simulation setup produces repeatable results independent of the available computing and network power. The results are demonstrated using an exemplary driving scenario and compared to paced real-time simulation.

Multipath QUIC is a promising protocol for future multipath communications. One early and often cited Multipath QUIC implementation has been presented by Quentin De Coninck and Olivier Bonaventure in 2017. Their implementation is based on quic-go and results are compared to Multipath TCP Version 0.91, showing performance benefits of Multipath QUIC in various scenarios. In this short paper, we reproduce their experiments and update Multipath TCP to Version 0.92 and later. More recent MPTCP implementations perform much better than Multipath TCP Version 0.91 and thus the performance gain compared to the mentioned Multipath QUIC implementation is not that significant anymore.

Proprietary field buses such as PROFINET IRT, EtherNet/IP, and EtherCAT are currently used to ensure realtime communication in industrial automation appliances. Despite the fact that they are mostly Ethernet-based, interoperability between them is limited or not feasible at all. In the Industry 4.0 context, the goal is to enable real-time communication with ultralow latencies, while also permitting non-critical traffic and ensuring interoperability between devices from different manufacturers. Time-Sensitive Networking (TSN) is a promising candidate to meet these requirements on layer 2 of the OSI reference model in conjunction with a suitable higher-level protocol. The IEEE 802.1 TSN Task Group has been developing extensions to the Ethernet technology to make it more deterministic and reliable since 2012. In this paper, we have evaluated the TSN mechanisms (1) Time-Aware Shaper (TAS) and (2) frame preemption (FP) and compared them to the existing strict priority scheduling (ST) mechanism in exemplary setups. The chosen comparison criteria are latency guarantee, jitter, and the influence on non-critical traffic. Since the support of TSN devices on the market is low, we have carried out the evaluation using the simulation framework OMNeT++ and the NeSTiNg library. Our evaluation shows that, all in all, TAS performs the best, but it introduces a great configuration overhead. FP and ST can keep up in certain scenarios and their performance can even be improved.

The architecture of modern vehicles is shifting from multiple rather loosely collaborating single-core systems to fewer tightly collaborating high-performance multi-core systems. This architectural shift is intended to be capable of managing the increased computational and communication effort that is required to run advanced driver assistance and automated driving software. The most widespread real-time execution model is based on bounded worst-case execution times of tasks, which can result in weaknesses with regard to data determinism in multi-core systems. Therefore, an alternative execution model in which the execution time is abstracted by a logical time has emerged. This execution model proved its feasibility in practical use cases on multi-core systems under the assumption, that data transmission can be performed in zero time. However, this assumption is difficult to meet, especially for distributed realtime systems. In this paper, we propose an execution model in which a logical transmission time is introduced in addition to the logical execution time. The proposed execution model not only allows the prerequisite of zero read and write times to be discarded, but also the scheduling of read and write processes at a fixed point in time is no longer necessary. Consequently, we validate the proposed execution model in a case study and discuss advantages and disadvantages.

The development and testing of automated driving functions in the real world is costly and time-consuming. For this reason, software for automated driving is at first developed and tested in a virtual environment. The provision of the virtual environment requires the coupling of several simulation tools and models. The result is a co-simulation with a feedback loop from the automated driving function to the virtual environment, that does not necessarily exhibit deterministic behavior. However, determinism is a desirable property for developers to be able to analyze the automated driving function and for testers to be able to implement test automation. This paper identifies sources of non-determinism in co-simulation setups for automated driving and proposes a master-less coupling approach that allows repeatable executions. The results are demonstrated using an exemplary driving scenario and compared to a traditional coupling method that uses physical clock synchronization. The presented approach adds some overhead to the simulation modeling and execution, but ensures repeatable simulation results in any case.

Multipath communication with heterogeneous channels is very challenging. In this paper, we describe the combination of terrestrial and satellite internet access links. We review related concepts for such a setup (MPTCP, CMT-SCTP, MPQUIC, Layer 3 Bonding). We propose that dynamically switching a single flow among heterogeneous communication links is a simple and efficient scheduling strategy. Finally, architectures and protocols (MOBIKE, PMIPv6, LISP) are discussed regarding their eligibility for multipath communication and heterogeneous link bonding.

Nowadays factory automation customers require that industrial automation networks are capable of transmitting real-time I/O data and non-time-critical data. PROFINET, one of the most used industrial Ethernet standards available today [1], promises that “everything [can be transmitted] on one cable” [2]. Nevertheless, the standard also recommends that the non-time-critical data sent by a device is limited to 3 kilobytes per millisecond to avoid network overload [3]. In this paper we discuss the need for controlling the non-time-critical data sent to PROFINET networks. Since Microsoft Windows is an often used operating system in industrial automation today [4] we show that its built-in traffic shaping capabilities are not appropriate for industrial automation networks since it is not possible to limit the bursts sent by applications. Thus, we developed our own industrial traffic shaper that is able to limit the burst and the rate per non-time-critical flow. This enables us to apply Network Calculus to obtain guaranteed performance bounds for PROFINET networks that both transmit real-time I/O and non-time-critical data [5]. Both the analysis and the Windows traffic shaper are now provided by the Siemens AG as Siemens Network Planner (SINETPLAN) [6]. Our work is not limited to PROFINET but also applicable to Industrial Ethernet technologies based on IEEE 802.3 such as EtherNet/IP.

Increasingly intelligent energy-management and safety systems are developed to realize safe and economic automobiles. The realization of these systems is only possible with complex and distributed software. This development poses a challenge for verification and validation. Upcoming standards like ISO 26262 provide requirements for verification and validation during development phases. Advanced test methods are requested for safety critical functions. Formal specification of requirements and appropriate testing strategies in different stages of the development cycle are part of it. In this paper we present our approach to formalize the requirements specification by test models. These models serve as basis for the following testing activities, including the automated derivation of executable test cases from it. Test cases can be derived statistically, randomly on the basis of operational profiles, and deterministically in order to perform different testing strategies. We have applied our approach with a large German OEM in different development stages of active safety and energy management functionalities. The test cases were executed in model-in-the-loop and in hardware-in-the-loop simulation. Errors were identified with our approach both in the requirement specification and in the implementation that were not discovered before.

Embedded systems become increasingly complex and distributed. Although there is necessity for thourough testing, exhaustive validiation and verification is hardly possible in industry due to time and resource restrictions. In the past the reason for this has often been that it was to time-consuming to specify, to execute, and to evaluate test cases for the first design models and the integrated embedded system. In the meantime methods have become popular in industry that allow the automated generation, execution, and evaluation of test cases. In order to be able to automate these steps all necessary information must be integrated into the models that are the basis for the following steps. The complexity of the system, however, makes the evaluation and assessment of the behavior of the system even more complex. The growth of information needed for this comes along with it. The integration of this information into the model which is used for the generation of test cases is hardly feasible. In this paper we describe how this issue can be addressed by the combination of reference models in MATLAB/Simulink® with test models. Time Usage Models (TUM) are employed as test models and provide the basis to generate all possible test scenarios. Model based statistical testing with consideration of time and durations is supported by TUMs. The reference models are used like an executable specification, providing information for the evaluation of the system to be tested. The test model can therefore be kept generic in order to be able to derive virtually any test case from the model, taking account of the potentially infinite sequence of inputs reactive systems might process. We applied the presented approach with a german automotive OEM for the validation and verification of the energy management system.

In the automotive industry requirements are often still composed of natural language text, spreadsheets, drawings, and formal models. Models are often used to describe partial aspects from the whole set of requirements. Hence, flaws and vagueness in requirements are common and hard to discover. Upcoming standards like ISO 26262 request the automotive industry to be more strict and formal on the requirements. Formal notation and unambiguitiy is explicitely stated. In the field of system and acceptance testing requirements are the basis for all activities. Although, requirements are often not close to testing. To overcome this we introduced the Timed Usage Model (TUM) as a formal representation of requirements specification. During the creation of the model the requirements are analyzed and brought into an unambiguous and formal representation. Traceability is achieved, as each path in the model must be based upon a requirement. The formulation of the requirements in form of an unambiguous model clarifies the requirements and helps to detect design errors. During the creation of the model omissions and flaws in the requirements are discovered. The model serves as a communication medium when functionality responsibles are involved to clarify these aspects. Timed Usage Models were created for power train functionality and the energy management. Moreover, the model as a formalized representation of the requirements served as the basis for the whole testing process, including test planning, test case generation, and test campaign analysis.

In embedded automotive real-time systems correct timing is required to effectively implement distributed applications, time-triggered communication, and integration of intelligent safety applications. We use formal schedulability analysis to validate the task timing for asynchronous real-time systems in an automotive context. Restricting the computational model to purely time-triggered task activation and applying application specific constraints enables us to perform exact response time analyses without pessimism. The analysis can be done efficiently since our computational model drastically limits the number of critical instant candidates to be considered for the worst-case scenario. Simulation results show that the original computational model incorporates a significant amount of response time overestimation and a high computational workload. The presented method can be applied in early design stages as well as in series development and helps to reduce development risks for electronic control units and embedded telematic systems.

The development of economic and hybrid vehicles is only possible with complex systems to control the energy management and engine. The amount of embedded electronics and software to realize these systems increases continuously, and hence the complexity of the integration. The functional validation and testing of these systems is of increasing importance, as these systems have an increasing impact on other systems like comfort and safety functionalities. So there is a need for thourough testing. However, exhaustive testing is not possible as testing in industry is limited by time and resources. So a method is needed to derive systematically the most significant test cases in order to be able to assess the behavior of the system. We applied test models to accomplish this task. A Timed Usage Model served as a formal requirement specification and was the basis for following test activities. The model provides the possibility to describe timing and data dependencies of the system to be tested. The test planning and test case generation was supported by the models. The appliance of models allowed the systematic generation of test cases and the assessment of the significance of the conducted test activities with respect to the coverage of requirements.

Modern cars comprise a multitude of electronic features which are implemented in tens of communicating control units. To connect these in-car embedded systems, the CAN bus offers a sustainable performance, hence it is used as a widespread communication infrastructure, even for safety critical applications. However, CAN media access is priority based and performed competitive and non-preemptive. Thus, assessing the worst case end-to-end delay is inevitable in order to provide safe and efficient operation of functions with hard real-time properties. In this paper, we use the analytical method of network calculus to determine guaranteed upper bounds for transmission delays of all CAN priorities. We demonstrate the applicability of our approach by investigating current real-life CAN communication data from the German car manufacturer Audi.