December 16–18, 2020 | Virtual Conference
|Wednesday, December 16||Thursday, December 17||Friday, December 18|
| 16:00 ‑ 16:10 (CET)
07:00 ‑ 07:10 (PST)
00:00+1 ‑ 00:10+1 (JST)
|G1: Welcome||S6: Simulation and V2X Applications||G6: poster and Demo Session|
| 16:10 ‑ 16:20 (CET)
07:10 ‑ 07:20 (PST)
00:10+1 ‑ 00:20+1 (JST)
|S1: Cellular V2X|
| 16:20 ‑ 16:30 (CET)
07:10 ‑ 07:30 (PST)
00:20+1 ‑ 00:30+1 (JST)
|16:30 ‑ 16:40 (CET)
07:30 ‑ 07:40 (PST)
00:30+1 ‑ 00:40+1 (JST)
|S2: Radio I|
|16:40 ‑ 16:50 (CET)
07:40 ‑ 07:50 (PST)
00:40+1 ‑ 00:50+1 (JST)
|16:50 ‑ 17:00 (CET)
07:50 ‑ 08:00 (PST)
00:50+1 ‑ 01:00+1 (JST)
|17:00 ‑ 17:10 (CET)
08:00 ‑ 08:10 (PST)
01:00+1 ‑ 01:10+1 (JST)
|17:10 ‑ 17:40 (CET)
08:10 ‑ 08:40 (PST)
01:10+1 ‑ 01:40+1 (JST)
|17:40 ‑ 17:50 (CET)
08:40 ‑ 08:50 (PST)
01:40+1 ‑ 01:50+1 (JST)
|17:50 ‑ 18:00 (CET)
08:50 ‑ 09:00 (PST)
01:50+1 ‑ 02:00+1 (JST)
|G2: break||G4: Panel Discussion|
|18:00 ‑ 18:10 (CET)
09:00 ‑ 09:10 (PST)
02:00+1 ‑ 02:10+1 (JST)
|S3: Radio II|
|18:10 ‑ 18:20 (CET)
09:10 ‑ 09:20 (PST)
02:10+1 ‑ 02:20+1 (JST)
|S8: E/E Architectures and In-Vehicle Communication|
|18:20 ‑ 18:30 (CET)
09:20 ‑ 09:30 (PST)
02:20+1 ‑ 02:30+1 (JST)
|S4: Congestion control|
|18:30 ‑ 18:40 (CET)
09:30 ‑ 09:40 (PST)
02:30+1 ‑ 02:40+1 (JST)
|S9: Machine Learning and Localization|
|18:40 ‑ 18:50 (CET)
09:40 ‑ 09:50 (PST)
02:40+1 ‑ 02:50+1 (JST)
|S5: Cooperative Driving and Sensing|
|18:50 ‑ 19:00 (CET)
09:50 ‑ 10:00 (PST)
02:50+1 ‑ 03:00+1 (JST)
|19:00 ‑ 20:00 (CET)
10:00 ‑ 11:00 (PST)
03:00+1 ‑ 04:00+1 (JST)
|G5: Virtual Hangout|
Abstract: Cellular-V2X aims at supporting road safety and traffic management applications and introduces sidelink (PC5)as a new interface. These applications primarily rely on periodic broadcast of small-sized periodic messages and event-triggered notification messages. A key requirement of the applications is to facilitate uninterrupted exchange of messages among the vehicles irrespective of whether the vehicles are inside a base station coverage or not. In Cellular-V2X standardization, two resource allocation modes have been specified: when a vehicle is located inside the coverage of the cellular network, the corresponding base station manages the transmission resources for the direct message exchange. When a vehicle is outside of cellular coverage,it allocates its resources autonomously. The modes are also referred to as "managed" and "unmanaged" mode, or mode 3and mode 4, respectively. In situations with imperfect cellular coverage by base stations, a vehicle needs to perform mode switching. The present work aims to understand the different phases of the mode switching procedure and to study the latency involved in each phase for both switching scenarios, mode 3 to 4and vice versa. Considering a realistic road traffic scenario with a highway tunnel, we assess the impact of mode switch latency on the data transmission for a selected mode switch strategy.
Abstract: Many vehicular use cases are already based on V2I (Vehicle-to-Infrastructure) and V2N (Vehicle-to-Network) driven applications. The importance of such applications will increase with the upcoming 5G mobile technology because different stakeholders, e.g., the 3rd Generation Partnership Project (3GPP) or the 5G Automotive Association (5GAA), define further use cases in that context. Regarding the evaluation of the application capabilities different Quality-of-Service (QoS) metrics are of interest. In this paper, we modified the recently introduced OMNeT++ framework 5G-Sim-V2I/N for such a use case evaluation. We modelled four different advanced vehicular use cases (video streaming, parallel running applications in several vehicles, remote driving and cooperative perception via V2I) and simulated all of them with different parameter sets in two traffic scenarios. We integrated the 5G HARQ mechanism and ran simulations with different numbers of active processes. The reliability and the jitter were measured for each simulated data packet in downlink and uplink direction. By conducting these experiments it turned out that an increasing number of HARQ processes can lead to a severe impairment of performance. The effects will be illustrated and potential reasons will be identified.
Abstract: It is expected that continuous development of Intelligent Transportation Systems (ITS) will further improve the safety and convenience of road traffic. Cellular V2X (C-V2X), as a next-generation V2X communication technique, has attracted much attention recently. C-V2X supports direct communications with sidelink via the PC5 interface as well as cellular communications via the Uu interface. Vehicles associated with the same operator can use both interfaces to communicate with each other, but the communication between vehicles associated with different operators has not been well studied yet. In this paper, we study how to use C-V2X to efficiently and reliably disseminate traffic event information among vehicles associated with different operators. First, in the basic method, we investigate the combination of sidelink and cellular communications. Next, we propose a sidelink based relay scheme so as to improve the dissemination rate across operators, and improve radio resource efficiency by reducing redundant communications from two aspects. Simulation results confirm that compared with the basic method, the proposed method improves event dissemination rate by up to 38.1% and reduces event duplicate rate and the number of RBs by up to 82.0% and 57.5%, respectively.
Abstract: Cellular networks currently support non-safety-critical Vehicle to Everything (V2X) services with relaxed latency and reliability requirements. 5G introduces novel technologies at the radio, transport and core networks that are expected to significantly reduce the latency and increase the flexibility and reliability of cellular networks. This has raised expectations on the possibility for 5G to support advanced V2X applications, including connected and automated applications such as advanced ADAS services, cooperative driving and remote driving. At the radio access network (RAN), 5G introduces the New Radio (NR) interface that incorporates flexible numerologies and new slot formats, channel coding schemes, and radio resource management processes. Previous studies have reported latency values of 5G NR below 2 ms when considering scenarios with limited users in the cell and with unlimited bandwidth. Supporting advanced V2X services using 5G requires a scalable network capable to support a larger number of users without degrading the required service level in scenarios with potentially limited spectrum. This study advances the current state of the art with the evaluation of the scalability of the 5G NR RAN. As a case study, the paper evaluates the capacity of 5G RAN to support the latency and reliability requirements of the cooperative lane change use case as the network load varies. The results show that the capacity of the 5G RAN to support advanced V2X services depends on the system configuration, network load and service requirements. These results call for a careful design, configuration and planning of 5G networks to support V2X services.
Abstract: "Limited feedback communication systems rely on codebook schemes for informing the base station (BS) with the favorable downlink precoding weights. The shortcoming of those methods is that the codebook size increases as the number of antennas at the BS scales up and/or there is rich scattering. The exhaustive search method proposed in most of the current literature should be replaced with new codebook indexing techniques that can reduce the search complexity without jeopardizing the sumrate performance. This paper presents a novel codebook search scheme which makes use of the spatial correlation structure in massive multi-user multi-input multi-output (MU-MIMO) systems to minimize the set complex operations needed to select the best codeword. Moreover, we derive the closed form analytical expression for the probability of indexing the whole codebook under the presented search scheme. Simulation results show that with the presented method we can achieve a considerable percentage of the sum-rate performance realized by the more complex brute force search method yet with less number of indexed codewords."
Abstract: A key challenge in the domain of Inter-Vehicle Communication (IVC) is to make the best use of limited channel capacity to achieve a continuous exchange of information. In this work, we propose the use of directional radio transmission and reception (known from modern WLAN standards, but following the IEEE 802.11p specifications) to lower the channel utilization. In this work, we use platooning as an example application, since this application offers a well-defined communication topology. Using extensive simulations, we show that directional communication can substantially lower the channel busy ratio to less then half for platoon vehicles and down to less than 1% for non-platoon vehicles. If care is not taken, however, it can drastically increase the probability for packet collisions.
Abstract: Millimeter Wave (mmWave) mobile communication networks are envisioned to provide high performance radio links as required by emerging vehicular applications such as remote driving and massive automotive sensing. However, the high penetration loss and thus a high blockage probability embody the greatest challenges in farming this frequency spectrum. For this reason, the specific utilization of reflecting surfaces, especially so-called reconfigurable intelligent surfaces (RIS) introduce a controlled and focused reflection towards the designated receiver, if the line-of-sight is obstructed. In this work, we present a latest extension to our vehicular mobility and mobile network simulation environment, which allows for coverage analysis, network planning and the evaluation of context-aware beam management strategies. Preliminary results prove, that even in a weak coverage scenario with one base station, the deployment of properly disposed RISs leads to the achievement of predefined coverage goals even with imperfect context information. Based on the outcomes, new rules to ease the placement of RISs for network planning tasks might be derived in future.
Abstract: In this paper, the investigation of the channel characteristics for mmWaves is addressed due to its recent sparked interest in modern communication technologies. Modern intelligent transportation systems require higher data rates and, therefore, more bandwidth, utilization of higher frequencies, which can provide these needs, get more and more in focus in the context of 5G. As part of our current research, the proposed work aims to give a deeper understanding of the channel characteristics and wave propagation at 77 GHz, especially for automotive use cases like platooning or intersection assistant. Typical traffic scenarios, starting with a rural street and proceeding with a highway scene, were reconstructed using the commercial simulator WinProp. Furthermore, the scenarios are evaluated by including additional obstacles, like vehicles and traffic signs. Time-varying environments with driving vehicles with either line-of-sight (LOS) as well as non-LOS (NLOS) connections were investigated. The results reveal significant changes in the channel impulse response for one time step, when, especially relevant in the NLOS case, a new reflection path gets available as scattering and reflection paths have more than 25 dB difference in receiving amplitude. Therefore, the antenna alignment and the environment had a strong influence. Thus, it could be shown that scatters like traffics signs cause power fading for the LOS connection, whereas they benefit the NLOS case, by enabling communication at all. In future research, the channel models will be further addressed and expanded to design a link that is capable of communicating safety-relevant information and hold the high demands on latency.
Lars Wolf is full professor for computer science at the TU Braunschweig and head of the Institute of Operating Systems and Computer Networks. Prof. Wolf is respectively was responsible in other roles such as being department head and speaker of informatics & information technologies centers at TU Braunschweig. Previous positions were with IBM, TU Darmstadt, and Univ. Karlsruhe.
Dr. Wolf served on editorial boards and as chair and member of committees of many international conferences and workshops. His current research interests include wireless networking in general, vehicular networks, sensor networks, internet of things, cyber-physical systems, delay-tolerant networks, and mobile systems in several application areas. The research experience of Lars Wolf on vehicular-related topics lasts back for more than 15 years including inter-vehicular communication, but also autonomous vehicles, e.g., being part of the TU Braunschweig team in the DARPA UrbanChallenge final in 2007.
Abstract: Due to their ubiquity and low use-cost, the opportunistic use of Wi-Fi networks to offload data from moving vehicles is enticing. However, due to their limited coverage and variable performance, choosing what Access Points (APs) to use in order to maximize the amount of data that can be offloaded is challenging. This difficulty is exacerbated by the heterogeneity created by the introduction of new Wi-Fi standards such as 802.11ad, which renders heuristics designed for homogeneous environments, e.g., signal quality, ineffective. In this work we test the hypothesis that historical network performance, indexed by vehicular mobility information, can be used to effectively forecast future network performance, and consequently help select APs for data offloading in a heterogeneous Wi-Fi environment. Our approach was to perform a trace-based analysis on experimental data collected in a realistic vehicular environment. Our results show that a practical algorithm based on data rate forecasting from mobility information was able to transfer at least 80% of the optimal amount of data, under the tested scenarios.
Abstract:In this paper, we evaluate the n-ray ground interference model based on field experiments. The n-ray model is an extension of the popular two-ray interference model for three-dimensional Vehicular Ad Hoc Network (VANET) simulation. Using commodity hardware, we measure the path loss on nine selected test tracks and replay the scenarios simulatively to compare it to the simulated path loss resulting from the n-ray model. The simulation results are in good agreement with the measurements for the most parts in all nine cases, indicating that the n-ray ground interference model is able to capture the effect of ground reflections for arbitrary terrain shapes. The two-ray interference model fails to do so as it neglects elevation information. Moreover, the dependence of the results on the parameterization of the model as well as on the horizontal resolution of the underlying elevation dataset is investigated. In order to ensure that the n-ray model is only applied under Line of Sight (LOS) conditions, we further present a possible combination with our environmental diffraction model.
Abstract: We explore the use of RADar based COMmunication (RADCOM) as a complementary communication technology for next-generation Intelligent Transportation Systems (ITS). RADCOM makes use of vehicular radar operating in the 77 GHz mmWave band. Using platooning of cars on the freeway as an example application, which could benefit substantially from mmWave high bandwidth transmissions with low latencies along the platoon, we propose to combine Full-Duplex Relaying (FDR) with RADCOM. Full-duplex relaying at the 77 GHz band benefits from the directionality of the signal, which leads to reduced Looped Self-Interference (LSI). We developed a real-time simulation model based on GNU Radio to study the mmWave propagation in platoons. In particular, we investigate the opportunities of inband FDR for RADCOM. Our first results clearly indicate the advantages of the proposed approach in terms of reduced physical layer latency and energy requirements while maintaining optimal channel link rates.
Abstract: To address the increasing communication demand driven by envisioned large-scale connected vehicle deployments, the dual use of the 76-81 GHz automotive radar frequency band for joint automotive radar and communication (JARC) system has gained significant interest given the wide and dedicated bandwidth. In this paper, we propose and evaluate a multi-range joint automotive radar and communication system based on pilot-based OFDM waveform. The proposed JARC system exploits the dynamic allocation of pilot subcarriers to switch between radar specifications for different ranges, e.g. short-range radar (SRR), mid-range radar (MRR) and long-range radar (LRR), to minimize the age of information of the covered regions, while dynamically adjusting the communication data rate based on traffic load.
Abstract: In vehicular ad hoc networks, congestion control prevents the overloading of the wireless channel and ensures a fair distribution of the transmission resources. For ITS-G5-based vehicular networks, the European standardization by ETSI has specified a Decentralized Congestion Control (DCC) function at the access layer. This function controls the medium occupancy of a network node by enforcing maximum values of message transmission parameters. In the present paper, we study the impact of DCC on the performance of the collective perception service. This communication service enables vehicles and roadside stations to exchange messages with pre-processed sensor data. Since collective perception can considerably contribute to the network load, the transmission restrictions imposed by DCC affect the performance of the information exchange and the quality of the perception. The current design of collective perception in ETSI does not adapt the messages to the actual DCC constraints. We propose a novel approach for DCC-aware collective perception, which enhances the object filtering process of collective perception by dynamically adapting the message size to the DCC constraints and implicitly the message generation rate. Compared to the current ETSI design, the obtained results show a better quality of perception and channel usage, with a reduced message generation rate.
Abstract: Communications between vehicles and their surroundings can improve the safety of road users. Other vehicles can be identified via radio broadcast messages and the safety in non-line-of-sight situations can be improved. One emerging technology is Cellular Vehicle-to-Everything (C-V2X) communications, which enables various direct communications applications, mainly safety use cases. Broadcast messages are received by all road users in the vicinity, while signal propagation determines the maximum distance over which messages can be decoded. Field tests often use test antennas, set up on the vehicles' roof to test the communications link. While this is suitable for testing general use cases, in a real-life deployment of the technology, the antenna placement will have to comply with design aspects. In this paper we have conducted different field tests with an antenna prototype integrated into the existing shark fin housing. We measured the achievable packet error rates (PER) for different distances and communication directions. The influence of the car on the transmission of radio waves asymmetrically affects the propagation. The maximum distance for reliable communication in C-V2X is influenced by characteristics and placement of the antenna and the shape of the vehicle. Additionally, we could observe a significant impact by panorama glass roofs. Antenna placement will, therefore, be an important factor for the capability of V2X communications.
Abstract: In vehicle-to-everything (V2X) communications, ensuring the scalability is one of the most important aspects of V2X safety applications. This paper analyzes and compares the scalability performance of dedicated short-range communication (DSRC) based on IEEE 802.11p and 3GPP Rel-14 LTE-V2X PC5 mode 4 (i.e., direct communication mode without the help of base stations) in transmitting the Basic Safety Messages using SAE congestion control mechanisms in freeway scenarios with up to 4000 vehicles over a 5-km road. Simulation results show that for V2V distances below 300 m, DSRC over 10-MHz bandwidth is comparable with LTE-V2X over 20-MHz bandwidth with HARQ retransmission. Also, we reveal that LTE-V2X has a consecutive packet loss issue due to its semi-persistent scheduling mechanism. The results show that the use of a wider bandwidth and packet retransmission in LTE-V2X can partly mitigate a consecutive packet loss issue.
Abstract: In vehicle-to-vehicle (V2V) communication, vehicles must periodically broadcast their kinematics information to reduce the chances of collision. When the communication channel utilization becomes high, however, congestion control for the periodic broadcast traffic is inevitable. So far, all standard or proposed congestion control schemes for V2X safety communication have been quality of service (QoS)-blind. Which aspect of the QoS is compromised by a given congestion control scheme has been purely the artefact of the mechanism(s) employed by the scheme, detached from the needs of the safety application running at the time. This paper demonstrates the feasibility of congestion control that allows V2X safety applications to choose a QoS class that preferentially protects the QoS aspect that is more prized for their requirement. And yet, it shows that such congestion control scheme can be engineered to reduce the channel utilization in a uniform manner as before, regardless of the individual choices made by each V2X application.
Abstract: Vehicle platooning can produce significant fuel saving due to reduced aerodynamic drag. In the design of future cooperative driving systems, quantifying such benefits will be of utmost importance, because it will need to be considered in designing, forming and managing platoons. This work aims at developing an Open Source framework for modeling aerodynamics effects. We perform a Computational Fluid Dynamics (CFD) study about the aerodynamics of vehicle platoons, describe a model that exploits the resulting measurements, and implement it inside a road traffic simulator to show how savings can be estimated by means of simulations. Furthermore, we publish the necessary tools required to reproduce the results, enabling further research. Strictly speaking, this work does not deal with vehicular networks, but it contributes to tackle the problem of optimizing the management and control of platoons through proper use of communications, from the point of view of fuel consumption and battery usage.
Abstract: Human drivers have several means to coordinate their maneuvers, which pose challenges for automated vehicles. Standardized means like the turning signal are relatively easy to interpret for an automated vehicle, but interpreting hand signaling is a challenge. Even for human drivers, these signals are sometimes ambiguous. In case of a false interpretation, such a signal can become a cause of a critical situation. With the upcoming of Vehicle-to-Vehicle (V2V) communications automated vehicle got another method for exchanging complex information. This can be used for maneuver coordination between vehicles, but is inaccessible for human drivers. Thus, it is important to consider whether critical situations can be caused by such new means. In previous work we proposed a Maneuver Coordination Protocol (MCP). In this paper, we present a systematic analysis of our proposed MCP and show that it does not introduce additional safety risk. All safety risks presented in this paper already exist without our protocol and our protocol reduces the occurrence probability of these risks.
Abstract: Automated driving is not possible everywhere. Limited by the Operational Design Domain (ODD) of vehicle automation functions, Transitions of Control (ToC) are required. If the ToCs fail, Minimum Risk Maneuvers (MRM) are executed, resulting in stopped vehicles on the road. As a result, traffic is negatively impacted, esp. when the number of automated vehicles (AVs) rises. To reduce such negative impacts, the EU-H2020 TransAID project has designed novel infrastructure-assisted traffic management measures using V2X communications, and evaluated them via simulations and field trials. This paper shows how prototypic real-world tests were performed to validate feasibility of the TransAID measures on public road and test track trials. The obtained results show that infrastructure support and V2X communication can contribute to drastically reduce the need to perform ToCs, MRMs, and hence the risk of blocked roads.
Abstract: This paper proposes the concept of perceived safety to quantitatively evaluate the safety benefits of collective perception to connected road users in safety critical scenarios. It is assumed that connected vehicles and infrastructure network transmit collective perception messages (CPMs) that contain not only basic information about the transmitter, but also key information about other objects detected via its on-board sensors. The detected objects may include non-connected road users that could be either occluded or out of the sensor range of the other connected road users. Our perceived safety framework provides a systematic way to determine when information perceived by a vehicle is sufficient to safely perform a given maneuver based on the sensory set, wireless connectivity, and dynamics of the ego vehicle. We illustrate the perceived safety framework by considering a lane change scenario, and show by simulations that CPMs enhance the ability of connected automated vehicles to perform safe lane changes.
Abstract: Vehicle-to-Everything (V2X) communication has become an integral component of Intelligent Transportation Systems (ITS) due to its ability to connect vehicles, pedestrians, infrastructure, and create situational awareness among vehicles. Cellular-Vehicle-to-Everything (C-V2X), based on 3rd Generation Partnership Project (3GPP) Release 14, is one such communication technology that has recently gained significant attention to cater the needs of V2X communication. However, for a successful deployment of C-V2X, it is of paramount significance to thoroughly test the performance of this technology. It is unfeasible to physically conduct a V2X communication experiment to test the performance of C-V2X by arranging hundreds of real vehicles and their transceiving on-board units. Although multiple simulators based on frameworks such as NS-3, OMNET++ and OPNET have proven to be reliable and economic alternatives to using real vehicles, all these simulators are time-consuming and require several orders of magnitudes longer than the actual simulation time. As opposed to physical field- and simulation-based testing, network emulators can provide more realistic and repeatable results for testing vehicular communication. This paper proposes a real-time, high-fidelity, hardware-in-the-loop network emulator (RVE-CV2X) based on C-V2X mode 4 that can provide scalable, reliable and repeatable testing scenarios for V2X communication. The accuracy of this emulator is verified by comparing it to an already validated C-V2X simulator based on the NS-3 framework.
Abstract: Autonomous vehicles promise a revolution in mobility, enabling safe, resource-efficient urban and inter-urban transport with a high degree of user convenience. To achieve optimal efficiency, autonomous vehicles must be viewed as a network of communicating cyber-physical systems which exchange information to optimize a utility function under strict security and safety requirements. Vehicles can exchange information to extend their perception horizon, exchange driving modes to enhance scene understanding and most importantly cooperate directly with other automated and autonomous vehicles in cooperative driving maneuvers such as platooning. In this paper we present a novel communication protocol built on the current vehicular communication standard IEEE 802.11p, which enables negotiation and execution of cooperative driving maneuvers based on distributed state machines. The main goal of this protocol is to achieve a common synchronized state and common state transitions while supporting fault-tolerance and self-supervision under security and safety constraints. This paper presents the Collaborative Maneuver Protocol (CMP) and provides a formal proof of correctness. We furthermore present an application in a platooning function and provide an evaluation of robustness in regard to packet loss.
Abstract: Prototyping V2X applications in large-scale scenarios is a challenging but demanded task. To address this challenge, this paper presents and evaluates a novel prototyping tool based on OpenAirInterface (OAI). This tool makes use of a new OAI stub interface located at the MAC level that is able to emulate all underlying software and hardware of the LTE V2X technology. The proposed solution delivers packets to the target UEs according to statistical packet loss and delay models. The OAI stub exposes flexible APIs to integrate different models, including different LTE-V2X schedulers, that enable the evaluation of V2X applications under high traffic density scenarios. To showcase the proposed OAI stub interface, this paper also presents two case studies based on two V2X schedulers, the Sensing-based Semi-persistent scheduling (SB-SPS), as well as a Self-organizing TDMA (STDMA) scheduler.
Abstract: In recent years, there has been a growing interest in researching and developing vehicular networking protocols and safety applications that consider the inclusion of Vulnerable Road Users (VRU) in their design. In this work, we present a methodology that enables the inclusion of connected VRU in the well-known VEINS framework. We describe the main problems found during the integration of VRU and the necessary fixes required at the time of launching the simulation environment. The integration contributes to better understand the behavior of both connected vehicles and connected VRU in Cooperative Intelligent Transportation Systems. The methodology has been tested with vehicular scenarios at intersections using DSRC in both vehicles and VRU, but it is also compatible with other technologies (e.g., vehicular networking supported by 5G and heterogeneous networking architectures), facilitating the connectivity from the VRU. Furthermore, with the VRU integration in the VEINS framework, it is possible to exploit the benefits of bidirectional simulations to gather traffic data, build communications networks and protocols that consider VRU, and decide over traffic dynamics at both vehicles and VRU. The tool is publicly available to the research community and continues under development.
Abstract: For increased safety and fuel-efficiency, vehicle platoons use Cooperative Adaptive Cruise Control (CACC) where vehicles adapt their state, incl. speed and position, based on information exchanged between vehicles. Intruders, however, may carry out attacks against CACC platoons by exploiting the communication channels used to cause harm, e.g., a vehicle crash. Therefore, during design-phase, engineers should provide evidence supporting platoon security. This paper proposes a formal framework for the security verification of CACC platoons to provide such evidence based on precise mathematical models. Our vehicle platoon models support the specification of both cyber, e.g., communication protocols, and physical, e.g., speeds, position, vehicle behaviors. Moreover, we propose intruder models that are parametric on his capabilities of manipulating communication channels, i.e., message injection and blocking. Our model is implemented enabling the automated formal verification involving both platoon and intruder models. We validate our machinery with a number of attacks taken from the literature and novel attacks discovered by using our formal machinery.
Abstract: "Cooperative Adaptive Cruise Control (CACC) is a typical example for cooperative maneuvers of connected, automated vehicles. Safe operation of any such vehicle highly depends on information emitted by the surrounding vehicles as it is processed by the vehicle's longitudinal controller. This opens an attack surface affecting vehicle safety. This paper studies how CACC controllers can be protected against such attacks to ensure the safety of the vehicle and its passengers. We discuss the current state of security in C-ITS standardization and the resulting protections, revealing the deficiencies regarding the protection of CACC. This raises the question of how to mitigate injection of false data into the controller by attacks and other forms of vehicle misbehavior. Based on a literature analysis, this paper defines a new categorization for mitigation strategies, discussing their individual strengths, weaknesses and their potential for attack mitigation. This categorization can be used to derive novel ideas for mitigations, for example, an approach with a parameter-dependent reaction of the controller based on a suspiciousness parameter signaled by the Misbehavior Detection System (MDS). Through a simulation study, we show that this approach improves safety and efficiency of the platoon under attack but also identify a fundamental trade-off between these two design goals."
Abstract: In this paper, we present a physical-layer cyberattack resilient OFDM radar design for automotive applications. A typical FMCW radar system sends multiple chirp signals in a coherent processing interval, whereas an OFDM radar sends multiple OFDM frames . OFDM radar approach has a potential advantage for automotive applications, which is the use same RF hardware and bandwidth for both AV radar and V2V/V2X communication . Cyberattack resilience of FMCW radars has been studied in , , , , in this paper we address the same problem for OFDM radars, propose a new radar algorithm called OFDM-i, and a cyberattack detector D 1. The proposed idea is based on selecting a fixed percentage of subcarriers as null, and changing these null bands randomly for each frame. To minimize the impact of this on range resolution, we also require no more than two neighboring null bands in each frame. We demonstrate the performance of OFDM-i both by using simulations, and then using real data obtained from a National Instruments 39 GHz mmwave system. We also provide an upper bound for the false cyberattack alarm rate, study the resilience of the proposed system using a combinatorial analysis, and then using simulated attacks. Finally, we summarize possible future research directions.
Abstract: "Previous work on misbehavior detection and trust management can identify falsified and malicious V2X messages, enabling the witness vehicles to report observations for certificate management. However, there may not exist enough ""benign"" vehicles with V2X connectivity or vehicle owners who are willing to opt-in in the early stages of connected vehicle deployment. In this paper, we propose a security protocol for V2I communication, titled Proof-of-Travel, to answer the research question: How can we transform the power of cryptography techniques embedded within the protocol into social and economic incentives to promote V2X adoption such that vehicle owners are more willing to participate in sharing traffic events and jointly mitigate malicious nodes? The key idea is to determine the trustworthiness of and the contribution made by a vehicle based on its distance traveled and the information it shared through V2X communication along the path of movement. Besides, the total vehicle mileage traveled for a vehicle must be testified by digital signatures signed by each infrastructure component in its trajectory path. While acquiring the chain of proofs of spatial movement creates burdens for malicious vehicles, the protocol does not result in extra cost for normal vehicles. Instead, the verifiable vehicle mileage traveled for a normal vehicle can be used to determine its contributions and the deserved reward. The altruistic behaviors of sharing its observations about traffic and road conditions can benefit the transportation network. We apply the protocol to assist infrastructure components in forming consensuses on the authenticity and correctness of vehicle-reported events and present initial simulation results."
Falko Dressler is full professor and Chair for Data Communications and Networking at the School of Electrical Engineering and Computer Science, TU Berlin. He received his M.Sc. and Ph.D. degrees from the Dept. of Computer Science, University of Erlangen in 1998 and 2003, respectively.
Dr. Dressler has been associate editor-in-chief for IEEE Trans. on Mobile Computing and Elsevier Computer Communications as well as an editor for journals such as IEEE/ACM Trans. on Networking, IEEE Trans. on Network Science and Engineering, Elsevier Ad Hoc Networks, and Elsevier Nano Communication Networks. He has been chairing conferences such as IEEE INFOCOM, ACM MobiSys, ACM MobiHoc, IEEE VNC, IEEE GLOBECOM. He authored the textbooks Self-Organization in Sensor and Actor Networks published by Wiley & Sons and Vehicular Networking published by Cambridge University Press. He has been an IEEE Distinguished Lecturer as well as an ACM Distinguished Speaker.
Dr. Dressler is an IEEE Fellow as well as an ACM Distinguished Member. He is a member of the German National Academy of Science and Engineering (acatech). He has been serving on the IEEE COMSOC Conference Council and the ACM SIGMOBILE Executive Committee. His research objectives include adaptive wireless networking (radio, visible light, molecular communications) and embedded system design (from microcontroller to Linux kernel) with applications in ad hoc and sensor networks, the Internet of Things, and cooperative autonomous driving systems.
Abstract: Railway transportation and in particular its safety-critical signalling equipment has experienced a wave of digitization in the recent years. Along with the utilization of off-the-shelf hardware and protocols the need for a resilient security concept rises to protect train operation against cyberattacks. We have build a testbed to emulate railway signalling network communication adhering to standards defined by EULYNX. The testbed allows us to test and evaluate a variety of cybersecurity controls in a realistic environment without risking to interrupt real-world railway operation.
Abstract: Climate has a wide range of effects on the roads and highways. Rain, snow, ice, fog, freezing rain, and other condition of weather can affect the driver's ability to drive their vehicles safely. It considerably decreases the capacity of roads as well as dramatically enhance the traveling time. Finnish Meteorological Institute (FMI), Finland delivers up-to-date real-time forecast data, road friction information, maintain roads as well as traffic management. In this article, we have discussed the developed advance ITS-assisted road-weather and traffic service architecture: we have established those ITS services for road weather to forecast accurate road-weather service parameters i.e. road condition (friction information), weather condition, traffic situation and road surface temperature etc. The condition of the road categorizes the surface of road into distinct classes: damp, dry, frost, snow, wet etc. Another index for road traffic is the classification of road- forecast which identifies whether the condition of driving is good, satisfactory, bad or very bad. The road traffic index is a mixture of the road-weather data and road condition. This state-of-the-art road weather service infrastructure and model aims to improve the road safety particularly in winter situation. The ITS infrastructure and road-weather services are tested and analyzed during pragmatically reproducible and reconfigurable field tests and analysis on our test track in Sodankyla, Finland.
Abstract: Message dissemination protocols for vehicular networks often build upon information maintained in neighbor tables. Due to high mobility and scarce channel capacity available in such networks, algorithms for maintaining neighbor tables must carefully select which information is transmitted at which time. Recent approaches rely on probabilistic data structures (e.g., Bloom filter) for transmitting such neighbor information in order to reduce channel load, yet they still suffer from scarce channel capacities. In this paper we propose to use multiple communication technologies (heterogeneous networking) for maintaining neighbor tables by building upon recent 2-hop neighbor management strategies. Promising results show an increase of up to 19 % of covered neighbors in comparison to a Baseline approach that only use one communication technology.
Abstract: Modern automotives and automated vehicles are increasingly adopting various sensors such as camera, LiDAR, etc. to assist drivers or enable certain automated features. While the on-board sensors provide the capability to sensing the environment, the sensors on a single vehicle are often susceptible to limitations due to occlusion or sensor range. To overcome the limitations, it is necessary to share sensor data or processed perception information among vehicles to extend the sensor coverage and avoid occlusions. In this paper, we take the sensor data sharing approach and present a real-time high-definition sensor data sharing using millimeter wave (mmWave) V2V communications. In particular, the implemented system allows vehicle to share high-definition sensor data including LiDAR and camera through an IEEE802.11ad communication link and for each vehicle to combine remote sensor data with its local sensor data.
Abstract: The recent research studies have unveiled the potential of vehicular micro clouds. The key idea is to have nearby vehicles collaborate with each other over wireless networks to form a virtual edge server on a road. This paper investigates the feasibility of leveraging vehicular micro clouds to provide an intelligent traffic management service. Through a field trial with multiple test vehicles traveling around a 4-way stop intersection, we succeeded in forming a vehicular micro cloud over real DSRC networks, demonstrating the benefit of an inter-vehicle coordination service offered by the virtual edge server.
Abstract: Vehicular on-board communication is the basis for advanced driver assistance, autonomous driving, over-the-air updates, and many more. If unprotected, this infrastructure is vulnerable to manipulation and various attacks. As any networked system, future connected cars require robust protection, monitoring, and incidence management against cyber-attacks during their lifetime. We demonstrate an infrastructure that secures the in-vehicle communication system and enables the security management of an entire vehicle fleet. Our prototype - a real-world production car - uses an Ethernet backbone network. It implements protective measures using software-defined networking, anomaly detection technologies, and is connected to a cyber defense center in the cloud. We demonstrate how this combination can reliably detect and mitigate common attacks on the vehicle - including its legacy components.
Abstract: Gossip Learning (GL) is a fully decentralized machine learning paradigm with the potential to enable highly scalability and to preserve user privacy. The majority of existing results however consider scenarios in which either each node communicates with all other nodes, or in which the connectivity graph is static, and they are therefore inapplicable in dynamic setups such as in VANETs. This work is a first attempt at designing and assessing GL schemes suited for scenarios with moving nodes with the application of predicting the trajectory of moving cars.
Abstract: Cooperative perception allows exchanging on-board sensor information through V2X networks for overcoming the limitations of local sensing. We present a complete cooperative perception platform based on COTS hardware, including an initial implementation of the ETSI TR 103 562 standard. Preliminary results show the performance of the system using two setups, namely the high impact of the 3D object detection task.
Abstract: Ultra-reliable and low-latency communication (URLLC) is one of the main requirements of the fifth-generation mobile communications system (5G), intended for realizing autonomous driving and telemedicine. Wireless transmission technologies that satisfy both reliability and low latency are required to realize URLLC. We have proposed an early-feedback hybrid automatic repeat request (early HARQ) that provides high reliability and low latency. However, we have only studied at the link level with ideal channel estimation, and the performance in a more practical environment was not clear. In this paper, we use reference signals for channel estimation and evaluate the early HARQ in a highway environment. Through the numerical simulations, we show the superiority of the early HARQ under real conditions to a conventional closed-loop HARQ.
Syed Rafiul Hussain is currently an Assistant Professor in the Department of Computer Science and Engineering at Pennsylvania State University. Before joining Penn State, he worked as a postdoctoral researcher at Purdue University from where he also received his Ph.D. in December 2018. His research interests broadly lie in network and systems security with a focus on the fundamental improvement of security and privacy analysis of emerging networks and cyber-physical systems, including cellular networks and Internet-of-Things. His papers have received awards and nominations, including ACSAC’19 distinguished paper award, NDSS’19 distinguished paper award honorable mention, and ACM SIGBED EWSN’17 best paper award nomination. He has been inducted twice in the Hall of Fame Mobile Security Research by GSMA for his contribution in identifying 20+ new protocol flaws in 4G and 5G cellular networks. His findings led to several changes in the 4G and 5G cellular protocol designs and in operational networks. His work has been featured by mass media outlets worldwide, including the New York Times, Washington Post, Forbes, MIT Technology Review, ACM TechNews, and The Register. More details can be found at https://syed-rafiul-hussain.github.io/.
Abstract: Traditional in-vehicle networks are based on low-bandwidth technologies like CAN. They are statically deployed and configured in the manufacturing process depending on the vehicle configuration. With the introduction of additional camera and entertainment applications and increased bandwidth demand, Ethernet technology becomes more relevant in the automotive sector. Use cases like trailer networks and integration of new applications requiring in-vehicle sensor data require re-configurable network architectures. In this work, we propose a flexible architecture for automotive Ethernet networks accommodating both high-bandwidth multimedia streams and time-critical low bandwidth data. Based on the software-defined networking paradigm, the in-vehicle network can be re-configured for the integration of additional hardware and applications. We illustrate its concepts for operations, management, safety, and security.
Abstract: Future vehicles will be more connected than ever. This leads to a dynamic growth and further development of vehicle on-board networks. Software-Defined Networking (SDN) is a promising technology to meet the emerging needs for flexibility and security in future automotive use cases. Although SDN controllers have been evaluated in data center networks, to the best of our knowledge there is a lack of an analysis and performance evaluation of SDN controllers for automotive use cases. In this work we provide a detailed requirements analysis for the use of SDN controllers in cars.Based on this requirements analysis we choose existing controller implementations for a performance analysis. Finally, we analyze automotive specific use cases for SDN controllers with controller application examples and show how these can fulfill additional requirements. Our evaluation provides a helpful basis for the redesign or development of SDN controllers that can be used in vehicles.
Abstract: Current In-Vehicle Networks (IVNs) connect Electronic Control Units (ECUs) via domain busses. A gateway forwards messages between these domains. Automotive Ethernet emerges as a flat, high-speed backbone technology for IVNs that carries the various control flows within Ethernet frames. Recently, Software-Defined Networking (SDN) has been identified as a useful building block of the vehicular domain, as it allows the differentiation of packets based on all header fields and thus can isolate unrelated control flows. In this work, we systematically explore the different strategies for integrating automotive control flows in switched Ether-networks and analyze their security impact for a software-defined IVN. We discuss how control flow identifiers can be embedded on different layers resulting in a range of solutions from fully exposed embedding to deep encapsulation. We evaluate these strategies in a realistic IVN based on the communication matrix of a production grade vehicle, which we map into a modern Ethernet topology. We find that visibility of automotive control flows within packet headers is essential for the network infrastructure to enable isolation and access control. With an exposed embedding, the SDN backbone can establish and survey trust zones within the IVN and largely reduce the attack surface of connected cars. An exposed embedding strategy also minimizes communication expenses.
Abstract: As the mobile and automotive industry move towards autonomous vehicles, many advanced driving applications have been developed. These driving applications may require different levels of intelligence, communication capabilities and processing power from the communication network and processing platform. These advanced driving applications can be grouped into static which runs all the time as the engine starts and dynamic which runs for a duration of time depending on the vehicle conditions. After the emergence of IEEE Time-Sensitive Networking (TSN) features for Ethernet technology, the automotive industry started to move towards the usage of TSN for advanced driving applications. However, IEEE TSN poses a challenge in streamlining the schedules and routes of the dynamic traffic since they require swift and fast determination of transmission schedules and routes on-the-fly. In this paper, we mainly focus on static traffic and device a novel static scheduling and routing algorithm which would be conducive for dynamic traffic requirements. In this approach, we have developed Mixed-integer programming (MIP) based joint scheduling and routing of static applications with the aim of load balancing such that more dynamic traffic would be schedulable as the vehicle drives off. We proposed two load-balancing based objective function and conducted an experimental analysis of objective functions with six different vehicle network configurations in two scales of zonal architecture. Experimental evaluations show the efficacy of our developed algorithm, in which the load is balanced in the egress port of the network, which in turn can schedule more dynamic traffic.
Abstract: Federated Learning (FL) is a recent distributed technique to extract knowledge, i.e. an abstract understanding obtained from a set of information through experience and analysis. Vehicular networks are highly mobile networks in which a large spectrum of data types is distributed. So far, no mechanisms have been defined that distribute FL model updates in vehicular networks based on which nodes are likely to hold the right data for training, and when. In turn, this potentially limits FL model training speed and accuracy. In this paper, we describe protocols to exchange model-based training requirements based on the Vehicular Knowledge Networking framework. Based on this understanding, we define vehicular mobility and data distribution-aware FL orchestration mechanisms. An evaluation of the approach using a federated variant of the MNIST dataset shows training speed and model accuracy improvements compared to traditional FL training approaches.
Abstract: Machine Learning (ML) is becoming ever more popular in many application domains, including vehicular networking. It has been shown already that Intelligent Transportation Systems (ITS) can greatly benefit from this approach, particularly from Reinforcement Learning (RL). To implement Vehicular Ad-hoc Network (VANET) environments for RL training, researchers often start from scratch. Because up until now, there is neither an established interface to ML toolkits nor a common scenario for VANET applications. Though such established standards would be a great benefit to research: Previous results would be easier to reproduce and different solutions could be compared in equal situations and using the same metrics. We developed Veins-Gym to bridge this gap. Veins-Gym combines the popular Veins vehicular networking simulator with OpenAI Gym. Using an exemplary VANET application, we show that RL techniques can be easily applied to ITSs with this framework. This enabled us to train an agent that outperformed hand-written algorithms.
Abstract: We present FALCON, a novel autonomous drone network system for sensing, localizing, and approaching RF targets/sources such as smartphone devices. A potential application of our system includes a disaster relief mission in which networked drones sense the Wi-Fi signal emitted from a victim's smartphone and dynamically navigate to accurately localize and quickly approach the victim, for instance, to deliver the time-critical first-aid kits. For that, we exploit Wi-Fi's recent Fine Time Measurement (FTM) protocol to realize the first on-drone FTM sensor network that enables accurate and dynamic ranging of targets in a mission. We propose a flight planning strategy that adapts the trajectory of the drones to concurrently favor localizing and approaching the target. Namely, our approach jointly optimizes the drones' diversity of observations while also approaching the target, while flexibly trading off the intensities of the potentially conflicting objectives. We implement FALCON via a custom-designed multi-drone platform and demonstrate up to 2x localization accuracy compared to a baseline flocking approach, while spending 30% less time localizing targets.
Abstract: Bicycles are becoming increasingly more equipped with embedded connected devices, by design or through after-market products, to support applications such as fitness monitoring and tracking. Bluetooth (BT) and BT Low Energy (BLE) technology is often embedded in such devices to support connectivity to a personal mobile device or a dock, when parked. BT/BLE transmit periodic beacons for node discovery that can be explored for V2X applications, such as safety and fleet management. We present a distributed system that explores periodic BT beacons sent by a module embedded in a bicycle to opportunistically locate nodes of interest (NOI). We address the particular application of stolen bicycle detection. In a scenario in which a bicycle is stolen and has its communication system tampered with but BLE remains functional, a service provider (e.g., fleet operator, authorities) is informed of this new NOI and shares an updated NOI list with the NOI detection-enabled bicycles. In turn, the bicycles flag contacts with stolen bicycles to the provider backoffice, at the earliest convenience (depending on available communications interfaces: immediately if cellular is available, or opportunistically when passing by a dock). We describe the operation and software architecture of the system, and an actual implementation in COTS equipment. Experimental measurements of the communication range and a demonstration of the system for proof-of-concept are also reported.