Hitachi (France)

Following the success of the 1st, 2nd, 3rd, 4th and 5th CHiME challenges we organize the 6th CHiME Speech Separation and Recognition Challenge (CHiME-6). The new challenge revisits the previous CHiME-5 challenge and further considers the problem of distant multi-microphone conversational speech diarization and recognition in everyday home environments. Speech material is the same as the previous CHiME-5 recordings except for accurate array synchronization. The material was elicited using a dinner party scenario with efforts taken to capture data that is representative of natural conversational speech. This paper provides a baseline description of the CHiME-6 challenge for both segmented multispeaker speech recognition (Track 1) and unsegmented multispeaker speech recognition (Track 2). Of note, Track 2 is the first challenge activity in the community to tackle an unsegmented multispeaker speech recognition scenario with a complete set of reproducible open source baselines providing speech enhancement, speaker diarization, and speech recognition modules.

Nowadays, researchers show more and more interests to vehicular ad hoc networks (VANETs), which are a specific instance of mobile ad hoc networks (MANETs) where nodes are vehicles. In VANETs, vehicles have no energy resource constraint which could extend coverage and network lifetime, but have a high mobility patterns that cause frequent and fast topology changes. Consequently, VANETs have particular research interests, like dedicated MAC and routing optimization. In our previous work, we have proposed movement prediction-based routing (MOPR) concept for VANETs, which improves the routing process by selecting the most stable route in terms of lifetime with respect to the movement of vehicles. And in this paper, we present how this MOPR concept can be applied to position-based routing protocols, and how it improves their performances. Based on simulation results we compare MOPR with the position-based routing protocol GPSR and another movement-based routing protocol called MORA.

Inter-vehicular communications (IVCs) are now considered as a way to realize active safety, for example, by providing the position information of each other or the potential danger warning by wireless communications. We have worked on a flooding protocol over vehicular ad hoc networks (VANETs) to efficiently disseminate the information for the sake of active safety applications, such as the positions and the velocities of the vehicles. We propose a flooding protocol with (i) congestion detection algorithm which suppresses unnecessary packets due to vehicular congested traffic and (ii) Backfire algorithm which efficiently forwards the packet through the network by selecting the adequate receiver node based on the distance from the original node. In this paper, we show simulation results over NS2 (network simulator 2). They show that the proposed flooding protocol significantly improves the performance of data dissemination over VANETs

This paper, focuses on the "enhancement" of multi- hop vehicular broadcast (MHVB). The protocol is fundamentally a flooding algorithm with special characteristics in order to efficiently disseminate information such as the positions and the velocities of the vehicles for the sake of active safety applications. The main purpose of this paper is to show the performance improvements obtained by adding more special characteristics to the existing version of MHVB. The enhancement procedure is carried out in two steps: by changing the shape of the backfire region in the algorithm and by introducing a new Dynamic Scheduling algorithm which prioritizes the packet transmission based upon "processing" of the received packets from the other vehicles. The key point in the proposal made to enhance the broadcast protocol is the balance between the application requirement and the performance of the protocol.

Wireless vehicular communications are attracting more and more interests for applied research in industries. Most of the efforts are spent in deploying Vehicular Mobile Ad-Hoc Networks (VANETs) for applications such as active safety and Internet services. This paper addresses routing problem in VANETs for applications related to comfort and infotainment for users where an unicast routing protocol optimized for fast topology changes is needed. In previous research work, we have proposed a new movement prediction-based routing concept for VANETs called MOPR which we have already applied to the reactive routing protocol AODV in order to improve its performances by exploiting vehicules movements patterns. In this work, we first propose a new design of this concept, then we apply it to the OLSR routing protocol by optimizing the procedure of selecting the MPR (Multipoint Relay) sets as well as that of determining the optimal path from each pair of vehicles. Basically, the connected MPR graph is composed of the most stable wireless links in the VANETs. We conduct several simulation scenarios to investigate the performance of the modified OLSR (OLSR-MOPR) by studying several metrics including the end-to-end average delay, the routing overhead, the packet delivery ratio, and the routing overhead ratio. The simulation results of the modified OLSR for various VANETs scenarios show great improvements comparing to the basic OLSR.

Automated vehicles require an adequate and reliable perception of the surrounding world in order to make good decisions. Using vehicle-to-vehicle (V2V) communication to exchange location data (i.e. time, position, heading and speed) can improve the perception beyond the capabilities of traditional on-board sensors (e.g. radar, lidar). However, it is vital to trust the data before it is being used. Cryptographic mechanisms can protect the exchange and authenticity of data but do not guarantee the correctness of the content. In this paper we present a vision-based multi-object tracking system for checking the plausibility of V2V communication. The system is addressing the challenge of fusing relative sensor observations as provided by a MobilEye vision-system with time-delayed absolute GNSS-based measurements from Cooperative Awareness Messages (CAMs) as provided by V2V. The plausibility check is implemented in a prototype and based on a state-of-the-art multiple-object tracking algorithm. The proposed system is evaluated and validated under real-world conditions by conducting several test drives under urban conditions.

The European Union (EU) Framework Program 7 (FP7) funded project, iTETRIS (An Integrated Wireless and Traffic Platform for Real-Time Road Traffic Management Solutions) targets to extend state of art in simulation of wireless vehicular cooperative systems for evaluation of road traffic management services and applications. In particular, iTETRIS addresses four important and distinct challenges: (a) road traffic and wireless integrated open-source simulation platform, (b) large scale trials, (c) realistic Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communication simulation and (d) dynamic, distributed and self-autonomous Intelligent Transport Systems (ITS) applications based on cooperative systems. iTETRIS will provide a standard compliant, open-source integrated communication and traffic platform suited for large scale scenario simulation. This Paper addresses extension of traffic simulator being used in iTETRIS, scalability studies of networking simulator, which have been achieved as project results. Further the paper presents the adaptation of the Intelligent Communications for Transport Systems (ICTS) reference architecture functionalities being developed by the European Telecommunication Standard Institute (ETSI) at its Technical Committee (TC) on ITS.

International audience

Engine Control Systems (ECS) for automobiles have numerous variants for many manufactures and different markets. To improve development efficiency, exploiting ECS commonalities and predicting their variability are mandatory. The concept of software product line engineering meets the business background of ECS. However, we should carefully investigate the expected technical, economical, and organizational effects of introducing this strategy into existing products.This paper explains an approach for assessing the potential of merging existing embedded software into a product line approach. The definition of an economically useful product line approach requires two things: analyzing return on investment (ROI) expectations of a product line and understanding the effort required for building reusable assets. We did a clone analysis to provide the basis for effort estimation for merge potential assessment of existing variants. We also report on a case study with ECS. We package the lessons learned and open issues that arose during the case study.

The decision of an organization to introduce product line engineering depends on a sound and careful analysis of risks and return on investment. The latter is computed by an economic model, which relies on high quality input and must reflect the envisioned migration strategy sufficiently. To facilitate risk analysis, this paper applies Monte-Carlo simulation to an existing product line economic model. Additionally, the model is extended by the support of product line generations that is, considering the degeneration of product line infrastructures and taking reinvestment into an existing product line into account. The practical application of the model is demonstrated by an industrial case study

Side channel attacks and more specifically fault, simple power attacks, constitute a pragmatic, potent mean of braking a cryptographic algorithm like RSA. For this reason, many researchers have proposed modifications on the arithmetic operation functions required for RSA in order to thwart those attacks. However, these modifications are applied on theoretic - algorithmic level and do not necessary result in high performance RSA designs. This paper constitute the first complete attempt for an efficient design approach on a fault and simple power attack resistant RSA based on the well known, for its high performance, Montgomery multiplication algorithm. To achieve this, a fault and simple power attack resistant modular exponentiation algorithm is proposed that is based on the Montgomery modular multiplication. In order to optimize this algorithm's performance we also propose a modified version of Montgomery modular multiplication algorithm that employs value precomputation and carry save logic in all input, output and intermediate values. We introduce a hardware architecture based on the proposed Montgomery modular multiplication algorithm and use it as a building block for the design of a fault and simple power attack resistant modular exponentiation unit. This unit is optimized by taking advantage of the inherit parallelism in the proposed fault and simple power attack resistant modular exponentiation algorithm. Realizing the proposed unit in FPGA technology very advantageous results are found when compared against other well known designs even though our design bears an extra computation cost due to its fault and simple power attack resistance characteristic.

Engine Control Systems (ECS) for automobiles have many variants for many manufactures and several markets. To improve their development efficiency, exploiting ECS commonalities and predicting their variability are mandatory. The concept of software product line engineering meets this ECS business background. However, we should carefully investigate the expected technical, economical, and organizational effects of introducing the strategy into existing products. Thereafter, a strategy will be derived systematically and realize the desired benefits.This paper reports an experience with the up-front investigation performed for Hitachi's ECS. We focus on the approach to plan the migration of the existing family of individual systems into a future product line. The approach assesses potential ways of merging software from existing variants and eventually defines a procedure for performing the migration. To get a high quality strategy, we integrate the approach of software measurement, the expertise of software architects, and reverse engineering techniques.

Passive network monitoring is very useful for the operation, maintenance, control and protection of communication networks, while in certain cases it provides the authorities with the means for law enforcement. Nevertheless, the flip side of monitoring activities is that they are natively surrounded by serious privacy implications and, therefore, they are subject to data protection legislation. Chapter 22 investigates the challenges related to privacy protection in passive network monitoring, based on a joint technical and regulatory analysis of the associated issues. After introducing the issue and its special characteristics, the chapter provides background knowledge regarding the corresponding legal and regulatory framework, as well as some related work. It then delves into the description of the legal and regulatory requirements that govern network monitoring systems, before providing an overview of a reference monitoring system, which has been designed with these requirements in mind.

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> This paper proposes a model for invariant resource sharing problems in dioid algebra. A strong motivation for investigating the issue is the absence of a general systematic technique which can be used to tackle these problems. <formula formulatype="inline"><tex Notation="TeX">$(\min, +)$</tex></formula> constraints have been developed to handle resource sharing in Discrete-Event Dynamic Systems. In particular, the part that can be modeled by a Timed Event Graph induce <formula formulatype="inline"><tex Notation="TeX">$(\min, +)$</tex></formula>-linear equations which are constrained by the resource availability. The proposed algebraic model has been proved to describe the actual behavior of the systems dealt with. This paper will show two examples of systems that are modeled and controlled by means of this approach. </para>

Vehicular communication is an important part of the Intelligent Transportation Systems (ITS). Geographic routing in vehicular ad hoc network (VANET) is becoming an interesting topic to deliver safety messages between cars but also between a car and a roadside infrastructure within a designated destination area. The Car2Car Communication Consortium specified C2CNet architecture as a geographic routing protocol. The results of GeoNet project are presented in the paper, which aims at combining IPv6 networking and C2CNet. The system with IPv6 and C2CNet is designed and implemented in Linux. The prototype implementation is first evaluated indoor testbed with the fixed positions. Then it is evaluated in the field testbed with three vehicles with various scenarios. For evaluation in field testbed, we have developed the AnaVANET evaluation tool to perform the evaluation taking into account all of geographic factors.

Data dissemination in vehicular networks has been a challenge due to unpredictable network dynamics and channel unreliability. In fact, conventional approaches that rely on TCP or UDP do not perform well and cannot consistently guarantee reliable communications. A potential solution to these problems involves using erasure-correcting codes to make UDP transmissions reliable. However, it is also important to note that, due to vehicular mobility and road obstacles, connections among nodes can be as short as a few seconds. Therefore, in a communication between two nodes, the delay between node discovery and data exchange must be minimized. In this paper we present CORP, a Cooperative Rateless Protocol that exploits the reliability of the rateless coding approach while performing fast and efficient dissemination through cooperating nodes within the network. The use of rateless codes and unicast connections places minimal constraints on the delay between node discovery and data transmission, thus simplifying and rendering feasible the content reliable download even in a highly dynamic scenario. Results from simulations reveal that performance improves as more and more nodes cooperate. In sum, CORP yields important improvements in terms of speed of dissemination when compared to traditional approaches such as TCP.

Efficient dissemination of messages in a vehicular ad hoc network (VANET) still face many challenges in the current research scenario. This paper addresses the problem of redundant forwarding of messages that occur during broadcast and proposes an adaptive forwarding mechanism which controls the amount of redundant messages thereby improving message dissemination over a VANET. The mechanism will enable a node to efficiently forward a message and at the same time refrain other potential forwarders of the same message in a chosen area to forward the message. The algorithm exploits the density of nodes on the network for its functionality and will make an adaptive sector under which the redundant nodes will be refrained from sending the same message. The area of the sector for a particular node will be proportional to the density of its neighbors that are within the node's communication range. The proposal assumes that the nodes in the VANET are aware of their surroundings up to a certain area by periodic exchange of location information from each of the nodes. The mechanism will enable routing algorithms and/or beaconing systems to consume lesser bandwidth and thereby enhance the efficiency of the algorithm itself.

Advanced Driver Assistance Systems (ADAS) rely on a combination of data acquired from on-board sensors and map data, the so-called Electronic Horizon. The Electronic Horizon provides an extended view on the surroundings of the ego vehicle, thereby enriching safety and efficiency applications. In this context, one key aspect is the application program interface (API) that is used between the Electronic Horizon and ADAS components. Although proposals of interfaces exist, none of them currently meets the requirements of future automated vehicles. To fill this gap, we propose extensions to the concepts used in ADASISv2 and ETSI LDM. Our solution comes in the form of an extended ADASIS protocol that uses a hybrid data representation approach: 1D lane-level path map and 2D detailed spatial map. We validated our protocol with real-world experiments being run in an automated vehicle and showed that such a hybrid approach can reduce up to 80% of the amount of data sent to applications, while maintaining compatibility with the current ADASISv2 protocol specification.

Owing to the increase in both heterogeneity and complexity in today's networking systems, the need arises for an architecture for network-based services that provides flexibility and efficiency in the definition, deployment and execution of the services and, at the same time, takes care of the adaptability and evolution of such services. In this paper we present an approach that applies a component model to GT4, a Web-service-based grid environment, that enables the provision of parallel applications as QoS-aware (grid) services, whose performance characteristics may be dynamically negotiated between a client application and service providers. Our component model allows context dependencies to be explicitly expressed and dynamically managed with respect to the hosting environment, computational resources, and dependencies on other components. Our work can be seen as a first step towards a component-based programming model for service-oriented infrastructures utilizing standard Web-service technologies

Green Light Optimal Speed Advisory (GLOSA) systems contribute to the reduction of CO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions and fuel consumption by giving speed advice to drivers based on current and future traffic light signal phase timings so they can avoid unneeded stopping and acceleration. These systems have been well investigated by means of simulations and real-world tests. In previous work we have shown that simulations tend to overestimate the communication quality to be expected in urban environments and that in a real-world test, IEEE 802.11p-based GLOSA cannot always reach the required information distance. Although multi-hop information dissemination can help alleviate this problem, it has not yet received much attention from the research community in the context of GLOSA systems. In this paper we present results from extensive field tests with almost 200 traffic light approaches. We find that two-hop dissemination of signal phase and timing information from traffic lights increases the maximum information distance by around 35% and is able to support continuous updates even in challenging environments.