1
Local Interactions for Cooperative ITS: Opportunities and Constraints

Jean-Marie BONNIN1 and Christophe COUTURIER2

1 IMT Atlantique, Rennes, France

2 YoGoKo, Rennes, France

1.1. Introduction

Since the advent of wireless communication and its integration into consumer devices, the concept of intelligent environment or pervasive application has emerged. The ability to communicate with all objects in our immediate environment makes it possible to take information or trigger actions. Information collection feeds a context that applications take into account to adapt their behavior to the situation.

For this type of application, direct interaction with objects in the environment greatly facilitates matters, since it is not necessary to rely on a precise location and database to associate information (or objects) with this location. If we need to know the room temperature, all that is needed is to discover a temperature sensor and query it directly. Acquiring the same information when a server is in charge of collecting and exposing the building’s temperature data firstly implies discovering the server that has the information at its disposal, then dialoging with it to retrieve the temperature of the room in which the sensor is located, and finding consequently a way to determine that the location is necessary. The machinery to be put in place is much more complex and yet it seems more intuitive, as the majority of the industry has been built on this model.

The difficulty when it comes to building services on direct (we will also use the term “local”) interactions is that this implies standardizing the method of communication, the frequency (or frequencies) used and the message format. For road or city applications, it is therefore necessary to bring many actors to agreement, and to impose choices on the entire ecosystem.

Direct interactions are widely used today for service discovery; for example, Wi-Fi devices continuously scan all frequencies used in the 2.4 GHz and 5 GHz bands to determine if there is an access point available in the environment. The presence of such an access point in no way indicates that the terminal will know how to connect to it, and even in the case where it is able to connect, whether it will be able to obtain a service (an Internet connection). The other technology widely used on consumer terminals is Bluetooth. Again, part of the terminals expose their presence by regularly sending messages at a determined frequency. All Bluetooth devices in proximity are able to see these messages and determine whether or not they know the correspondent. They can then either establish a connection to perform the service (e.g. hands-free kit) by taking advantage of the keying material previously established during pairing, or ask to perform a pairing, which requires the user’s intervention.

It should be noted that even when the two correspondents know each other, whether via Wi-Fi or Bluetooth, the discovery and connection establishment time frame is far too long for services with significant time constraints. We will return to this when we examine how the specificities of ITS-G5 make it possible to significantly reduce the time required to exchange information for road safety-related services.

In the second part of this chapter, we will present the concept of ephemeral local interactions, giving examples of services based entirely (or partially) on this type of interaction. We will describe how the first services that will be deployed in the context of cooperative ITS (awareness) are based on this type of interaction and the advantages/constraints of this approach. Lastly, before concluding, we will explore the place infrastructure holds in the implementation of services, based on ephemeral local interactions.

1.2. Ephemeral local interactions: concept and examples

1.2.1. Examples of services using ephemeral local interactions

Once it has been established that the different devices in interaction use the same communication technology on a subset of frequencies well known to all, it is necessary to specify the type of interaction targeted. Indeed, we will focus more specifically on interactions where no connection is established. When two devices are in proximity, they can “see” each other because of their technology community; they have at their disposal information that is spontaneously sent by their peers without having to go through the time-consuming establishment of a connection. When the communication technology has a fairly short range, simply being in communication and seeing a device gives an indication of co-spatiality that can form an integral part of the service. Therefore, when a telephone receives an advertisement on one of the three Bluetooth Low-Energy (BLE) channels, it knows that it is in close proximity to the tag whose identity is transmitted in the message, in addition to the information contained in the message itself. In a supermarket, the reception of its advertisements makes it possible to locate the mobile as long as the service provider has the precise placement of the tags in the store at its disposal. However, tags can also directly send information that may be used by the smartphone itself, such as a price, promotions or a link to the page describing a product.

Within the framework of the TousAntiCovid backtracking application and other mechanisms for identifying at-risk contacts, developed in the context of the Covid-19 pandemic, this co-spatiality property was used to identify transmission risks (Roca 2022). The complexity of the application comes mainly from the need to protect the privacy of users, while ensuring contact identification that is as accurate as possible. It was therefore necessary to avoid storing the list of contacts but to transmit within the advertisement messages the information required to make it possible to determine a posteriori whether their smartphone had been in contact with that of a contaminated person.

The case of contactless payment applications is rather different, since it instead involves establishing as secure a connection as possible to carry out a monetary transaction. It is therefore absolutely necessary to ensure that we are faced with the right device, and to prove that a valid transaction has taken place. However, the co-spatiality property is used to ensure that the payment card with which the transaction is carried out is in immediate proximity to the payment terminal. NFC (Near Field Contact) technology has been specifically adapted to reduce range and impose “near-contact”. The operation of radio transmissions makes this work quite complex because of the propagation of waves in the frequency bands used. It poses security problems, since it makes it possible, for example, to use relays. It then becomes necessary to go beyond controlling the transmission power to limit the range and to very finely control the transmission time, which also depends on the distance; this makes it possible, when it is excessive, to detect an attempt to relay the signal.

Prior to the emergence of the Bluetooth Low-Energy (BLE) version, applications used RFID technology, which has the great advantage of being able to install devices in the environment at very low cost, capable of “responding” to a request and sending previously configured data. This is generally a simple unique identifier, relatively similar to that of eBeacons in BLE. These RFID tags also have the property of being passive most of the time and of using the energy of the reader, which lights them up to wake up and respond. They therefore do not require a battery but are, however, inactive as long as they are not lit up. Readers also need to consume a fairly significant amount of energy to power the tags remotely.

In the different examples that we have seen, local interactions are essentially used to transmit an identifier, which makes it possible to establish our position by referring to prior knowledge of the position of the various devices. Richer applications make it possible to transmit information that the correspondent can directly use and that is most often linked to the position of the sender (a URL describing a product). This somewhat removes the need to maintain a geographic information system (GIS). In the case of BLE, this information is transmitted regularly, whether or not there is a correspondent to listen to it and to do something with it. The information broadcast in this way forms part of the environment and enriches it. The outlines of what we will call “ephemeral local interactions” are given below.

1.2.2. Characteristics of ephemeral local interactions

The first characteristic of local interactions is that they are established in the event of a close contact supported by a short- or medium-range wireless communication (BLE, NFC, RFID, ITS-G5, etc.).

The examples presented above highlight the opportunistic nature of these contacts. The objects considered evolve within a very large scope. They interact, sometimes ephemerally, with many other objects that they do not know beforehand. As stated above, from the outset, this excludes communication technologies requiring a form of pairing (e.g. Bluetooth) or a connection to a network (e.g. Wi-Fi or cellular networks). Indeed, beyond the fact that the necessary establishment time would often be prohibitive with regards to the applications envisaged, it is simply impossible for objects to memorize specific association parameters for each of these...