Members

Michel Banatre, Paul Couderc, Jean-Marc Menaud, Frédéric Weis.

Ph.D. positions

The Solidor project is now hiring three PhD students to contribute to the ongoing research work, about Spontaneous Information Systems.

Motivations

Development and progress of appliances enable to envision the implementation of new application classes that exploit direct communication with a limited range. Mobile users will be able to exchange information dynamically and spontaneously when they are physically close together.

In order to illustrate our proposal, we consider the following example : a set of people are taking part in a conference. Several sessions occur simultaneously in separate rooms. The participants are moving freely from one session to an other, depending on the exposed subject and their center of interests. We also suppose that these people have wireless appliances that can communicate directly together in a limited range. Each appliance contains a set of information and documents, for instance the owner's identity, their center of interests classified by key words, a copy of its presentation, a set of bibliographic references and URLs … In that specific context, each people is placed in an environment where a lot of information can be potentially exchanged with the other people that are standing in the same surrounding physical space. But these information’s can only be accessed at a given instant and during a limited time that can not be bounded because of the users mobility. The idea is to create spontaneously an information system that allows to users that are close enough to discover and exploit in an efficient way all these “distributed information”. It is clear that this idea can be also applied to other application classes like: mobile robotics or car information systems.

Let us introduce the concept of Spontaneous Information System (SIS), it can be roughly defined as follows : a SIS is made of a set of mobile users that carry an appliance with a wireless communication interface. The system is spontaneously created as soon as, at least, two mobile entities approach one from the other. Afterward the entities have the ability to establish a communication, and can exchange implicitly or explicitly some information. The SIS disappears as soon as the entities move away one from each other. In that context, the challenge is to ensure efficient information processing in spite of the user mobility. In the worst case, the occurrence of a disconnection between two users that are taking part in the same SIS can be definitive. So masking disconnection is impossible. Solutions to this problem has to be based on an information systems that takes into account the user mobility. In other words physical mobility belongs to the « semantics » of the application.

Objectives

Currently we are working on three research topics.

Information discovery. When one entity meets another one, data can potentially be exchanged. The visibility of such discovered information has to be governed by protection and security rules. This task will provide the software components enabling complex information discovery. In particular we have to cope with the following problems related to naming, addressing complex information. At the same time, we have also to address the fundamental issue of information confidentiality according to a user defined policy.
 

Predictive schema The second task under consideration is related to the design of a predictive schema to evaluate the communication time between entities of the same SIS. For example, we have to take into account, at the physical level, parameters such as the distance between two appliances or their relative speed. More precisely, we have first to design the following components in order to reach our objectives :

• Design of the presence detection components which allow one autonomous entity Ei to detect the other one that are close enough to initiate information exchanges with it.
• Design of the position evaluation components which provide physical characteristics of the position of each entity Ei relative to an other detected one Ej. Such a position is ideally defined by three parameters : the distance between entities, the speed vector of the entity, and the angle between the speed vectors of the two entities.
• Design of the power components which deliver information related to the power consumption and the power range of each considered entity.
  

Information provided by these modules will be stored by the different SIS components in order to support efficient information system discovery and information access.

Neighborhood communication protocol. Another one is about information transfer protocols between SIS entities (Neighborhood Communication Protocol or NCP). The goal of this task is to design and implement the basic communication primitives (send and receive). These primitives will be designed to benefit of the predictive schema built from environment awareness (transmission power, speed, …). In other words the NCP will provide mechanisms to transfer structured information between SIS entities and stabilize them despite the highly volatile environment (user mobility and limited coverage of the communication interface,…).

Related areas

• Communication architecture
  Piconet
  Bluetooth

• Ad hoc networking and embedded systems
  SCADDS project
  Monarch project

• Wearable and Ubiquitous computing
  Wearable Computing Research Group (University of Oregon)
  Parctab

• Information discovery
  JINI
  Salutation