SP2 Data Communication

SP-Leader: Guillaume Chelius (CITI, Lyon)

The design of data communication techniques in sensor and self-organized networks has challenges at all layers of communication: physical, medium access control (MAC), network and transport layers. This part of the research project concentrates on the network layer, with occasional `visits' to medium access or transport layers. In data communication problems, such as routing, quality-of-service routing, geocasting, multicasting, and broadcasting, the primary goal is to fulfill a given communication task successfully between nodes in ad hoc network. The secondary task is to minimize the communication overhead (since bandwidth in wireless communication is typically limited) and power consumption by battery operated nodes. These problems can be defined as follows:

• Routing (sending a message from a source node to a destination node),
• Quality-of-Service routing (selecting a network path between source and destination with sufficient resources to satisfy the delay and bandwidth requirements),
• Geocasting and multicasting (sending a message from a source to all the nodes inside a geographic region, {e.g. circle, or to a group of users, respectively),
• Broadcasting (sending a message from a node or central facility to all the nodes in the network).
• Location updates for efficient routing (finding reasonably accurate destination addresses).

These problems are related to each other, and the solution for one impacts the solution for the other (e.g. broadcasting is frequently used for route discovery). Routing problem also encompasses a wide range of applications. For instance, service access problems are basically the problems of routing from a given node to a fixed destination ({e.g. NAP). Thus service access problems are primarily routing problems. There are normally some variations to this problem in a particular scenario. For instance, sensors reporting temperature may aggregate individual measurements along the route toward monitoring station.

Mobility Models

The problems associated with routing messages and transporting information in a high mobile environment are tightly related to the state and evolution of the radio links throughout time, hence to the spatial repartition of people throughout this same time. While the precise displacements of persons cannot be predicted, identifying certain characteristics of the populations movements can help improve heuristics for the choice of actions to take at the node level.

Our goal here is to measure the comportment at the individual level and the evolution of spatial density surrounding it. Using sensor networking nodes, we will gather information from each other node they encounter, relative positions to anchor points as they pass by, as well as other information on the place they are at (luminosity, sound level, acceleration, magnetic orientation, etc.).

This data will allow a study of the network links evolution, as well as a first sketch of users comportment. This information will be used in building heuristics for data relaying decisions, and help us refine models used to test new routing algorithms.