The wireless sensor networks contain large number of nodes that can collect and transmit data autonomously. These sensors are usually equipped with an energy source (battery). Generally this resource is limited, thus reducing energy consumption represents a real challenge for this type of networks.
The aim of the thesis is to optimize data collection from sensors used in the EverBlu project of Itron company, which consequently increases the life of these sensors. EverBlu is an automatic meter reading systems suitable for all types of urban, suburban or semi- rural environments.
We propose a new medium access technique to reduce the number of packet retransmission in case of packets collision or data loss. To reduce the number of packet exchange in the network, we improve the data aggregation technique in EverBlu network.
EverBlu Deployment is a very important process to ensure 100% recovered data. To optimize the data collection several algorithms have been proposed to predict the positions of the collectors to be deployed on field. After deployment, some other algorithms must be executed to balance the sensors load on each collector.
The proposals will be implemented and tested on different sites where Itron deploys EverBlu network.