Packet Loss Performance and MAC Design for Estimation in Wireless Sensor Networks
Abstract: Recent advances in wireless communications and electronics has enabled the development of low-cost sensor networks. Wireless sensor networks can be used for various application areas (e.g., military, home, environment). Each application poses its own particular technical issues. One key problem that arises in many applications is how to organize the computation and communication in a sensor network to enable accurate estimation of the state of a dynamic system. As a starting point in this thesis, we demonstrate that packet loss and delays have a strong influence on the performance of optimal estimators. Thus, when we want to use a sensor network platform to collect and forward measurements to a data collector for estimation, it is critical to understand the loss performance of current hardware platforms and, possibly, to design MAC schemes that eliminate losses and minimize data collection latency. We investigate on the problem of packet loss, with practical experiments on a real wireless sensor network built up with Telos platform. The experiments in this thesis are made with single hop and multi-hop protocols in a star and linear network topology. To solve the problem of data delay and to avoid the collisions that lead to packet losses, we propose an overlay TDMA over CSMA. The underlying idea is to have a distributed protocol which is able to schedule the transmission of the nodes in order to have a data flow that converges from the furthest nodes toward the fusion center. To show the benefits of our protocol we simulate the behavior of a network in which nodes are randomly scheduled by the fusion center. Our protocol permits to obtain, in a distributed way, the optimal solution of such a network giving a strong improvement over a casual schedule.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)