Analysis of synchronization in mobile sensor networks using time-varying poles

Abstract

Time Synchronization is an important feature in mobile sensor networks since it leads to more efficient data fusion, more efficient power saving schemes and more efficient access of the communication medium. The objective of time synchronization in a network is to provide a common time scale for the local clocks of all the nodes in the network. In this thesis, the problem of time synchronization in mobile sensor networks is modeled as an interconnection of Linear Time-Varying (LTV) systems operating over a graph. It is well known that in the case of Linear Time-Invariant Systems operating over a graph, the eigenvalues of the graph Laplacian provide useful information regarding the convergence of the network. This paper explores the use of time-varying analogs, i.e. LTV poles obtained using a factorization approach, in analyzing LTV systems interacting over a graph. The influence of simultaneous power transmit control and synchronization is also studied, and it is observed that in some regions of the parameter space, power transmit control can speed up synchronization, while in other regions of the parameter space, it can slow down synchronization.