On accuracy of region based localization algorithms for wireless sensor networks
Computer Communications
| Hi-index | 0.00 |
Availability of node positions in ad hoc networks is of a twofold interest. First, it enables the use of position centric methods of addressing and routing. Second, node positions are a goal in themselves for applications such as tracking, sensing and reporting. This thesis makes contributions in both these directions. Trajectory based forwarding (TBF) is a novel method to forward packets in a dense ad hoc network that makes it possible to route packets along a predefined curve. The fundamental aspect of TBF is that it makes the transition from a discrete to a continuous view of the network. TBF's main advantages are that it provides cheap path diversity, it trades off computation for communication, and it decouples path naming from the actual path. Trajectories are a natural namespace for describing route paths when the topology of the network mirrors the topography of the physical surroundings in which it is deployed. I show how simple trajectories can be used to implement important networking protocols such as unicast, broadcast, multicast, multipath routing, and discovery in a quick and approximate way, as it needs only one support service: positioning. TBF requires that nodes know their position relative to a common coordinate system. Ad hoc positioning system (APS) is a family of methods that extends the self positioning capabilities of a fraction of landmark nodes to the rest of the network. It makes use of other node capabilities, such as angle of arrival, range estimations, and compasses. I characterize APS's performance by proving a Cramer-Rao bound for position error, and by simulating tracking and routing on top of obtained positions. Localized positioning system is a method that positions only the nodes along a trajectory, allowing TBF to function without landmarks, and even without a global coordinate system.