Wireless sensor networks (WSN) localization relies on measurements. Availability of, and the information content in, these measurements depend on the network architecture, connectivity, node time synchronization and the signaling bandwidth between the sensor nodes. This chapter addresses wireless sensor networks measurements in a general framework based on a set of nodes, where each node either emits or receives signals. The emitted signal can for example be a radio, acoustic, seismic, infrared or sonic wave that is propagated in a certain media to the receiver. This general observation model does not make any difference between localization of sensor network nodes or unknown objects, or whether the nodes or objects are stationary or mobile. The information available for localization in wireless cellular networks (WCN) is in literature classified as direction of arrival (DOA), time of arrival (TOA), time difference of arrival (TDOA) and received signal strength (RSS). This chapter generalizes these concepts to the more general wireless sensor networks.
General Observation Models
We start with a high level characterization of a wireless sensor network (WSN), and proceed with more detailed sensor models. An as generic vocabulary as possible will be used, and the particular application dependent notation will be introduced later on.
The WSN consists in general of nodes i = 1, . . ., M positioned at pit at time t. A general observation model of a signal sj(t) emitted from node j and received at node i with an amplification aij(t) and a delay τij observed in noise wij(t) is: (1)
A node can in this context be characterized with the following properties: