Abstract
Wireless sensor networks can provide large amounts of data that, when combined with pre-processing and data analysis processes, can generate large amounts of data that may be difficult to present in visual forms. Often, understanding of the data and how it spatially and temporally changes as well as the patterns suggested by the data are of interest to human viewers. This chapter considers the issues involved in the visual presentations of such data and includes an analysis of data set sizes generated by wireless sensor networks and a survey of existing wireless sensor network visualization systems. A novel model is presented that can include not only the raw data but also derived data indicating certain patterns that the raw data may indicate. The model is informally presented and a simulation-based example illustrates its use and potential.
TopBackground
The primary purpose of sensor networks is to acquire information about some environment. Sensor data is obtained both spatially and temporally, and for purposes of this chapter, is assumed to be transmitted to a computational base station for pre-processing and visual displaying. The first part of this section discusses the significant large data set sizes that wireless sensor networks impose upon visualization systems based upon a simple analysis. The second part discusses several wireless sensor applications in context of current day realistic data set sizes. And the third part discusses several existing sensor visualization systems.
Key Terms in this Chapter
Wireless Sensor Networks: Networks of sensor nodes capable of acquiring sensed information about the environment and communicating that information via wireless data links to base stations.
Data Level Visualization: Visualization aimed at displaying the values and patterns of the sensed data, may be combined with derived data visualizations, that is, visualizations of pre-processed sensed data.
Underlying Model: Dynamic system model composed of state-space parameters either observable or not which provides semantics about the sensed environment; observable parameters are sensed by the wireless sensor network.
Underlying Model Level Visualization: Visualization aimed at displaying the state-space transitions and behavior described by the underlying model.
Orthogonal Organized Finite State Machine (OOFSM): A special finite state machine abstraction used to represent the state-space transitions and state-space regions of behavior of the underlying model.
Multiple Level Visualization (MLV) Model: New visualization model that combines data level and underlying model level visualizations so as to provide underlying model semantics coupled with standard data visualizations of the sensed environment.
Visualization: Displaying information appropriately to facilitate human understanding leading to decision making about the sensed environment; usually, pictorial or graphical displays.