Wireless sensor networks [1] are an exciting new area of research.
They belong to the class of ad-hoc networks, where the
individual nodes have limited sensing, computation, communication
and energy. The (envisaged) large scale of such networks prohibits
human intervention for network maintenance. One of the
very scarce resources for these types of networks is energy. These
networks are expected to have a long lifetime (weeks to years) without
human intervention for energy replenishment (recharging or
changing the batteries). Human intervention is undesirable since
large number of nodes imply high operational cost.
Current approaches to energy management mainly focus on low
power architecture and low power network design at different communication
layers. These include ( documents are a mixture of topics, where a topic
is defined as a probability distribution over words. These models are interesting
because they provide a simple probabilistic procedure for generating documents.
Such a procedure can be inverted using standard statistical techniques, allowing
us to infer a set of topics from which a particular document was generated. We
then associate the inferred topic distributions with class labels based on real-world
quantities such as company-level financial indicators for the classification task