Abstract: In some wireless sensor network applications, multiple sensors can be used to measure the same variable, while differing in their sampling cost, for example in their power requirements. This raises the problem of automatically controlling heterogeneous sensor suites in wireless sensor network applications, in a manner that balances cost and accuracy of sensors. We apply genetic programming (GP) to this problem, considering two basic approaches. First, we construct a hierarchy of models, where increasing levels in the hierarchy use sensors of increasing cost. If a model that polls low cost sensors exhibits too much prediction uncertainty, the burden of prediction is automatically transferred to a higher level model using more expensive sensors. Second, we train models with cost as an optimization objective, called non-hierarchical models, that use conditionals to automatically select sensors based on both cost and accuracy. We compare these approaches in a setting where the available budget for sampling is considered to remain constant, and in a setting where the system is sensitive to a fluctuating budget, for example available battery power. We show that in both settings, for increasingly challenging datasets, hierarchical models makes predictions with equivalent accuracy yet lower cost than non-hierarchical models.
Abstract: Resource constrained sensor systems are an increasingly attractive option in a variety of environmental monitoring domains, due to continued improvements in sensor technology. However, sensors for the same measurement application can differ in terms of cost and accuracy, while fluctuations in environmental conditions can impact both application requirements and available energy. This raises the problem of automatically controlling heterogeneous sensor suites in resource constrained sensor system applications, in a manner that balances cost and accuracy of available sensors. We present a method that employs a hierarchy of model ensembles trained by genetic programming (GP): if model ensembles that poll low-cost sensors exhibit too much prediction uncertainty, they automatically transfer the burden of prediction to other GP-trained model ensembles that poll more expensive and accurate sensors. We show that, for increasingly challenging datasets, this hierarchical approach makes predictions with equivalent accuracy yet lower cost than a similar yet non-hierarchical method in which a single GP-generated model determines which sensors to poll at any given time. Our results thus show that a hierarchy of GP-trained ensembles can serve as a control algorithm for heterogeneous sensor suites in resource constrained sensor system applications that balances cost and accuracy.