Energy-Aware Feature and Model Selection for Onboard Behavior Classification in Low-Power Animal Borne Sensor Applications

We present an energy-aware feature and classifier selection technique for low-power sensor applications. The aim is to minimize the combined energy consumed by feature extraction and statistical classification while minimizing the associated loss in classifier accuracy. Our particular application is the development of animal borne sensors with onboard behavior classification to support conservation efforts of endangered wildlife. Our technique makes use of cross-validated sequential forward feature selection to identify a shortlist of classifiers and feature sets that offer a favorable trade off between energy consumption and classification accuracy. This shortlist is subsequently re-ranked by incorporating the energy cost of the classifier itself, which is often disregarded in related studies. Our technique, therefore, favors classifiers and features which are in combination less energy expensive to compute at runtime. We apply this technique to datasets of accelerometer data, which has been complied for sheep and rhinoceros. For the sheep dataset, we are able to achieve a 6.8-fold reduction in energy consumption relative to a baseline while suffering a loss in classification accuracy among five behavioral classes from 89.6% to 88.4%. For the rhinoceros dataset, we achieve a 363-fold reduction in energy requirements while suffering a loss in classification accuracy among three behavioral classes from 99.6% to 99.3%. We conclude that the presented technique offers a feasible and successful means to achieve the principled design of statistical classifiers intended for low-power embedded sensor applications.