Development of a pinger for classification of feeding behavior of fish based on axis-free acceleration data

An acceleration pinger was developed for classification of feeding behaviors of sit-and-wait fish, using red-spotted grouper (Epinephelus akaara) as a model species. The pinger incorporates a classification algorithm constructed on the basis of axis-free acceleration data. The acceleration data were obtained in tank experiments using acceleration data loggers implanted in the abdominal cavity of fish. The acceleration pinger was designed such that it could sample acceleration data at high sampling frequency and could process the data to output the results of classification. The pinger would resolve one of key problems of acceleration pingers, i.e., the amount of data a pinger could transmit was crucially limited.