Energy-efficient adaptive communication by preference-based routing and forecasting

Abstract

Life simplifying fully automated factories are only feasible with a high accuracy and quality checking sensor density. In order to realize a huge amount of sensor nodes a single node must require minimal amount of energy for cost and uptime reasons. In this paper, a behavior adaptive routing protocol is presented which utilizes factory structures, resulting order relations, and application specific process data relevance to reduce energy requirements of nodes and increase their applicability. The reduction of data amount is realized by a parameter compressing data format and time-behavior adapted buffering. The reduction of number of transmissions is achieved by prevention of periodic updates and usage of echoes of other nodes. In order to react to dynamic environmental influences and energy fluctuations, sensor nodes improve their behavior. Change in behavior results from a compressed preference value. The accuracy of mapping of parameters to the preference is validated by several a priori rule checks and error adaptions.