Heterogeneous Wireless Sensor Nodes that Target the Internet of Things.

Wireless sensor networks (WSNs) combine embedded sensing and processing capabilities with a wireless communication infrastructure, all of which are essential for upcoming Internet-of-Things (IoT) applications. The limited communication bandwidth and energy budget of the nodes necessitate decentralized smart feature extraction and data aggregation algorithms to reduce the required communication bandwidth. Energy-efficient processing elements are thus required to meet the ever-growing computational demands of the IoT motes. The proposed Hardware-Accelerated Low-Power Mote (HaLoMote) combines a field-programmable gate array for hardware-accelerated data aggregation with a radio system on chip for network management. This article summarizes the architectural design decisions of HaLoMote, with a focus on its dynamic power management concepts. Furthermore, the authors show the possible power consumption improvements of HaLoMote compared to typical WSN software processors for two use cases based on general data compression and application-specific feature extraction.