Acoustic Ranging in Resource Constrained Sensor Networks

Fine-grained geographic localization of nodes is essential for an extensive range of distributed sensor applications. To compute geographic coordinates, localization algorithms commonly use pair-wise distance estimates between nodes. In this technical report we present a noise tolerant acoustic ranging mechanism for wireless sensors that employs digital signal processing techniques on standard MICA hardware. We describe how noise canceling, digital filtering and peak detection can be applied to meet the severe resource constraints of the platform, yet yielding average range estimation errors below 10cm independently from the actual node-to-node distances