HarmonicID: An Identification System for Low-Power Analog Backscatter Tags

Backscatter tags consume extremely low power and are activated by an external carrier for sensing and communication purposes. Analog backscatter tags achieve further power and cost reductions by moving digitization and computation overhead to a remote receiver. Analog transmissions do not have a digital frame structure to convey control information such as addresses. Therefore, when multiple tags are deployed without such ID it is challenging to associate received data with specific tags. We present a novel fully analog tag identification mechanism that is inspired by musical instruments. Each instrument has a unique harmonic content and therefore the human ear can distinguish different musical instruments even when they are playing at the same frequency. By tuning the tags to reflect a unique harmonic content we enable identification of purely analog tags. Our results show that we can create identities for tags by leveraging signal conditioning to modify harmonics, without the need to add dedicated digital blocks for identification purposes. Experiments on our hardware demonstrate that we can successfully identify the tags even in challenging scenarios such as when they are blocked by walls or in motion.