Collaborative Channel Estimation in Backscattering Tag-to-Tag Networks

Backscattering Tag-to-Tag Networking (BTTN) represents a rapidly emerging paradigm enabling passive, radio-less tags to communicate directly with each other by reflecting (backscattering) an RF signal supplied by an external un-coordinated exciter. Recent advancements have taken the capability of BTTN beyond basic communication, empowering the networks with the ability to collaboratively sense and recognize human activities in the deployment space. The key to this ability is a novel passive channel estimation, allowing individual tags to measure tag-to-tag wireless channel parameters without involvement of any active radio. Previously reported techniques for this suffer from a limitation in that, they are unable to isolate the tag-to-tag channel of interest from the wider-range exciter-to-tag channels. As a result, the channel estimates and the analytics based thereof are susceptible to dynamic variations and clutter in the overall deployment environment, outside the range of the tag-to-tag link. In this paper, we overcome these limitations using a novel collaborative technique thus greatly enhancing the utility of passive channel estimation in BTTN. We elucidate our proposed technique using analytical modeling and validate with in-lab experiments using tag hardware built from discrete components.