Privacy-Aware Design and Analysis of Drone Remote Identification Systems

A key component to enabling the safe, secure, and efficient operations of Uncrewed Aerial Systems (UAS), or drones, is the ability to surveil such vehicles in flight. Remote identification systems (e.g., Remote ID) are designed to serve such purposes, and lay the foundations for denser, more advanced UAS operations. However, given the individualized nature of many commonly-envisioned UAS use cases (e.g., door-to-door package delivery), the design of remote identification systems must consider the trade-offs between surveillance versus consumer privacy. To this end, we present a comprehensive method for first characterizing the surveillance coverage of remote identification systems, then integrating in privacy from the perspective of plausible deniability: Surveillance gaps present opportunities for enhanced trajectory privacy, but too many gaps hinder the original purpose-surveillance and safety-of remote identification systems. We apply our method to realistic urban geographies (San Francisco, New York City, Los Angeles) where future UAS operations may perform door-to-door package deliveries between vendors and consumers. We provide ranges for the number of remote identification receivers needed to cover 50-95% of simulated UAS trajectories, then compute a privacy score based on the level of afforded plausible deniability. Our design and analysis framework contribute to the integration of UAS into the future intelligent transportation and smart city landscape, and will be useful for airspace designers, UAS fleet operators, air navigation service providers, and city planners interested in ensuring vehicle safety while considering data usage and an individual's privacy.